├── .flake8 ├── .github ├── PULL_REQUEST_TEMPLATE.md ├── scripts │ └── pre_build_script.sh └── workflows │ ├── build_docs.yaml │ ├── build_linux_wheels.yaml │ ├── export.yaml │ ├── gpu_test.yaml │ ├── lint.yaml │ ├── regression_test.yaml │ ├── rl_test.yaml │ └── unit_test.yaml ├── .gitignore ├── .pre-commit-config.yaml ├── CITATION.cff ├── CODE_OF_CONDUCT.md ├── CONTRIBUTING.md ├── LICENSE ├── MANIFEST.in ├── README.md ├── docs ├── Makefile ├── license_header.txt ├── requirements.txt └── source │ ├── _static │ ├── css │ │ └── custom_torchtune.css │ └── img │ │ ├── card-background.svg │ │ ├── comet_torchtune_project.png │ │ ├── generic-pytorch-logo.png │ │ ├── kd-finetune-student.png │ │ ├── kd-finetune-teacher.png │ │ ├── kd-hyperparam-kd-ratio.png │ │ ├── kd-hyperparam-lr.png │ │ ├── kd-qwen2-res.png │ │ ├── kd-simplified.png │ │ ├── lora_diagram.png │ │ ├── lora_experiment_loss_curves.png │ │ ├── pytorch-logo-dark.png │ │ ├── pytorch-logo-dark.svg │ │ ├── pytorch-logo-flame.png │ │ ├── pytorch-logo-flame.svg │ │ ├── qat_diagram.png │ │ ├── qlora_exp.png │ │ ├── torchtune_datasets.svg │ │ └── torchtune_workspace.png │ ├── _templates │ ├── autosummary │ │ ├── class.rst │ │ └── function.rst │ └── layout.html │ ├── api_ref_config.rst │ ├── api_ref_data.rst │ ├── api_ref_datasets.rst │ ├── api_ref_generation.rst │ ├── api_ref_models.rst │ ├── api_ref_modules.rst │ ├── api_ref_rlhf.rst │ ├── api_ref_training.rst │ ├── api_ref_utilities.rst │ ├── basics │ ├── chat_datasets.rst │ ├── custom_components.rst │ ├── datasets_overview.rst │ ├── instruct_datasets.rst │ ├── message_transforms.rst │ ├── messages.rst │ ├── model_transforms.rst │ ├── multimodal_datasets.rst │ ├── packing.rst │ ├── preference_datasets.rst │ ├── prompt_templates.rst │ ├── text_completion_datasets.rst │ └── tokenizers.rst │ ├── conf.py │ ├── custom_directives.py │ ├── deep_dives │ ├── README.txt │ ├── checkpointer.rst │ ├── comet_logging.rst │ ├── configs.rst │ ├── recipe_deepdive.rst │ └── wandb_logging.rst │ ├── index.rst │ ├── install.rst │ ├── overview.rst │ ├── recipes │ ├── dpo.rst │ ├── lora_finetune_single_device.rst │ ├── qat_distributed.rst │ └── recipes_overview.rst │ ├── tune_cli.rst │ └── tutorials │ ├── README.txt │ ├── chat.rst │ ├── e2e_flow.rst │ ├── first_finetune_tutorial.rst │ ├── llama3.rst │ ├── llama_kd_tutorial.rst │ ├── lora_finetune.rst │ ├── memory_optimizations.rst │ ├── multinode.rst │ ├── qat_finetune.rst │ └── qlora_finetune.rst ├── pyproject.toml ├── recipes ├── __init__.py ├── configs │ ├── dev │ │ ├── 11B_lora_multi_dataset.yaml │ │ ├── 3B_full_grpo.yaml │ │ ├── 3B_sft_for_grpo.yaml │ │ ├── 8B_full_experimental.yaml │ │ ├── qwen3B_async_grpo.yaml │ │ └── qwen3B_sync_grpo.yaml │ ├── eleuther_evaluation.yaml │ ├── gemma │ │ ├── 2B_full.yaml │ │ ├── 2B_lora.yaml │ │ ├── 2B_lora_single_device.yaml │ │ ├── 2B_qlora_single_device.yaml │ │ ├── 7B_full.yaml │ │ ├── 7B_lora.yaml │ │ ├── 7B_lora_single_device.yaml │ │ ├── 7B_qlora_single_device.yaml │ │ └── evaluation.yaml │ ├── gemma2 │ │ ├── 27B_full.yaml │ │ ├── 27B_lora.yaml │ │ ├── 27B_lora_single_device.yaml │ │ ├── 27B_qlora_single_device.yaml │ │ ├── 2B_full.yaml │ │ ├── 2B_lora.yaml │ │ ├── 2B_lora_single_device.yaml │ │ ├── 2B_qlora_single_device.yaml │ │ ├── 9B_full.yaml │ │ ├── 9B_lora.yaml │ │ ├── 9B_lora_single_device.yaml │ │ └── 9B_qlora_single_device.yaml │ ├── generation.yaml │ ├── llama2 │ │ ├── 13B_full.yaml │ │ ├── 13B_lora.yaml │ │ ├── 13B_qlora_single_device.yaml │ │ ├── 1B_full_ppo_low_memory_single_device.yaml │ │ ├── 1B_qat_single_device.yaml │ │ ├── 70B_lora.yaml │ │ ├── 70B_qlora.yaml │ │ ├── 7B_full.yaml │ │ ├── 7B_full_low_memory.yaml │ │ ├── 7B_lora.yaml │ │ ├── 7B_lora_dpo.yaml │ │ ├── 7B_lora_dpo_single_device.yaml │ │ ├── 7B_lora_single_device.yaml │ │ ├── 7B_qat_full.yaml │ │ ├── 7B_qlora.yaml │ │ ├── 7B_qlora_single_device.yaml │ │ └── generation_v2.yaml │ ├── llama3 │ │ ├── 70B_full.yaml │ │ ├── 70B_generation_distributed.yaml │ │ ├── 70B_lora.yaml │ │ ├── 8B_dora.yaml │ │ ├── 8B_dora_single_device.yaml │ │ ├── 8B_full.yaml │ │ ├── 8B_full_single_device.yaml │ │ ├── 8B_lora.yaml │ │ ├── 8B_lora_single_device.yaml │ │ ├── 8B_qat_full.yaml │ │ ├── 8B_qat_lora.yaml │ │ ├── 8B_qdora_single_device.yaml │ │ └── 8B_qlora_single_device.yaml │ ├── llama3_1 │ │ ├── 405B_qlora.yaml │ │ ├── 70B_full.yaml │ │ ├── 70B_generation_distributed.yaml │ │ ├── 70B_lora.yaml │ │ ├── 8B_full.yaml │ │ ├── 8B_full_dpo.yaml │ │ ├── 8B_full_single_device.yaml │ │ ├── 8B_lora.yaml │ │ ├── 8B_lora_dpo.yaml │ │ ├── 8B_lora_dpo_single_device.yaml │ │ ├── 8B_lora_single_device.yaml │ │ ├── 8B_qat_full.yaml │ │ ├── 8B_qat_lora.yaml │ │ ├── 8B_qlora_single_device.yaml │ │ └── evaluation.yaml │ ├── llama3_2 │ │ ├── 1B_full.yaml │ │ ├── 1B_full_single_device.yaml │ │ ├── 1B_lora.yaml │ │ ├── 1B_lora_single_device.yaml │ │ ├── 1B_qat_lora.yaml │ │ ├── 1B_qlora_single_device.yaml │ │ ├── 3B_full.yaml │ │ ├── 3B_full_single_device.yaml │ │ ├── 3B_lora.yaml │ │ ├── 3B_lora_single_device.yaml │ │ ├── 3B_qat_full.yaml │ │ ├── 3B_qat_lora.yaml │ │ ├── 3B_qlora_single_device.yaml │ │ ├── 8B_to_1B_KD_lora_distributed.yaml │ │ ├── 8B_to_1B_KD_lora_single_device.yaml │ │ └── evaluation.yaml │ ├── llama3_2_vision │ │ ├── 11B_evaluation.yaml │ │ ├── 11B_full.yaml │ │ ├── 11B_full_single_device.yaml │ │ ├── 11B_generation_v2.yaml │ │ ├── 11B_lora.yaml │ │ ├── 11B_lora_single_device.yaml │ │ ├── 11B_qlora.yaml │ │ ├── 11B_qlora_single_device.yaml │ │ ├── 90B_full.yaml │ │ ├── 90B_lora.yaml │ │ └── 90B_qlora.yaml │ ├── llama3_3 │ │ ├── 70B_full.yaml │ │ ├── 70B_full_multinode.yaml │ │ ├── 70B_generation_distributed.yaml │ │ ├── 70B_lora.yaml │ │ └── 70B_qlora.yaml │ ├── llama4 │ │ ├── maverick_17B_128E_full.yaml │ │ ├── scout_17B_16E_full.yaml │ │ ├── scout_17B_16E_generation_distributed.yaml │ │ └── scout_17B_16E_lora.yaml │ ├── mistral │ │ ├── 7B_full.yaml │ │ ├── 7B_full_low_memory.yaml │ │ ├── 7B_full_ppo_low_memory.yaml │ │ ├── 7B_lora.yaml │ │ ├── 7B_lora_single_device.yaml │ │ ├── 7B_qlora_single_device.yaml │ │ └── evaluation.yaml │ ├── phi3 │ │ ├── evaluation.yaml │ │ ├── mini_full.yaml │ │ ├── mini_full_low_memory.yaml │ │ ├── mini_lora.yaml │ │ ├── mini_lora_single_device.yaml │ │ └── mini_qlora_single_device.yaml │ ├── phi4 │ │ ├── 14B_full.yaml │ │ ├── 14B_full_low_memory.yaml │ │ ├── 14B_lora.yaml │ │ ├── 14B_lora_single_device.yaml │ │ ├── 14B_qlora_single_device.yaml │ │ └── evaluation.yaml │ ├── quantization.yaml │ ├── qwen2 │ │ ├── 0.5B_full.yaml │ │ ├── 0.5B_full_single_device.yaml │ │ ├── 0.5B_lora.yaml │ │ ├── 0.5B_lora_single_device.yaml │ │ ├── 1.5B_full.yaml │ │ ├── 1.5B_full_single_device.yaml │ │ ├── 1.5B_lora.yaml │ │ ├── 1.5B_lora_single_device.yaml │ │ ├── 1.5_to_0.5B_KD_lora_distributed.yaml │ │ ├── 1.5_to_0.5B_KD_lora_single_device.yaml │ │ ├── 7B_full.yaml │ │ ├── 7B_full_single_device.yaml │ │ ├── 7B_lora.yaml │ │ ├── 7B_lora_single_device.yaml │ │ └── evaluation.yaml │ ├── qwen2_5 │ │ ├── 0.5B_full.yaml │ │ ├── 0.5B_full_single_device.yaml │ │ ├── 0.5B_lora.yaml │ │ ├── 0.5B_lora_single_device.yaml │ │ ├── 1.5B_full.yaml │ │ ├── 1.5B_full_single_device.yaml │ │ ├── 1.5B_lora.yaml │ │ ├── 1.5B_lora_single_device.yaml │ │ ├── 1.5B_qat_single_device.yaml │ │ ├── 14B_lora_single_device.yaml │ │ ├── 14B_to_7B_KD_lora_single_device.yaml │ │ ├── 32B_lora.yaml │ │ ├── 3B_full.yaml │ │ ├── 3B_full_single_device.yaml │ │ ├── 3B_lora.yaml │ │ ├── 3B_lora_single_device.yaml │ │ ├── 3B_qat_single_device.yaml │ │ ├── 72B_lora.yaml │ │ ├── 7B_full.yaml │ │ ├── 7B_full_single_device.yaml │ │ ├── 7B_lora.yaml │ │ ├── 7B_lora_single_device.yaml │ │ └── evaluation.yaml │ └── qwen3 │ │ ├── 0.6B_full.yaml │ │ ├── 0.6B_full_single_device.yaml │ │ ├── 0.6B_lora.yaml │ │ ├── 0.6B_lora_single_device.yaml │ │ ├── 0.6B_qat_full.yaml │ │ ├── 0.6B_qat_lora.yaml │ │ ├── 1.7B_full.yaml │ │ ├── 1.7B_full_single_device.yaml │ │ ├── 1.7B_lora.yaml │ │ ├── 1.7B_lora_single_device.yaml │ │ ├── 1.7B_qat_full.yaml │ │ ├── 1.7B_qat_lora.yaml │ │ ├── 14B_lora_single_device.yaml │ │ ├── 14B_to_8B_KD_lora_single_device.yaml │ │ ├── 32B_lora.yaml │ │ ├── 32B_qat_lora.yaml │ │ ├── 4B_full.yaml │ │ ├── 4B_full_single_device.yaml │ │ ├── 4B_lora.yaml │ │ ├── 4B_lora_single_device.yaml │ │ ├── 4B_qat_full.yaml │ │ ├── 4B_qat_lora.yaml │ │ ├── 8B_full.yaml │ │ ├── 8B_full_single_device.yaml │ │ ├── 8B_lora.yaml │ │ ├── 8B_lora_single_device.yaml │ │ ├── 8B_qat_full.yaml │ │ ├── 8B_qat_lora.yaml │ │ └── evaluation.yaml ├── dev │ ├── 7B_full_early_exit.yaml │ ├── async_grpo.md │ ├── async_grpo_full_finetune_distributed.py │ ├── early_exit_finetune_distributed.py │ ├── generate_v2.py │ ├── generate_v2_distributed.py │ ├── grpo_full_finetune_distributed.py │ ├── gsm8k_sft.sbatch │ ├── lora_finetune_distributed_multi_dataset.py │ └── multinode_grpo.sbatch ├── eleuther_eval.py ├── full_dpo_distributed.py ├── full_finetune_distributed.py ├── full_finetune_multinode.slurm ├── full_finetune_single_device.py ├── generate.py ├── knowledge_distillation_distributed.py ├── knowledge_distillation_single_device.py ├── lora_dpo_distributed.py ├── lora_dpo_single_device.py ├── lora_finetune_distributed.py ├── lora_finetune_single_device.py ├── ppo_full_finetune_single_device.py ├── qat_distributed.py ├── qat_lora_finetune_distributed.py ├── qat_single_device.py ├── quantization.md └── quantize.py ├── tests ├── __init__.py ├── assets │ ├── README.md │ ├── alpaca_tiny.json │ ├── chat_tiny.json │ ├── dog_on_skateboard.jpg │ ├── generation_config.json │ ├── generation_config_gemma.json │ ├── hh_rlhf_tiny.json │ ├── instruct_tiny.json │ ├── invalid_dummy_config.yaml │ ├── m.model │ ├── merges.txt │ ├── rgb_pytorch.png │ ├── sentencepiece.model │ ├── stack_exchange_paired_tiny.json │ ├── tiktoken_small.model │ ├── tiktoken_small_llama4.model │ ├── tiny_bpe_merges.txt │ ├── tiny_bpe_tokenizer.json │ ├── tiny_bpe_vocab.json │ ├── tiny_fair_checkpoint.pt │ ├── tiny_state_dict_with_one_key.pt │ ├── tokenizer.json │ ├── tokenizer_config.json │ ├── tokenizer_config_gemma.json │ ├── tokenizer_gemma_cropped.json │ ├── valid_dummy_config.yaml │ ├── vocab.json │ └── vqa_tiny.json ├── cache_artifacts.sh ├── common.py ├── conftest.py ├── recipes │ ├── __init__.py │ ├── common.py │ ├── dev │ │ ├── test_async_grpo_full_finetune_distributed.py │ │ └── test_generate_v2.py │ ├── test_configs.py │ ├── test_eleuther_eval.py │ ├── test_full_dpo_distributed.py │ ├── test_full_finetune_distributed.py │ ├── test_full_finetune_single_device.py │ ├── test_knowledge_distillation_distributed.py │ ├── test_knowledge_distillation_single_device.py │ ├── test_lora_dpo_distributed.py │ ├── test_lora_dpo_single_device.py │ ├── test_lora_finetune_distributed.py │ ├── test_lora_finetune_single_device.py │ ├── test_ppo_full_finetune_single_device.py │ ├── test_qat_distributed.py │ ├── test_qat_lora_finetune_distributed.py │ ├── test_qat_single_device.py │ └── utils.py ├── regression_tests │ └── test_llama2_7b.py ├── test_import_recipes.py ├── test_utils.py └── torchtune │ ├── __init__.py │ ├── _cli │ ├── __init__.py │ ├── test_cat.py │ ├── test_cp.py │ ├── test_download.py │ ├── test_ls.py │ ├── test_run.py │ ├── test_tune.py │ └── test_validate.py │ ├── config │ ├── test_config_utils.py │ ├── test_instantiate.py │ ├── test_parse.py │ └── test_validate.py │ ├── data │ ├── test_collate.py │ ├── test_data_utils.py │ ├── test_messages.py │ └── test_prompt_templates.py │ ├── datasets │ ├── __init__.py │ ├── multimodal │ │ ├── test_llava_instruct_dataset.py │ │ ├── test_multimodal_chat_dataset.py │ │ ├── test_the_cauldron_dataset.py │ │ └── test_vqa_dataset.py │ ├── test_alpaca_dataset.py │ ├── test_chat_dataset.py │ ├── test_cnn_dailymail_dataset.py │ ├── test_concat_dataset.py │ ├── test_grammar_dataset.py │ ├── test_hh_rlhf_helpful_dataset.py │ ├── test_instruct_dataset.py │ ├── test_packed_dataset.py │ ├── test_preference_dataset.py │ ├── test_samsum_dataset.py │ ├── test_sft_dataset.py │ ├── test_slimorca_dataset.py │ ├── test_stack_exchange_paired_dataset.py │ ├── test_text_completion_dataset.py │ └── test_wikitext_dataset.py │ ├── dev │ └── rl │ │ ├── rewards │ │ ├── __init__.py │ │ └── test_rewards.py │ │ └── workers │ │ └── test_postprocessing.py │ ├── generation │ ├── __init__.py │ └── test_generation.py │ ├── models │ ├── __init__.py │ ├── clip │ │ ├── __init__.py │ │ ├── test_clip_image_transform.py │ │ ├── test_clip_text_encoder.py │ │ ├── test_clip_tokenizer.py │ │ ├── test_pos_embedding_interpolation.py │ │ └── test_positional_embeddings.py │ ├── flamingo │ │ ├── test_flamingo_decoder.py │ │ └── test_flamingo_encoder.py │ ├── flux │ │ ├── __init__.py │ │ └── test_flux_autoencoder.py │ ├── gemma │ │ └── test_gemma_tokenizer.py │ ├── llama2 │ │ ├── scripts │ │ │ ├── README.md │ │ │ ├── __init__.py │ │ │ ├── compare_attention.py │ │ │ ├── compare_decoder.py │ │ │ ├── compare_decoder_layer.py │ │ │ ├── compare_dora.py │ │ │ ├── compare_feed_forward.py │ │ │ ├── compare_fused_attention.py │ │ │ ├── compare_lora.py │ │ │ ├── compare_lora_attention.py │ │ │ └── compare_lora_llama2.py │ │ ├── test_llama2_prompt_template.py │ │ ├── test_llama2_tokenizer.py │ │ └── test_lora_llama2.py │ ├── llama3 │ │ ├── test_llama3.py │ │ └── test_llama3_tokenizer.py │ ├── llama3_1 │ │ └── test_position_embeddings.py │ ├── llama3_2_vision │ │ └── test_llama_vision_lora.py │ ├── llama4 │ │ ├── test_llama4.py │ │ ├── test_llama4_tokenizer.py │ │ └── test_llama4_transform.py │ ├── mistral │ │ ├── scripts │ │ │ ├── README.md │ │ │ ├── __init__.py │ │ │ ├── compare_feed_forward.py │ │ │ ├── compare_mistral.py │ │ │ ├── compare_mistral_classifier.py │ │ │ ├── mistral_reference.py │ │ │ └── mistral_test_config.py │ │ ├── test_mistral.py │ │ ├── test_mistral_classifier.py │ │ ├── test_mistral_prompt_template.py │ │ └── test_mistral_tokenizer.py │ ├── phi3 │ │ ├── __init__.py │ │ ├── test_lora_phi3.py │ │ ├── test_phi3.py │ │ ├── test_phi3_position_embeddings.py │ │ └── test_phi3_tokenizer.py │ ├── phi4 │ │ ├── __init__.py │ │ └── test_phi4_tokenizer.py │ ├── qwen2 │ │ ├── __init__.py │ │ ├── test_lora_qwen2.py │ │ ├── test_qwen2.py │ │ └── test_qwen2_tokenizer.py │ ├── qwen2_5 │ │ ├── __init__.py │ │ └── test_tokenizer.py │ ├── qwen3 │ │ ├── __init__.py │ │ └── test_tokenizer.py │ └── t5 │ │ ├── __init__.py │ │ ├── test_t5_encoder.py │ │ └── test_t5_tokenizer.py │ ├── modules │ ├── __init__.py │ ├── _export │ │ ├── test_attention.py │ │ └── test_export_position_embeddings.py │ ├── loss │ │ ├── test_ce_chunked_output_loss.py │ │ ├── test_cross_entropy_loss.py │ │ └── test_kd_losses.py │ ├── low_precision │ │ ├── __init__.py │ │ ├── test_nf4_dispatch_registration.py │ │ └── test_nf4_linear.py │ ├── model_fusion │ │ ├── __init__.py │ │ ├── test_deep_fusion.py │ │ ├── test_early_fusion.py │ │ ├── test_fusion_layers.py │ │ └── test_fusion_utils.py │ ├── moe │ │ ├── test_experts.py │ │ └── test_moe.py │ ├── peft │ │ ├── __init__.py │ │ ├── test_dora.py │ │ ├── test_lora.py │ │ └── test_utils.py │ ├── test_attention.py │ ├── test_attention_utils.py │ ├── test_classifier.py │ ├── test_common_utils.py │ ├── test_cosine_with_warmup.py │ ├── test_early_exit_loss.py │ ├── test_embedding_utils.py │ ├── test_feed_forward.py │ ├── test_kv_cache.py │ ├── test_layer_dropout.py │ ├── test_layernorm.py │ ├── test_optim.py │ ├── test_position_embeddings.py │ ├── test_rms_norm.py │ ├── test_transformer_decoder.py │ ├── test_vision_transformer.py │ ├── test_vq_embeddings.py │ ├── tokenizers │ │ └── test_gpt2.py │ └── transforms │ │ ├── test_get_canvas_best_fit.py │ │ ├── test_get_inscribed_size.py │ │ ├── test_pad_dim_to_size.py │ │ ├── test_resize_with_pad.py │ │ ├── test_tile_crop.py │ │ ├── test_transforms.py │ │ └── tokenizers │ │ ├── test_hf_tokenizer.py │ │ ├── test_sentencepiece.py │ │ ├── test_tiktoken.py │ │ └── test_utils.py │ ├── rl │ └── test_parameter_server.py │ ├── rlhf │ ├── __init__.py │ ├── loss │ │ ├── __init__.py │ │ ├── test_dpo_loss.py │ │ └── test_ppo_loss.py │ ├── test_rewards.py │ └── test_sequence_processing.py │ ├── training │ ├── checkpointing │ │ ├── test_checkpointer.py │ │ ├── test_checkpointer_utils.py │ │ └── test_distributed_checkpointer.py │ ├── test_activation_offloading.py │ ├── test_distributed.py │ ├── test_memory.py │ ├── test_metric_logging.py │ ├── test_model_util.py │ ├── test_pooling.py │ ├── test_precision.py │ ├── test_profiler.py │ ├── test_quantization.py │ └── test_seed.py │ └── utils │ ├── __init__.py │ ├── test_device.py │ └── test_logging.py ├── torchtune ├── __init__.py ├── _cli │ ├── __init__.py │ ├── cat.py │ ├── cp.py │ ├── download.py │ ├── ls.py │ ├── run.py │ ├── subcommand.py │ ├── tune.py │ └── validate.py ├── _recipe_registry.py ├── config │ ├── __init__.py │ ├── _errors.py │ ├── _instantiate.py │ ├── _parse.py │ ├── _utils.py │ └── _validate.py ├── data │ ├── __init__.py │ ├── _collate.py │ ├── _common.py │ ├── _messages.py │ ├── _prompt_templates.py │ ├── _torchdata.py │ └── _utils.py ├── datasets │ ├── __init__.py │ ├── _alpaca.py │ ├── _chat.py │ ├── _cnn_dailymail.py │ ├── _concat.py │ ├── _grammar.py │ ├── _hh_rlhf_helpful.py │ ├── _instruct.py │ ├── _packed.py │ ├── _preference.py │ ├── _samsum.py │ ├── _sft.py │ ├── _slimorca.py │ ├── _stack_exchange_paired.py │ ├── _text_completion.py │ ├── _wikitext.py │ └── multimodal │ │ ├── __init__.py │ │ ├── _llava_instruct.py │ │ ├── _multimodal.py │ │ ├── _the_cauldron.py │ │ └── _vqa.py ├── dev │ ├── README.md │ ├── __init__.py │ ├── grpo │ │ ├── __init__.py │ │ ├── data.py │ │ ├── generation.py │ │ ├── gsm8k.py │ │ ├── loss.py │ │ ├── rewards.py │ │ └── types.py │ └── rl │ │ ├── __init__.py │ │ ├── data.py │ │ ├── datatypes │ │ ├── __init__.py │ │ ├── request_output.py │ │ ├── trajectory.py │ │ └── vllm_completion_output.py │ │ ├── generation.py │ │ ├── gsm8k.py │ │ ├── linear_grpo_loss.py │ │ ├── rewards.py │ │ ├── types.py │ │ ├── utils │ │ ├── __init__.py │ │ └── dist.py │ │ └── workers │ │ ├── __init__.py │ │ ├── datacollectors │ │ ├── __init__.py │ │ └── sync.py │ │ ├── metric_logger.py │ │ ├── parameter_servers │ │ ├── __init__.py │ │ └── vllm.py │ │ ├── postprocessing.py │ │ ├── trainers │ │ ├── __init__.py │ │ └── training.py │ │ └── weight_updaters │ │ ├── __init__.py │ │ └── weight_updater.py ├── generation │ ├── __init__.py │ └── _generation.py ├── models │ ├── __init__.py │ ├── clip │ │ ├── __init__.py │ │ ├── _component_builders.py │ │ ├── _convert_weights.py │ │ ├── _model_builders.py │ │ ├── _position_embeddings.py │ │ ├── _text_encoder.py │ │ ├── _tokenizer.py │ │ ├── _transform.py │ │ └── inference │ │ │ └── _transform.py │ ├── convert_weights.py │ ├── flux │ │ ├── __init__.py │ │ ├── _autoencoder.py │ │ ├── _convert_weights.py │ │ └── _model_builders.py │ ├── gemma │ │ ├── __init__.py │ │ ├── _component_builders.py │ │ ├── _model_builders.py │ │ ├── _tokenizer.py │ │ ├── gemma_norm_embedding.py │ │ └── rms_norm.py │ ├── gemma2 │ │ ├── __init__.py │ │ ├── _attention.py │ │ ├── _component_builders.py │ │ ├── _convert_weights.py │ │ └── _model_builders.py │ ├── llama2 │ │ ├── __init__.py │ │ ├── _component_builders.py │ │ ├── _model_builders.py │ │ ├── _model_utils.py │ │ ├── _prompt_template.py │ │ └── _tokenizer.py │ ├── llama3 │ │ ├── __init__.py │ │ ├── _component_builders.py │ │ ├── _model_builders.py │ │ ├── _model_utils.py │ │ ├── _parallelism.py │ │ └── _tokenizer.py │ ├── llama3_1 │ │ ├── __init__.py │ │ ├── _component_builders.py │ │ ├── _model_builders.py │ │ └── _position_embeddings.py │ ├── llama3_2 │ │ ├── __init__.py │ │ ├── _component_builders.py │ │ └── _model_builders.py │ ├── llama3_2_vision │ │ ├── __init__.py │ │ ├── _component_builders.py │ │ ├── _convert_weights.py │ │ ├── _encoder.py │ │ ├── _model_builders.py │ │ └── _transform.py │ ├── llama3_3 │ │ ├── __init__.py │ │ └── _model_builders.py │ ├── llama4 │ │ ├── __init__.py │ │ ├── _chunked_attention.py │ │ ├── _component_builders.py │ │ ├── _convert_weights.py │ │ ├── _encoder.py │ │ ├── _model_builders.py │ │ ├── _parallelism.py │ │ ├── _position_embeddings.py │ │ ├── _tokenizer.py │ │ └── _transform.py │ ├── mistral │ │ ├── __init__.py │ │ ├── _component_builders.py │ │ ├── _model_builders.py │ │ ├── _prompt_template.py │ │ └── _tokenizer.py │ ├── phi3 │ │ ├── __init__.py │ │ ├── _component_builders.py │ │ ├── _convert_weights.py │ │ ├── _model_builders.py │ │ ├── _position_embeddings.py │ │ └── _tokenizer.py │ ├── phi4 │ │ ├── __init__.py │ │ ├── _model_builders.py │ │ └── _tokenizer.py │ ├── qwen2 │ │ ├── __init__.py │ │ ├── _component_builders.py │ │ ├── _convert_weights.py │ │ ├── _model_builders.py │ │ ├── _positional_embeddings.py │ │ └── _tokenizer.py │ ├── qwen2_5 │ │ ├── __init__.py │ │ ├── _model_builders.py │ │ └── _tokenizer.py │ ├── qwen3 │ │ ├── __init__.py │ │ ├── _attention.py │ │ ├── _component_builders.py │ │ ├── _convert_weights.py │ │ ├── _model_builders.py │ │ └── _tokenizer.py │ ├── smol │ │ ├── __init__.py │ │ ├── _component_builders.py │ │ └── _model_builders.py │ └── t5 │ │ ├── __init__.py │ │ ├── _component_builders.py │ │ ├── _convert_weights.py │ │ ├── _encoder.py │ │ ├── _model_builders.py │ │ └── _tokenizer.py ├── modules │ ├── __init__.py │ ├── _export │ │ ├── README.md │ │ ├── _position_embeddings.py │ │ ├── attention.py │ │ ├── install_requirements.sh │ │ └── kv_cache.py │ ├── attention.py │ ├── attention_utils.py │ ├── classifier.py │ ├── common_utils.py │ ├── early_exit_loss.py │ ├── embedding_utils.py │ ├── feed_forward.py │ ├── kv_cache.py │ ├── layer_dropout.py │ ├── layer_norm.py │ ├── loss │ │ ├── __init__.py │ │ ├── ce_chunked_output_loss.py │ │ ├── cross_entropy_loss.py │ │ ├── kd_losses.py │ │ └── loss_types.py │ ├── low_precision │ │ ├── __init__.py │ │ ├── _register_nf4_dispatch_ops.py │ │ └── nf4_linear.py │ ├── model_fusion │ │ ├── __init__.py │ │ ├── _deep_fusion.py │ │ ├── _early_fusion.py │ │ ├── _fusion_layers.py │ │ └── _fusion_utils.py │ ├── moe │ │ ├── __init__.py │ │ ├── _parallelism.py │ │ ├── experts.py │ │ ├── indices.py │ │ ├── moe.py │ │ └── utils.py │ ├── optim.py │ ├── peft │ │ ├── __init__.py │ │ ├── _utils.py │ │ ├── dora.py │ │ └── lora.py │ ├── position_embeddings.py │ ├── rms_norm.py │ ├── tanh_gate.py │ ├── tied_linear.py │ ├── tokenizers │ │ └── __init__.py │ ├── transformer.py │ ├── transforms │ │ ├── __init__.py │ │ ├── _transforms.py │ │ ├── tokenizers │ │ │ ├── __init__.py │ │ │ ├── _gpt2.py │ │ │ ├── _hf_tokenizer.py │ │ │ ├── _sentencepiece.py │ │ │ ├── _tiktoken.py │ │ │ └── _utils.py │ │ └── vision_utils │ │ │ ├── __init__.py │ │ │ ├── get_canvas_best_fit.py │ │ │ ├── get_inscribed_size.py │ │ │ ├── pad_dim_to_size.py │ │ │ ├── resize_with_pad.py │ │ │ └── tile_crop.py │ ├── vision_transformer.py │ └── vq_embeddings.py ├── recipe_interfaces.py ├── rlhf │ ├── __init__.py │ ├── _types.py │ ├── loss │ │ ├── __init__.py │ │ ├── dpo.py │ │ └── ppo.py │ ├── rewards.py │ ├── sequence_processing.py │ └── utils │ │ ├── __init__.py │ │ └── _convert_weights.py ├── training │ ├── __init__.py │ ├── _activation_offloading.py │ ├── _compile.py │ ├── _distributed.py │ ├── _grad_scaler.py │ ├── _model_util.py │ ├── _profiler.py │ ├── activations.py │ ├── checkpointing │ │ ├── __init__.py │ │ ├── _checkpoint_client.py │ │ ├── _checkpointer.py │ │ └── _utils.py │ ├── lr_schedulers.py │ ├── memory.py │ ├── metric_logging.py │ ├── pooling.py │ ├── precision.py │ ├── quantization.py │ └── seed.py └── utils │ ├── __init__.py │ ├── _device.py │ ├── _import_guard.py │ ├── _logging.py │ └── _version.py └── version.txt /.flake8: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/.flake8 -------------------------------------------------------------------------------- /.github/PULL_REQUEST_TEMPLATE.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/.github/PULL_REQUEST_TEMPLATE.md -------------------------------------------------------------------------------- /.github/scripts/pre_build_script.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/.github/scripts/pre_build_script.sh -------------------------------------------------------------------------------- /.github/workflows/build_docs.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/.github/workflows/build_docs.yaml -------------------------------------------------------------------------------- /.github/workflows/build_linux_wheels.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/.github/workflows/build_linux_wheels.yaml -------------------------------------------------------------------------------- /.github/workflows/export.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/.github/workflows/export.yaml -------------------------------------------------------------------------------- /.github/workflows/gpu_test.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/.github/workflows/gpu_test.yaml -------------------------------------------------------------------------------- /.github/workflows/lint.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/.github/workflows/lint.yaml -------------------------------------------------------------------------------- /.github/workflows/regression_test.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/.github/workflows/regression_test.yaml -------------------------------------------------------------------------------- /.github/workflows/rl_test.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/.github/workflows/rl_test.yaml -------------------------------------------------------------------------------- /.github/workflows/unit_test.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/.github/workflows/unit_test.yaml -------------------------------------------------------------------------------- /.gitignore: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/.gitignore -------------------------------------------------------------------------------- /.pre-commit-config.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/.pre-commit-config.yaml -------------------------------------------------------------------------------- /CITATION.cff: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/CITATION.cff -------------------------------------------------------------------------------- /CODE_OF_CONDUCT.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/CODE_OF_CONDUCT.md -------------------------------------------------------------------------------- /CONTRIBUTING.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/CONTRIBUTING.md -------------------------------------------------------------------------------- /LICENSE: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/LICENSE -------------------------------------------------------------------------------- /MANIFEST.in: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/MANIFEST.in -------------------------------------------------------------------------------- /README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/README.md -------------------------------------------------------------------------------- /docs/Makefile: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/docs/Makefile -------------------------------------------------------------------------------- /docs/license_header.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/docs/license_header.txt -------------------------------------------------------------------------------- /docs/requirements.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/docs/requirements.txt -------------------------------------------------------------------------------- /docs/source/_static/css/custom_torchtune.css: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/docs/source/_static/css/custom_torchtune.css -------------------------------------------------------------------------------- /docs/source/_static/img/card-background.svg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/docs/source/_static/img/card-background.svg -------------------------------------------------------------------------------- /docs/source/_static/img/comet_torchtune_project.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/docs/source/_static/img/comet_torchtune_project.png -------------------------------------------------------------------------------- /docs/source/_static/img/generic-pytorch-logo.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/docs/source/_static/img/generic-pytorch-logo.png -------------------------------------------------------------------------------- /docs/source/_static/img/kd-finetune-student.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/docs/source/_static/img/kd-finetune-student.png -------------------------------------------------------------------------------- /docs/source/_static/img/kd-finetune-teacher.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/docs/source/_static/img/kd-finetune-teacher.png -------------------------------------------------------------------------------- /docs/source/_static/img/kd-hyperparam-kd-ratio.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/docs/source/_static/img/kd-hyperparam-kd-ratio.png -------------------------------------------------------------------------------- /docs/source/_static/img/kd-hyperparam-lr.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/docs/source/_static/img/kd-hyperparam-lr.png -------------------------------------------------------------------------------- /docs/source/_static/img/kd-qwen2-res.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/docs/source/_static/img/kd-qwen2-res.png -------------------------------------------------------------------------------- /docs/source/_static/img/kd-simplified.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/docs/source/_static/img/kd-simplified.png -------------------------------------------------------------------------------- /docs/source/_static/img/lora_diagram.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/docs/source/_static/img/lora_diagram.png -------------------------------------------------------------------------------- /docs/source/_static/img/lora_experiment_loss_curves.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/docs/source/_static/img/lora_experiment_loss_curves.png -------------------------------------------------------------------------------- /docs/source/_static/img/pytorch-logo-dark.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/docs/source/_static/img/pytorch-logo-dark.png -------------------------------------------------------------------------------- /docs/source/_static/img/pytorch-logo-dark.svg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/docs/source/_static/img/pytorch-logo-dark.svg -------------------------------------------------------------------------------- /docs/source/_static/img/pytorch-logo-flame.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/docs/source/_static/img/pytorch-logo-flame.png -------------------------------------------------------------------------------- /docs/source/_static/img/pytorch-logo-flame.svg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/docs/source/_static/img/pytorch-logo-flame.svg -------------------------------------------------------------------------------- /docs/source/_static/img/qat_diagram.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/docs/source/_static/img/qat_diagram.png -------------------------------------------------------------------------------- /docs/source/_static/img/qlora_exp.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/docs/source/_static/img/qlora_exp.png -------------------------------------------------------------------------------- /docs/source/_static/img/torchtune_datasets.svg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/docs/source/_static/img/torchtune_datasets.svg -------------------------------------------------------------------------------- /docs/source/_static/img/torchtune_workspace.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/docs/source/_static/img/torchtune_workspace.png -------------------------------------------------------------------------------- /docs/source/_templates/autosummary/class.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/docs/source/_templates/autosummary/class.rst -------------------------------------------------------------------------------- /docs/source/_templates/autosummary/function.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/docs/source/_templates/autosummary/function.rst -------------------------------------------------------------------------------- /docs/source/_templates/layout.html: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/docs/source/_templates/layout.html -------------------------------------------------------------------------------- /docs/source/api_ref_config.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/docs/source/api_ref_config.rst -------------------------------------------------------------------------------- /docs/source/api_ref_data.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/docs/source/api_ref_data.rst -------------------------------------------------------------------------------- /docs/source/api_ref_datasets.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/docs/source/api_ref_datasets.rst -------------------------------------------------------------------------------- /docs/source/api_ref_generation.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/docs/source/api_ref_generation.rst -------------------------------------------------------------------------------- /docs/source/api_ref_models.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/docs/source/api_ref_models.rst -------------------------------------------------------------------------------- /docs/source/api_ref_modules.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/docs/source/api_ref_modules.rst -------------------------------------------------------------------------------- /docs/source/api_ref_rlhf.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/docs/source/api_ref_rlhf.rst -------------------------------------------------------------------------------- /docs/source/api_ref_training.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/docs/source/api_ref_training.rst -------------------------------------------------------------------------------- /docs/source/api_ref_utilities.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/docs/source/api_ref_utilities.rst -------------------------------------------------------------------------------- /docs/source/basics/chat_datasets.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/docs/source/basics/chat_datasets.rst -------------------------------------------------------------------------------- /docs/source/basics/custom_components.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/docs/source/basics/custom_components.rst -------------------------------------------------------------------------------- /docs/source/basics/datasets_overview.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/docs/source/basics/datasets_overview.rst -------------------------------------------------------------------------------- /docs/source/basics/instruct_datasets.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/docs/source/basics/instruct_datasets.rst -------------------------------------------------------------------------------- /docs/source/basics/message_transforms.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/docs/source/basics/message_transforms.rst -------------------------------------------------------------------------------- /docs/source/basics/messages.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/docs/source/basics/messages.rst -------------------------------------------------------------------------------- /docs/source/basics/model_transforms.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/docs/source/basics/model_transforms.rst -------------------------------------------------------------------------------- /docs/source/basics/multimodal_datasets.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/docs/source/basics/multimodal_datasets.rst -------------------------------------------------------------------------------- /docs/source/basics/packing.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/docs/source/basics/packing.rst -------------------------------------------------------------------------------- /docs/source/basics/preference_datasets.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/docs/source/basics/preference_datasets.rst -------------------------------------------------------------------------------- /docs/source/basics/prompt_templates.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/docs/source/basics/prompt_templates.rst -------------------------------------------------------------------------------- /docs/source/basics/text_completion_datasets.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/docs/source/basics/text_completion_datasets.rst -------------------------------------------------------------------------------- /docs/source/basics/tokenizers.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/docs/source/basics/tokenizers.rst -------------------------------------------------------------------------------- /docs/source/conf.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/docs/source/conf.py -------------------------------------------------------------------------------- /docs/source/custom_directives.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/docs/source/custom_directives.py -------------------------------------------------------------------------------- /docs/source/deep_dives/README.txt: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /docs/source/deep_dives/checkpointer.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/docs/source/deep_dives/checkpointer.rst -------------------------------------------------------------------------------- /docs/source/deep_dives/comet_logging.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/docs/source/deep_dives/comet_logging.rst -------------------------------------------------------------------------------- /docs/source/deep_dives/configs.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/docs/source/deep_dives/configs.rst -------------------------------------------------------------------------------- /docs/source/deep_dives/recipe_deepdive.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/docs/source/deep_dives/recipe_deepdive.rst -------------------------------------------------------------------------------- /docs/source/deep_dives/wandb_logging.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/docs/source/deep_dives/wandb_logging.rst -------------------------------------------------------------------------------- /docs/source/index.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/docs/source/index.rst -------------------------------------------------------------------------------- /docs/source/install.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/docs/source/install.rst -------------------------------------------------------------------------------- /docs/source/overview.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/docs/source/overview.rst -------------------------------------------------------------------------------- /docs/source/recipes/dpo.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/docs/source/recipes/dpo.rst -------------------------------------------------------------------------------- /docs/source/recipes/lora_finetune_single_device.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/docs/source/recipes/lora_finetune_single_device.rst -------------------------------------------------------------------------------- /docs/source/recipes/qat_distributed.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/docs/source/recipes/qat_distributed.rst -------------------------------------------------------------------------------- /docs/source/recipes/recipes_overview.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/docs/source/recipes/recipes_overview.rst -------------------------------------------------------------------------------- /docs/source/tune_cli.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/docs/source/tune_cli.rst -------------------------------------------------------------------------------- /docs/source/tutorials/README.txt: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /docs/source/tutorials/chat.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/docs/source/tutorials/chat.rst -------------------------------------------------------------------------------- /docs/source/tutorials/e2e_flow.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/docs/source/tutorials/e2e_flow.rst -------------------------------------------------------------------------------- /docs/source/tutorials/first_finetune_tutorial.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/docs/source/tutorials/first_finetune_tutorial.rst -------------------------------------------------------------------------------- /docs/source/tutorials/llama3.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/docs/source/tutorials/llama3.rst -------------------------------------------------------------------------------- /docs/source/tutorials/llama_kd_tutorial.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/docs/source/tutorials/llama_kd_tutorial.rst -------------------------------------------------------------------------------- /docs/source/tutorials/lora_finetune.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/docs/source/tutorials/lora_finetune.rst -------------------------------------------------------------------------------- /docs/source/tutorials/memory_optimizations.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/docs/source/tutorials/memory_optimizations.rst -------------------------------------------------------------------------------- /docs/source/tutorials/multinode.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/docs/source/tutorials/multinode.rst -------------------------------------------------------------------------------- /docs/source/tutorials/qat_finetune.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/docs/source/tutorials/qat_finetune.rst -------------------------------------------------------------------------------- /docs/source/tutorials/qlora_finetune.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/docs/source/tutorials/qlora_finetune.rst -------------------------------------------------------------------------------- /pyproject.toml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/pyproject.toml -------------------------------------------------------------------------------- /recipes/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/__init__.py -------------------------------------------------------------------------------- /recipes/configs/dev/11B_lora_multi_dataset.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/dev/11B_lora_multi_dataset.yaml -------------------------------------------------------------------------------- /recipes/configs/dev/3B_full_grpo.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/dev/3B_full_grpo.yaml -------------------------------------------------------------------------------- /recipes/configs/dev/3B_sft_for_grpo.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/dev/3B_sft_for_grpo.yaml -------------------------------------------------------------------------------- /recipes/configs/dev/8B_full_experimental.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/dev/8B_full_experimental.yaml -------------------------------------------------------------------------------- /recipes/configs/dev/qwen3B_async_grpo.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/dev/qwen3B_async_grpo.yaml -------------------------------------------------------------------------------- /recipes/configs/dev/qwen3B_sync_grpo.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/dev/qwen3B_sync_grpo.yaml -------------------------------------------------------------------------------- /recipes/configs/eleuther_evaluation.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/eleuther_evaluation.yaml -------------------------------------------------------------------------------- /recipes/configs/gemma/2B_full.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/gemma/2B_full.yaml -------------------------------------------------------------------------------- /recipes/configs/gemma/2B_lora.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/gemma/2B_lora.yaml -------------------------------------------------------------------------------- /recipes/configs/gemma/2B_lora_single_device.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/gemma/2B_lora_single_device.yaml -------------------------------------------------------------------------------- /recipes/configs/gemma/2B_qlora_single_device.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/gemma/2B_qlora_single_device.yaml -------------------------------------------------------------------------------- /recipes/configs/gemma/7B_full.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/gemma/7B_full.yaml -------------------------------------------------------------------------------- /recipes/configs/gemma/7B_lora.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/gemma/7B_lora.yaml -------------------------------------------------------------------------------- /recipes/configs/gemma/7B_lora_single_device.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/gemma/7B_lora_single_device.yaml -------------------------------------------------------------------------------- /recipes/configs/gemma/7B_qlora_single_device.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/gemma/7B_qlora_single_device.yaml -------------------------------------------------------------------------------- /recipes/configs/gemma/evaluation.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/gemma/evaluation.yaml -------------------------------------------------------------------------------- /recipes/configs/gemma2/27B_full.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/gemma2/27B_full.yaml -------------------------------------------------------------------------------- /recipes/configs/gemma2/27B_lora.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/gemma2/27B_lora.yaml -------------------------------------------------------------------------------- /recipes/configs/gemma2/27B_lora_single_device.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/gemma2/27B_lora_single_device.yaml -------------------------------------------------------------------------------- /recipes/configs/gemma2/27B_qlora_single_device.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/gemma2/27B_qlora_single_device.yaml -------------------------------------------------------------------------------- /recipes/configs/gemma2/2B_full.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/gemma2/2B_full.yaml -------------------------------------------------------------------------------- /recipes/configs/gemma2/2B_lora.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/gemma2/2B_lora.yaml -------------------------------------------------------------------------------- /recipes/configs/gemma2/2B_lora_single_device.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/gemma2/2B_lora_single_device.yaml -------------------------------------------------------------------------------- /recipes/configs/gemma2/2B_qlora_single_device.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/gemma2/2B_qlora_single_device.yaml -------------------------------------------------------------------------------- /recipes/configs/gemma2/9B_full.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/gemma2/9B_full.yaml -------------------------------------------------------------------------------- /recipes/configs/gemma2/9B_lora.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/gemma2/9B_lora.yaml -------------------------------------------------------------------------------- /recipes/configs/gemma2/9B_lora_single_device.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/gemma2/9B_lora_single_device.yaml -------------------------------------------------------------------------------- /recipes/configs/gemma2/9B_qlora_single_device.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/gemma2/9B_qlora_single_device.yaml -------------------------------------------------------------------------------- /recipes/configs/generation.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/generation.yaml -------------------------------------------------------------------------------- /recipes/configs/llama2/13B_full.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama2/13B_full.yaml -------------------------------------------------------------------------------- /recipes/configs/llama2/13B_lora.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama2/13B_lora.yaml -------------------------------------------------------------------------------- /recipes/configs/llama2/13B_qlora_single_device.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama2/13B_qlora_single_device.yaml -------------------------------------------------------------------------------- /recipes/configs/llama2/1B_qat_single_device.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama2/1B_qat_single_device.yaml -------------------------------------------------------------------------------- /recipes/configs/llama2/70B_lora.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama2/70B_lora.yaml -------------------------------------------------------------------------------- /recipes/configs/llama2/70B_qlora.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama2/70B_qlora.yaml -------------------------------------------------------------------------------- /recipes/configs/llama2/7B_full.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama2/7B_full.yaml -------------------------------------------------------------------------------- /recipes/configs/llama2/7B_full_low_memory.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama2/7B_full_low_memory.yaml -------------------------------------------------------------------------------- /recipes/configs/llama2/7B_lora.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama2/7B_lora.yaml -------------------------------------------------------------------------------- /recipes/configs/llama2/7B_lora_dpo.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama2/7B_lora_dpo.yaml -------------------------------------------------------------------------------- /recipes/configs/llama2/7B_lora_dpo_single_device.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama2/7B_lora_dpo_single_device.yaml -------------------------------------------------------------------------------- /recipes/configs/llama2/7B_lora_single_device.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama2/7B_lora_single_device.yaml -------------------------------------------------------------------------------- /recipes/configs/llama2/7B_qat_full.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama2/7B_qat_full.yaml -------------------------------------------------------------------------------- /recipes/configs/llama2/7B_qlora.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama2/7B_qlora.yaml -------------------------------------------------------------------------------- /recipes/configs/llama2/7B_qlora_single_device.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama2/7B_qlora_single_device.yaml -------------------------------------------------------------------------------- /recipes/configs/llama2/generation_v2.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama2/generation_v2.yaml -------------------------------------------------------------------------------- /recipes/configs/llama3/70B_full.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama3/70B_full.yaml -------------------------------------------------------------------------------- /recipes/configs/llama3/70B_generation_distributed.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama3/70B_generation_distributed.yaml -------------------------------------------------------------------------------- /recipes/configs/llama3/70B_lora.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama3/70B_lora.yaml -------------------------------------------------------------------------------- /recipes/configs/llama3/8B_dora.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama3/8B_dora.yaml -------------------------------------------------------------------------------- /recipes/configs/llama3/8B_dora_single_device.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama3/8B_dora_single_device.yaml -------------------------------------------------------------------------------- /recipes/configs/llama3/8B_full.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama3/8B_full.yaml -------------------------------------------------------------------------------- /recipes/configs/llama3/8B_full_single_device.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama3/8B_full_single_device.yaml -------------------------------------------------------------------------------- /recipes/configs/llama3/8B_lora.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama3/8B_lora.yaml -------------------------------------------------------------------------------- /recipes/configs/llama3/8B_lora_single_device.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama3/8B_lora_single_device.yaml -------------------------------------------------------------------------------- /recipes/configs/llama3/8B_qat_full.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama3/8B_qat_full.yaml -------------------------------------------------------------------------------- /recipes/configs/llama3/8B_qat_lora.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama3/8B_qat_lora.yaml -------------------------------------------------------------------------------- /recipes/configs/llama3/8B_qdora_single_device.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama3/8B_qdora_single_device.yaml -------------------------------------------------------------------------------- /recipes/configs/llama3/8B_qlora_single_device.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama3/8B_qlora_single_device.yaml -------------------------------------------------------------------------------- /recipes/configs/llama3_1/405B_qlora.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama3_1/405B_qlora.yaml -------------------------------------------------------------------------------- /recipes/configs/llama3_1/70B_full.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama3_1/70B_full.yaml -------------------------------------------------------------------------------- /recipes/configs/llama3_1/70B_generation_distributed.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama3_1/70B_generation_distributed.yaml -------------------------------------------------------------------------------- /recipes/configs/llama3_1/70B_lora.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama3_1/70B_lora.yaml -------------------------------------------------------------------------------- /recipes/configs/llama3_1/8B_full.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama3_1/8B_full.yaml -------------------------------------------------------------------------------- /recipes/configs/llama3_1/8B_full_dpo.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama3_1/8B_full_dpo.yaml -------------------------------------------------------------------------------- /recipes/configs/llama3_1/8B_full_single_device.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama3_1/8B_full_single_device.yaml -------------------------------------------------------------------------------- /recipes/configs/llama3_1/8B_lora.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama3_1/8B_lora.yaml -------------------------------------------------------------------------------- /recipes/configs/llama3_1/8B_lora_dpo.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama3_1/8B_lora_dpo.yaml -------------------------------------------------------------------------------- /recipes/configs/llama3_1/8B_lora_dpo_single_device.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama3_1/8B_lora_dpo_single_device.yaml -------------------------------------------------------------------------------- /recipes/configs/llama3_1/8B_lora_single_device.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama3_1/8B_lora_single_device.yaml -------------------------------------------------------------------------------- /recipes/configs/llama3_1/8B_qat_full.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama3_1/8B_qat_full.yaml -------------------------------------------------------------------------------- /recipes/configs/llama3_1/8B_qat_lora.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama3_1/8B_qat_lora.yaml -------------------------------------------------------------------------------- /recipes/configs/llama3_1/8B_qlora_single_device.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama3_1/8B_qlora_single_device.yaml -------------------------------------------------------------------------------- /recipes/configs/llama3_1/evaluation.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama3_1/evaluation.yaml -------------------------------------------------------------------------------- /recipes/configs/llama3_2/1B_full.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama3_2/1B_full.yaml -------------------------------------------------------------------------------- /recipes/configs/llama3_2/1B_full_single_device.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama3_2/1B_full_single_device.yaml -------------------------------------------------------------------------------- /recipes/configs/llama3_2/1B_lora.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama3_2/1B_lora.yaml -------------------------------------------------------------------------------- /recipes/configs/llama3_2/1B_lora_single_device.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama3_2/1B_lora_single_device.yaml -------------------------------------------------------------------------------- /recipes/configs/llama3_2/1B_qat_lora.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama3_2/1B_qat_lora.yaml -------------------------------------------------------------------------------- /recipes/configs/llama3_2/1B_qlora_single_device.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama3_2/1B_qlora_single_device.yaml -------------------------------------------------------------------------------- /recipes/configs/llama3_2/3B_full.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama3_2/3B_full.yaml -------------------------------------------------------------------------------- /recipes/configs/llama3_2/3B_full_single_device.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama3_2/3B_full_single_device.yaml -------------------------------------------------------------------------------- /recipes/configs/llama3_2/3B_lora.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama3_2/3B_lora.yaml -------------------------------------------------------------------------------- /recipes/configs/llama3_2/3B_lora_single_device.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama3_2/3B_lora_single_device.yaml -------------------------------------------------------------------------------- /recipes/configs/llama3_2/3B_qat_full.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama3_2/3B_qat_full.yaml -------------------------------------------------------------------------------- /recipes/configs/llama3_2/3B_qat_lora.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama3_2/3B_qat_lora.yaml -------------------------------------------------------------------------------- /recipes/configs/llama3_2/3B_qlora_single_device.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama3_2/3B_qlora_single_device.yaml -------------------------------------------------------------------------------- /recipes/configs/llama3_2/8B_to_1B_KD_lora_distributed.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama3_2/8B_to_1B_KD_lora_distributed.yaml -------------------------------------------------------------------------------- /recipes/configs/llama3_2/8B_to_1B_KD_lora_single_device.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama3_2/8B_to_1B_KD_lora_single_device.yaml -------------------------------------------------------------------------------- /recipes/configs/llama3_2/evaluation.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama3_2/evaluation.yaml -------------------------------------------------------------------------------- /recipes/configs/llama3_2_vision/11B_evaluation.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama3_2_vision/11B_evaluation.yaml -------------------------------------------------------------------------------- /recipes/configs/llama3_2_vision/11B_full.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama3_2_vision/11B_full.yaml -------------------------------------------------------------------------------- /recipes/configs/llama3_2_vision/11B_full_single_device.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama3_2_vision/11B_full_single_device.yaml -------------------------------------------------------------------------------- /recipes/configs/llama3_2_vision/11B_generation_v2.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama3_2_vision/11B_generation_v2.yaml -------------------------------------------------------------------------------- /recipes/configs/llama3_2_vision/11B_lora.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama3_2_vision/11B_lora.yaml -------------------------------------------------------------------------------- /recipes/configs/llama3_2_vision/11B_lora_single_device.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama3_2_vision/11B_lora_single_device.yaml -------------------------------------------------------------------------------- /recipes/configs/llama3_2_vision/11B_qlora.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama3_2_vision/11B_qlora.yaml -------------------------------------------------------------------------------- /recipes/configs/llama3_2_vision/11B_qlora_single_device.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama3_2_vision/11B_qlora_single_device.yaml -------------------------------------------------------------------------------- /recipes/configs/llama3_2_vision/90B_full.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama3_2_vision/90B_full.yaml -------------------------------------------------------------------------------- /recipes/configs/llama3_2_vision/90B_lora.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama3_2_vision/90B_lora.yaml -------------------------------------------------------------------------------- /recipes/configs/llama3_2_vision/90B_qlora.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama3_2_vision/90B_qlora.yaml -------------------------------------------------------------------------------- /recipes/configs/llama3_3/70B_full.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama3_3/70B_full.yaml -------------------------------------------------------------------------------- /recipes/configs/llama3_3/70B_full_multinode.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama3_3/70B_full_multinode.yaml -------------------------------------------------------------------------------- /recipes/configs/llama3_3/70B_generation_distributed.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama3_3/70B_generation_distributed.yaml -------------------------------------------------------------------------------- /recipes/configs/llama3_3/70B_lora.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama3_3/70B_lora.yaml -------------------------------------------------------------------------------- /recipes/configs/llama3_3/70B_qlora.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama3_3/70B_qlora.yaml -------------------------------------------------------------------------------- /recipes/configs/llama4/maverick_17B_128E_full.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama4/maverick_17B_128E_full.yaml -------------------------------------------------------------------------------- /recipes/configs/llama4/scout_17B_16E_full.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama4/scout_17B_16E_full.yaml -------------------------------------------------------------------------------- /recipes/configs/llama4/scout_17B_16E_lora.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/llama4/scout_17B_16E_lora.yaml -------------------------------------------------------------------------------- /recipes/configs/mistral/7B_full.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/mistral/7B_full.yaml -------------------------------------------------------------------------------- /recipes/configs/mistral/7B_full_low_memory.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/mistral/7B_full_low_memory.yaml -------------------------------------------------------------------------------- /recipes/configs/mistral/7B_full_ppo_low_memory.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/mistral/7B_full_ppo_low_memory.yaml -------------------------------------------------------------------------------- /recipes/configs/mistral/7B_lora.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/mistral/7B_lora.yaml -------------------------------------------------------------------------------- /recipes/configs/mistral/7B_lora_single_device.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/mistral/7B_lora_single_device.yaml -------------------------------------------------------------------------------- /recipes/configs/mistral/7B_qlora_single_device.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/mistral/7B_qlora_single_device.yaml -------------------------------------------------------------------------------- /recipes/configs/mistral/evaluation.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/mistral/evaluation.yaml -------------------------------------------------------------------------------- /recipes/configs/phi3/evaluation.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/phi3/evaluation.yaml -------------------------------------------------------------------------------- /recipes/configs/phi3/mini_full.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/phi3/mini_full.yaml -------------------------------------------------------------------------------- /recipes/configs/phi3/mini_full_low_memory.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/phi3/mini_full_low_memory.yaml -------------------------------------------------------------------------------- /recipes/configs/phi3/mini_lora.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/phi3/mini_lora.yaml -------------------------------------------------------------------------------- /recipes/configs/phi3/mini_lora_single_device.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/phi3/mini_lora_single_device.yaml -------------------------------------------------------------------------------- /recipes/configs/phi3/mini_qlora_single_device.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/phi3/mini_qlora_single_device.yaml -------------------------------------------------------------------------------- /recipes/configs/phi4/14B_full.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/phi4/14B_full.yaml -------------------------------------------------------------------------------- /recipes/configs/phi4/14B_full_low_memory.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/phi4/14B_full_low_memory.yaml -------------------------------------------------------------------------------- /recipes/configs/phi4/14B_lora.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/phi4/14B_lora.yaml -------------------------------------------------------------------------------- /recipes/configs/phi4/14B_lora_single_device.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/phi4/14B_lora_single_device.yaml -------------------------------------------------------------------------------- /recipes/configs/phi4/14B_qlora_single_device.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/phi4/14B_qlora_single_device.yaml -------------------------------------------------------------------------------- /recipes/configs/phi4/evaluation.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/phi4/evaluation.yaml -------------------------------------------------------------------------------- /recipes/configs/quantization.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/quantization.yaml -------------------------------------------------------------------------------- /recipes/configs/qwen2/0.5B_full.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/qwen2/0.5B_full.yaml -------------------------------------------------------------------------------- /recipes/configs/qwen2/0.5B_full_single_device.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/qwen2/0.5B_full_single_device.yaml -------------------------------------------------------------------------------- /recipes/configs/qwen2/0.5B_lora.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/qwen2/0.5B_lora.yaml -------------------------------------------------------------------------------- /recipes/configs/qwen2/0.5B_lora_single_device.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/qwen2/0.5B_lora_single_device.yaml -------------------------------------------------------------------------------- /recipes/configs/qwen2/1.5B_full.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/qwen2/1.5B_full.yaml -------------------------------------------------------------------------------- /recipes/configs/qwen2/1.5B_full_single_device.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/qwen2/1.5B_full_single_device.yaml -------------------------------------------------------------------------------- /recipes/configs/qwen2/1.5B_lora.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/qwen2/1.5B_lora.yaml -------------------------------------------------------------------------------- /recipes/configs/qwen2/1.5B_lora_single_device.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/qwen2/1.5B_lora_single_device.yaml -------------------------------------------------------------------------------- /recipes/configs/qwen2/1.5_to_0.5B_KD_lora_distributed.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/qwen2/1.5_to_0.5B_KD_lora_distributed.yaml -------------------------------------------------------------------------------- /recipes/configs/qwen2/1.5_to_0.5B_KD_lora_single_device.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/qwen2/1.5_to_0.5B_KD_lora_single_device.yaml -------------------------------------------------------------------------------- /recipes/configs/qwen2/7B_full.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/qwen2/7B_full.yaml -------------------------------------------------------------------------------- /recipes/configs/qwen2/7B_full_single_device.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/qwen2/7B_full_single_device.yaml -------------------------------------------------------------------------------- /recipes/configs/qwen2/7B_lora.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/qwen2/7B_lora.yaml -------------------------------------------------------------------------------- /recipes/configs/qwen2/7B_lora_single_device.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/qwen2/7B_lora_single_device.yaml -------------------------------------------------------------------------------- /recipes/configs/qwen2/evaluation.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/qwen2/evaluation.yaml -------------------------------------------------------------------------------- /recipes/configs/qwen2_5/0.5B_full.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/qwen2_5/0.5B_full.yaml -------------------------------------------------------------------------------- /recipes/configs/qwen2_5/0.5B_full_single_device.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/qwen2_5/0.5B_full_single_device.yaml -------------------------------------------------------------------------------- /recipes/configs/qwen2_5/0.5B_lora.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/qwen2_5/0.5B_lora.yaml -------------------------------------------------------------------------------- /recipes/configs/qwen2_5/0.5B_lora_single_device.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/qwen2_5/0.5B_lora_single_device.yaml -------------------------------------------------------------------------------- /recipes/configs/qwen2_5/1.5B_full.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/qwen2_5/1.5B_full.yaml -------------------------------------------------------------------------------- /recipes/configs/qwen2_5/1.5B_full_single_device.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/qwen2_5/1.5B_full_single_device.yaml -------------------------------------------------------------------------------- /recipes/configs/qwen2_5/1.5B_lora.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/qwen2_5/1.5B_lora.yaml -------------------------------------------------------------------------------- /recipes/configs/qwen2_5/1.5B_lora_single_device.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/qwen2_5/1.5B_lora_single_device.yaml -------------------------------------------------------------------------------- /recipes/configs/qwen2_5/1.5B_qat_single_device.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/qwen2_5/1.5B_qat_single_device.yaml -------------------------------------------------------------------------------- /recipes/configs/qwen2_5/14B_lora_single_device.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/qwen2_5/14B_lora_single_device.yaml -------------------------------------------------------------------------------- /recipes/configs/qwen2_5/14B_to_7B_KD_lora_single_device.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/qwen2_5/14B_to_7B_KD_lora_single_device.yaml -------------------------------------------------------------------------------- /recipes/configs/qwen2_5/32B_lora.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/qwen2_5/32B_lora.yaml -------------------------------------------------------------------------------- /recipes/configs/qwen2_5/3B_full.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/qwen2_5/3B_full.yaml -------------------------------------------------------------------------------- /recipes/configs/qwen2_5/3B_full_single_device.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/qwen2_5/3B_full_single_device.yaml -------------------------------------------------------------------------------- /recipes/configs/qwen2_5/3B_lora.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/qwen2_5/3B_lora.yaml -------------------------------------------------------------------------------- /recipes/configs/qwen2_5/3B_lora_single_device.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/qwen2_5/3B_lora_single_device.yaml -------------------------------------------------------------------------------- /recipes/configs/qwen2_5/3B_qat_single_device.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/qwen2_5/3B_qat_single_device.yaml -------------------------------------------------------------------------------- /recipes/configs/qwen2_5/72B_lora.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/qwen2_5/72B_lora.yaml -------------------------------------------------------------------------------- /recipes/configs/qwen2_5/7B_full.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/qwen2_5/7B_full.yaml -------------------------------------------------------------------------------- /recipes/configs/qwen2_5/7B_full_single_device.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/qwen2_5/7B_full_single_device.yaml -------------------------------------------------------------------------------- /recipes/configs/qwen2_5/7B_lora.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/qwen2_5/7B_lora.yaml -------------------------------------------------------------------------------- /recipes/configs/qwen2_5/7B_lora_single_device.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/qwen2_5/7B_lora_single_device.yaml -------------------------------------------------------------------------------- /recipes/configs/qwen2_5/evaluation.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/qwen2_5/evaluation.yaml -------------------------------------------------------------------------------- /recipes/configs/qwen3/0.6B_full.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/qwen3/0.6B_full.yaml -------------------------------------------------------------------------------- /recipes/configs/qwen3/0.6B_full_single_device.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/qwen3/0.6B_full_single_device.yaml -------------------------------------------------------------------------------- /recipes/configs/qwen3/0.6B_lora.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/qwen3/0.6B_lora.yaml -------------------------------------------------------------------------------- /recipes/configs/qwen3/0.6B_lora_single_device.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/qwen3/0.6B_lora_single_device.yaml -------------------------------------------------------------------------------- /recipes/configs/qwen3/0.6B_qat_full.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/qwen3/0.6B_qat_full.yaml -------------------------------------------------------------------------------- /recipes/configs/qwen3/0.6B_qat_lora.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/qwen3/0.6B_qat_lora.yaml -------------------------------------------------------------------------------- /recipes/configs/qwen3/1.7B_full.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/qwen3/1.7B_full.yaml -------------------------------------------------------------------------------- /recipes/configs/qwen3/1.7B_full_single_device.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/qwen3/1.7B_full_single_device.yaml -------------------------------------------------------------------------------- /recipes/configs/qwen3/1.7B_lora.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/qwen3/1.7B_lora.yaml -------------------------------------------------------------------------------- /recipes/configs/qwen3/1.7B_lora_single_device.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/qwen3/1.7B_lora_single_device.yaml -------------------------------------------------------------------------------- /recipes/configs/qwen3/1.7B_qat_full.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/qwen3/1.7B_qat_full.yaml -------------------------------------------------------------------------------- /recipes/configs/qwen3/1.7B_qat_lora.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/qwen3/1.7B_qat_lora.yaml -------------------------------------------------------------------------------- /recipes/configs/qwen3/14B_lora_single_device.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/qwen3/14B_lora_single_device.yaml -------------------------------------------------------------------------------- /recipes/configs/qwen3/14B_to_8B_KD_lora_single_device.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/qwen3/14B_to_8B_KD_lora_single_device.yaml -------------------------------------------------------------------------------- /recipes/configs/qwen3/32B_lora.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/qwen3/32B_lora.yaml -------------------------------------------------------------------------------- /recipes/configs/qwen3/32B_qat_lora.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/qwen3/32B_qat_lora.yaml -------------------------------------------------------------------------------- /recipes/configs/qwen3/4B_full.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/qwen3/4B_full.yaml -------------------------------------------------------------------------------- /recipes/configs/qwen3/4B_full_single_device.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/qwen3/4B_full_single_device.yaml -------------------------------------------------------------------------------- /recipes/configs/qwen3/4B_lora.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/qwen3/4B_lora.yaml -------------------------------------------------------------------------------- /recipes/configs/qwen3/4B_lora_single_device.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/qwen3/4B_lora_single_device.yaml -------------------------------------------------------------------------------- /recipes/configs/qwen3/4B_qat_full.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/qwen3/4B_qat_full.yaml -------------------------------------------------------------------------------- /recipes/configs/qwen3/4B_qat_lora.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/qwen3/4B_qat_lora.yaml -------------------------------------------------------------------------------- /recipes/configs/qwen3/8B_full.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/qwen3/8B_full.yaml -------------------------------------------------------------------------------- /recipes/configs/qwen3/8B_full_single_device.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/qwen3/8B_full_single_device.yaml -------------------------------------------------------------------------------- /recipes/configs/qwen3/8B_lora.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/qwen3/8B_lora.yaml -------------------------------------------------------------------------------- /recipes/configs/qwen3/8B_lora_single_device.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/qwen3/8B_lora_single_device.yaml -------------------------------------------------------------------------------- /recipes/configs/qwen3/8B_qat_full.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/qwen3/8B_qat_full.yaml -------------------------------------------------------------------------------- /recipes/configs/qwen3/8B_qat_lora.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/qwen3/8B_qat_lora.yaml -------------------------------------------------------------------------------- /recipes/configs/qwen3/evaluation.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/configs/qwen3/evaluation.yaml -------------------------------------------------------------------------------- /recipes/dev/7B_full_early_exit.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/dev/7B_full_early_exit.yaml -------------------------------------------------------------------------------- /recipes/dev/async_grpo.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/dev/async_grpo.md -------------------------------------------------------------------------------- /recipes/dev/async_grpo_full_finetune_distributed.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/dev/async_grpo_full_finetune_distributed.py -------------------------------------------------------------------------------- /recipes/dev/early_exit_finetune_distributed.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/dev/early_exit_finetune_distributed.py -------------------------------------------------------------------------------- /recipes/dev/generate_v2.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/dev/generate_v2.py -------------------------------------------------------------------------------- /recipes/dev/generate_v2_distributed.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/dev/generate_v2_distributed.py -------------------------------------------------------------------------------- /recipes/dev/grpo_full_finetune_distributed.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/dev/grpo_full_finetune_distributed.py -------------------------------------------------------------------------------- /recipes/dev/gsm8k_sft.sbatch: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/dev/gsm8k_sft.sbatch -------------------------------------------------------------------------------- /recipes/dev/lora_finetune_distributed_multi_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/dev/lora_finetune_distributed_multi_dataset.py -------------------------------------------------------------------------------- /recipes/dev/multinode_grpo.sbatch: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/dev/multinode_grpo.sbatch -------------------------------------------------------------------------------- /recipes/eleuther_eval.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/eleuther_eval.py -------------------------------------------------------------------------------- /recipes/full_dpo_distributed.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/full_dpo_distributed.py -------------------------------------------------------------------------------- /recipes/full_finetune_distributed.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/full_finetune_distributed.py -------------------------------------------------------------------------------- /recipes/full_finetune_multinode.slurm: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/full_finetune_multinode.slurm -------------------------------------------------------------------------------- /recipes/full_finetune_single_device.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/full_finetune_single_device.py -------------------------------------------------------------------------------- /recipes/generate.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/generate.py -------------------------------------------------------------------------------- /recipes/knowledge_distillation_distributed.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/knowledge_distillation_distributed.py -------------------------------------------------------------------------------- /recipes/knowledge_distillation_single_device.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/knowledge_distillation_single_device.py -------------------------------------------------------------------------------- /recipes/lora_dpo_distributed.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/lora_dpo_distributed.py -------------------------------------------------------------------------------- /recipes/lora_dpo_single_device.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/lora_dpo_single_device.py -------------------------------------------------------------------------------- /recipes/lora_finetune_distributed.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/lora_finetune_distributed.py -------------------------------------------------------------------------------- /recipes/lora_finetune_single_device.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/lora_finetune_single_device.py -------------------------------------------------------------------------------- /recipes/ppo_full_finetune_single_device.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/ppo_full_finetune_single_device.py -------------------------------------------------------------------------------- /recipes/qat_distributed.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/qat_distributed.py -------------------------------------------------------------------------------- /recipes/qat_lora_finetune_distributed.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/qat_lora_finetune_distributed.py -------------------------------------------------------------------------------- /recipes/qat_single_device.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/qat_single_device.py -------------------------------------------------------------------------------- /recipes/quantization.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/quantization.md -------------------------------------------------------------------------------- /recipes/quantize.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/recipes/quantize.py -------------------------------------------------------------------------------- /tests/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/__init__.py -------------------------------------------------------------------------------- /tests/assets/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/assets/README.md -------------------------------------------------------------------------------- /tests/assets/alpaca_tiny.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/assets/alpaca_tiny.json -------------------------------------------------------------------------------- /tests/assets/chat_tiny.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/assets/chat_tiny.json -------------------------------------------------------------------------------- /tests/assets/dog_on_skateboard.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/assets/dog_on_skateboard.jpg -------------------------------------------------------------------------------- /tests/assets/generation_config.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/assets/generation_config.json -------------------------------------------------------------------------------- /tests/assets/generation_config_gemma.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/assets/generation_config_gemma.json -------------------------------------------------------------------------------- /tests/assets/hh_rlhf_tiny.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/assets/hh_rlhf_tiny.json -------------------------------------------------------------------------------- /tests/assets/instruct_tiny.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/assets/instruct_tiny.json -------------------------------------------------------------------------------- /tests/assets/invalid_dummy_config.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/assets/invalid_dummy_config.yaml -------------------------------------------------------------------------------- /tests/assets/m.model: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/assets/m.model -------------------------------------------------------------------------------- /tests/assets/merges.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/assets/merges.txt -------------------------------------------------------------------------------- /tests/assets/rgb_pytorch.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/assets/rgb_pytorch.png -------------------------------------------------------------------------------- /tests/assets/sentencepiece.model: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/assets/sentencepiece.model -------------------------------------------------------------------------------- /tests/assets/stack_exchange_paired_tiny.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/assets/stack_exchange_paired_tiny.json -------------------------------------------------------------------------------- /tests/assets/tiktoken_small.model: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/assets/tiktoken_small.model -------------------------------------------------------------------------------- /tests/assets/tiktoken_small_llama4.model: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/assets/tiktoken_small_llama4.model -------------------------------------------------------------------------------- /tests/assets/tiny_bpe_merges.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/assets/tiny_bpe_merges.txt -------------------------------------------------------------------------------- /tests/assets/tiny_bpe_tokenizer.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/assets/tiny_bpe_tokenizer.json -------------------------------------------------------------------------------- /tests/assets/tiny_bpe_vocab.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/assets/tiny_bpe_vocab.json -------------------------------------------------------------------------------- /tests/assets/tiny_fair_checkpoint.pt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/assets/tiny_fair_checkpoint.pt -------------------------------------------------------------------------------- /tests/assets/tiny_state_dict_with_one_key.pt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/assets/tiny_state_dict_with_one_key.pt -------------------------------------------------------------------------------- /tests/assets/tokenizer.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/assets/tokenizer.json -------------------------------------------------------------------------------- /tests/assets/tokenizer_config.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/assets/tokenizer_config.json -------------------------------------------------------------------------------- /tests/assets/tokenizer_config_gemma.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/assets/tokenizer_config_gemma.json -------------------------------------------------------------------------------- /tests/assets/tokenizer_gemma_cropped.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/assets/tokenizer_gemma_cropped.json -------------------------------------------------------------------------------- /tests/assets/valid_dummy_config.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/assets/valid_dummy_config.yaml -------------------------------------------------------------------------------- /tests/assets/vocab.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/assets/vocab.json -------------------------------------------------------------------------------- /tests/assets/vqa_tiny.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/assets/vqa_tiny.json -------------------------------------------------------------------------------- /tests/cache_artifacts.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/cache_artifacts.sh -------------------------------------------------------------------------------- /tests/common.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/common.py -------------------------------------------------------------------------------- /tests/conftest.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/conftest.py -------------------------------------------------------------------------------- /tests/recipes/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/recipes/__init__.py -------------------------------------------------------------------------------- /tests/recipes/common.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/recipes/common.py -------------------------------------------------------------------------------- /tests/recipes/dev/test_async_grpo_full_finetune_distributed.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/recipes/dev/test_async_grpo_full_finetune_distributed.py -------------------------------------------------------------------------------- /tests/recipes/dev/test_generate_v2.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/recipes/dev/test_generate_v2.py -------------------------------------------------------------------------------- /tests/recipes/test_configs.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/recipes/test_configs.py -------------------------------------------------------------------------------- /tests/recipes/test_eleuther_eval.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/recipes/test_eleuther_eval.py -------------------------------------------------------------------------------- /tests/recipes/test_full_dpo_distributed.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/recipes/test_full_dpo_distributed.py -------------------------------------------------------------------------------- /tests/recipes/test_full_finetune_distributed.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/recipes/test_full_finetune_distributed.py -------------------------------------------------------------------------------- /tests/recipes/test_full_finetune_single_device.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/recipes/test_full_finetune_single_device.py -------------------------------------------------------------------------------- /tests/recipes/test_knowledge_distillation_distributed.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/recipes/test_knowledge_distillation_distributed.py -------------------------------------------------------------------------------- /tests/recipes/test_knowledge_distillation_single_device.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/recipes/test_knowledge_distillation_single_device.py -------------------------------------------------------------------------------- /tests/recipes/test_lora_dpo_distributed.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/recipes/test_lora_dpo_distributed.py -------------------------------------------------------------------------------- /tests/recipes/test_lora_dpo_single_device.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/recipes/test_lora_dpo_single_device.py -------------------------------------------------------------------------------- /tests/recipes/test_lora_finetune_distributed.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/recipes/test_lora_finetune_distributed.py -------------------------------------------------------------------------------- /tests/recipes/test_lora_finetune_single_device.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/recipes/test_lora_finetune_single_device.py -------------------------------------------------------------------------------- /tests/recipes/test_ppo_full_finetune_single_device.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/recipes/test_ppo_full_finetune_single_device.py -------------------------------------------------------------------------------- /tests/recipes/test_qat_distributed.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/recipes/test_qat_distributed.py -------------------------------------------------------------------------------- /tests/recipes/test_qat_lora_finetune_distributed.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/recipes/test_qat_lora_finetune_distributed.py -------------------------------------------------------------------------------- /tests/recipes/test_qat_single_device.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/recipes/test_qat_single_device.py -------------------------------------------------------------------------------- /tests/recipes/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/recipes/utils.py -------------------------------------------------------------------------------- /tests/regression_tests/test_llama2_7b.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/regression_tests/test_llama2_7b.py -------------------------------------------------------------------------------- /tests/test_import_recipes.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/test_import_recipes.py -------------------------------------------------------------------------------- /tests/test_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/test_utils.py -------------------------------------------------------------------------------- /tests/torchtune/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/__init__.py -------------------------------------------------------------------------------- /tests/torchtune/_cli/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/_cli/__init__.py -------------------------------------------------------------------------------- /tests/torchtune/_cli/test_cat.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/_cli/test_cat.py -------------------------------------------------------------------------------- /tests/torchtune/_cli/test_cp.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/_cli/test_cp.py -------------------------------------------------------------------------------- /tests/torchtune/_cli/test_download.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/_cli/test_download.py -------------------------------------------------------------------------------- /tests/torchtune/_cli/test_ls.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/_cli/test_ls.py -------------------------------------------------------------------------------- /tests/torchtune/_cli/test_run.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/_cli/test_run.py -------------------------------------------------------------------------------- /tests/torchtune/_cli/test_tune.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/_cli/test_tune.py -------------------------------------------------------------------------------- /tests/torchtune/_cli/test_validate.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/_cli/test_validate.py -------------------------------------------------------------------------------- /tests/torchtune/config/test_config_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/config/test_config_utils.py -------------------------------------------------------------------------------- /tests/torchtune/config/test_instantiate.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/config/test_instantiate.py -------------------------------------------------------------------------------- /tests/torchtune/config/test_parse.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/config/test_parse.py -------------------------------------------------------------------------------- /tests/torchtune/config/test_validate.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/config/test_validate.py -------------------------------------------------------------------------------- /tests/torchtune/data/test_collate.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/data/test_collate.py -------------------------------------------------------------------------------- /tests/torchtune/data/test_data_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/data/test_data_utils.py -------------------------------------------------------------------------------- /tests/torchtune/data/test_messages.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/data/test_messages.py -------------------------------------------------------------------------------- /tests/torchtune/data/test_prompt_templates.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/data/test_prompt_templates.py -------------------------------------------------------------------------------- /tests/torchtune/datasets/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/datasets/__init__.py -------------------------------------------------------------------------------- /tests/torchtune/datasets/multimodal/test_vqa_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/datasets/multimodal/test_vqa_dataset.py -------------------------------------------------------------------------------- /tests/torchtune/datasets/test_alpaca_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/datasets/test_alpaca_dataset.py -------------------------------------------------------------------------------- /tests/torchtune/datasets/test_chat_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/datasets/test_chat_dataset.py -------------------------------------------------------------------------------- /tests/torchtune/datasets/test_cnn_dailymail_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/datasets/test_cnn_dailymail_dataset.py -------------------------------------------------------------------------------- /tests/torchtune/datasets/test_concat_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/datasets/test_concat_dataset.py -------------------------------------------------------------------------------- /tests/torchtune/datasets/test_grammar_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/datasets/test_grammar_dataset.py -------------------------------------------------------------------------------- /tests/torchtune/datasets/test_hh_rlhf_helpful_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/datasets/test_hh_rlhf_helpful_dataset.py -------------------------------------------------------------------------------- /tests/torchtune/datasets/test_instruct_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/datasets/test_instruct_dataset.py -------------------------------------------------------------------------------- /tests/torchtune/datasets/test_packed_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/datasets/test_packed_dataset.py -------------------------------------------------------------------------------- /tests/torchtune/datasets/test_preference_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/datasets/test_preference_dataset.py -------------------------------------------------------------------------------- /tests/torchtune/datasets/test_samsum_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/datasets/test_samsum_dataset.py -------------------------------------------------------------------------------- /tests/torchtune/datasets/test_sft_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/datasets/test_sft_dataset.py -------------------------------------------------------------------------------- /tests/torchtune/datasets/test_slimorca_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/datasets/test_slimorca_dataset.py -------------------------------------------------------------------------------- /tests/torchtune/datasets/test_stack_exchange_paired_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/datasets/test_stack_exchange_paired_dataset.py -------------------------------------------------------------------------------- /tests/torchtune/datasets/test_text_completion_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/datasets/test_text_completion_dataset.py -------------------------------------------------------------------------------- /tests/torchtune/datasets/test_wikitext_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/datasets/test_wikitext_dataset.py -------------------------------------------------------------------------------- /tests/torchtune/dev/rl/rewards/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/dev/rl/rewards/__init__.py -------------------------------------------------------------------------------- /tests/torchtune/dev/rl/rewards/test_rewards.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/dev/rl/rewards/test_rewards.py -------------------------------------------------------------------------------- /tests/torchtune/dev/rl/workers/test_postprocessing.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/dev/rl/workers/test_postprocessing.py -------------------------------------------------------------------------------- /tests/torchtune/generation/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/generation/__init__.py -------------------------------------------------------------------------------- /tests/torchtune/generation/test_generation.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/generation/test_generation.py -------------------------------------------------------------------------------- /tests/torchtune/models/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/models/__init__.py -------------------------------------------------------------------------------- /tests/torchtune/models/clip/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/models/clip/__init__.py -------------------------------------------------------------------------------- /tests/torchtune/models/clip/test_clip_image_transform.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/models/clip/test_clip_image_transform.py -------------------------------------------------------------------------------- /tests/torchtune/models/clip/test_clip_text_encoder.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/models/clip/test_clip_text_encoder.py -------------------------------------------------------------------------------- /tests/torchtune/models/clip/test_clip_tokenizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/models/clip/test_clip_tokenizer.py -------------------------------------------------------------------------------- /tests/torchtune/models/clip/test_positional_embeddings.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/models/clip/test_positional_embeddings.py -------------------------------------------------------------------------------- /tests/torchtune/models/flamingo/test_flamingo_decoder.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/models/flamingo/test_flamingo_decoder.py -------------------------------------------------------------------------------- /tests/torchtune/models/flamingo/test_flamingo_encoder.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/models/flamingo/test_flamingo_encoder.py -------------------------------------------------------------------------------- /tests/torchtune/models/flux/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/models/flux/__init__.py -------------------------------------------------------------------------------- /tests/torchtune/models/flux/test_flux_autoencoder.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/models/flux/test_flux_autoencoder.py -------------------------------------------------------------------------------- /tests/torchtune/models/gemma/test_gemma_tokenizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/models/gemma/test_gemma_tokenizer.py -------------------------------------------------------------------------------- /tests/torchtune/models/llama2/scripts/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/models/llama2/scripts/README.md -------------------------------------------------------------------------------- /tests/torchtune/models/llama2/scripts/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/models/llama2/scripts/__init__.py -------------------------------------------------------------------------------- /tests/torchtune/models/llama2/scripts/compare_attention.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/models/llama2/scripts/compare_attention.py -------------------------------------------------------------------------------- /tests/torchtune/models/llama2/scripts/compare_decoder.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/models/llama2/scripts/compare_decoder.py -------------------------------------------------------------------------------- /tests/torchtune/models/llama2/scripts/compare_decoder_layer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/models/llama2/scripts/compare_decoder_layer.py -------------------------------------------------------------------------------- /tests/torchtune/models/llama2/scripts/compare_dora.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/models/llama2/scripts/compare_dora.py -------------------------------------------------------------------------------- /tests/torchtune/models/llama2/scripts/compare_lora.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/models/llama2/scripts/compare_lora.py -------------------------------------------------------------------------------- /tests/torchtune/models/llama2/scripts/compare_lora_llama2.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/models/llama2/scripts/compare_lora_llama2.py -------------------------------------------------------------------------------- /tests/torchtune/models/llama2/test_llama2_prompt_template.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/models/llama2/test_llama2_prompt_template.py -------------------------------------------------------------------------------- /tests/torchtune/models/llama2/test_llama2_tokenizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/models/llama2/test_llama2_tokenizer.py -------------------------------------------------------------------------------- /tests/torchtune/models/llama2/test_lora_llama2.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/models/llama2/test_lora_llama2.py -------------------------------------------------------------------------------- /tests/torchtune/models/llama3/test_llama3.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/models/llama3/test_llama3.py -------------------------------------------------------------------------------- /tests/torchtune/models/llama3/test_llama3_tokenizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/models/llama3/test_llama3_tokenizer.py -------------------------------------------------------------------------------- /tests/torchtune/models/llama3_1/test_position_embeddings.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/models/llama3_1/test_position_embeddings.py -------------------------------------------------------------------------------- /tests/torchtune/models/llama4/test_llama4.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/models/llama4/test_llama4.py -------------------------------------------------------------------------------- /tests/torchtune/models/llama4/test_llama4_tokenizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/models/llama4/test_llama4_tokenizer.py -------------------------------------------------------------------------------- /tests/torchtune/models/llama4/test_llama4_transform.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/models/llama4/test_llama4_transform.py -------------------------------------------------------------------------------- /tests/torchtune/models/mistral/scripts/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/models/mistral/scripts/README.md -------------------------------------------------------------------------------- /tests/torchtune/models/mistral/scripts/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/models/mistral/scripts/__init__.py -------------------------------------------------------------------------------- /tests/torchtune/models/mistral/scripts/compare_mistral.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/models/mistral/scripts/compare_mistral.py -------------------------------------------------------------------------------- /tests/torchtune/models/mistral/scripts/mistral_reference.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/models/mistral/scripts/mistral_reference.py -------------------------------------------------------------------------------- /tests/torchtune/models/mistral/test_mistral.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/models/mistral/test_mistral.py -------------------------------------------------------------------------------- /tests/torchtune/models/mistral/test_mistral_classifier.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/models/mistral/test_mistral_classifier.py -------------------------------------------------------------------------------- /tests/torchtune/models/mistral/test_mistral_tokenizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/models/mistral/test_mistral_tokenizer.py -------------------------------------------------------------------------------- /tests/torchtune/models/phi3/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/models/phi3/__init__.py -------------------------------------------------------------------------------- /tests/torchtune/models/phi3/test_lora_phi3.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/models/phi3/test_lora_phi3.py -------------------------------------------------------------------------------- /tests/torchtune/models/phi3/test_phi3.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/models/phi3/test_phi3.py -------------------------------------------------------------------------------- /tests/torchtune/models/phi3/test_phi3_position_embeddings.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/models/phi3/test_phi3_position_embeddings.py -------------------------------------------------------------------------------- /tests/torchtune/models/phi3/test_phi3_tokenizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/models/phi3/test_phi3_tokenizer.py -------------------------------------------------------------------------------- /tests/torchtune/models/phi4/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/models/phi4/__init__.py -------------------------------------------------------------------------------- /tests/torchtune/models/phi4/test_phi4_tokenizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/models/phi4/test_phi4_tokenizer.py -------------------------------------------------------------------------------- /tests/torchtune/models/qwen2/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/models/qwen2/__init__.py -------------------------------------------------------------------------------- /tests/torchtune/models/qwen2/test_lora_qwen2.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/models/qwen2/test_lora_qwen2.py -------------------------------------------------------------------------------- /tests/torchtune/models/qwen2/test_qwen2.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/models/qwen2/test_qwen2.py -------------------------------------------------------------------------------- /tests/torchtune/models/qwen2/test_qwen2_tokenizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/models/qwen2/test_qwen2_tokenizer.py -------------------------------------------------------------------------------- /tests/torchtune/models/qwen2_5/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/models/qwen2_5/__init__.py -------------------------------------------------------------------------------- /tests/torchtune/models/qwen2_5/test_tokenizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/models/qwen2_5/test_tokenizer.py -------------------------------------------------------------------------------- /tests/torchtune/models/qwen3/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/models/qwen3/__init__.py -------------------------------------------------------------------------------- /tests/torchtune/models/qwen3/test_tokenizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/models/qwen3/test_tokenizer.py -------------------------------------------------------------------------------- /tests/torchtune/models/t5/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/models/t5/__init__.py -------------------------------------------------------------------------------- /tests/torchtune/models/t5/test_t5_encoder.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/models/t5/test_t5_encoder.py -------------------------------------------------------------------------------- /tests/torchtune/models/t5/test_t5_tokenizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/models/t5/test_t5_tokenizer.py -------------------------------------------------------------------------------- /tests/torchtune/modules/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/modules/__init__.py -------------------------------------------------------------------------------- /tests/torchtune/modules/_export/test_attention.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/modules/_export/test_attention.py -------------------------------------------------------------------------------- /tests/torchtune/modules/loss/test_ce_chunked_output_loss.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/modules/loss/test_ce_chunked_output_loss.py -------------------------------------------------------------------------------- /tests/torchtune/modules/loss/test_cross_entropy_loss.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/modules/loss/test_cross_entropy_loss.py -------------------------------------------------------------------------------- /tests/torchtune/modules/loss/test_kd_losses.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/modules/loss/test_kd_losses.py -------------------------------------------------------------------------------- /tests/torchtune/modules/low_precision/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/modules/low_precision/__init__.py -------------------------------------------------------------------------------- /tests/torchtune/modules/low_precision/test_nf4_linear.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/modules/low_precision/test_nf4_linear.py -------------------------------------------------------------------------------- /tests/torchtune/modules/model_fusion/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/modules/model_fusion/__init__.py -------------------------------------------------------------------------------- /tests/torchtune/modules/model_fusion/test_deep_fusion.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/modules/model_fusion/test_deep_fusion.py -------------------------------------------------------------------------------- /tests/torchtune/modules/model_fusion/test_early_fusion.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/modules/model_fusion/test_early_fusion.py -------------------------------------------------------------------------------- /tests/torchtune/modules/model_fusion/test_fusion_layers.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/modules/model_fusion/test_fusion_layers.py -------------------------------------------------------------------------------- /tests/torchtune/modules/model_fusion/test_fusion_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/modules/model_fusion/test_fusion_utils.py -------------------------------------------------------------------------------- /tests/torchtune/modules/moe/test_experts.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/modules/moe/test_experts.py -------------------------------------------------------------------------------- /tests/torchtune/modules/moe/test_moe.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/modules/moe/test_moe.py -------------------------------------------------------------------------------- /tests/torchtune/modules/peft/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/modules/peft/__init__.py -------------------------------------------------------------------------------- /tests/torchtune/modules/peft/test_dora.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/modules/peft/test_dora.py -------------------------------------------------------------------------------- /tests/torchtune/modules/peft/test_lora.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/modules/peft/test_lora.py -------------------------------------------------------------------------------- /tests/torchtune/modules/peft/test_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/modules/peft/test_utils.py -------------------------------------------------------------------------------- /tests/torchtune/modules/test_attention.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/modules/test_attention.py -------------------------------------------------------------------------------- /tests/torchtune/modules/test_attention_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/modules/test_attention_utils.py -------------------------------------------------------------------------------- /tests/torchtune/modules/test_classifier.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/modules/test_classifier.py -------------------------------------------------------------------------------- /tests/torchtune/modules/test_common_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/modules/test_common_utils.py -------------------------------------------------------------------------------- /tests/torchtune/modules/test_cosine_with_warmup.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/modules/test_cosine_with_warmup.py -------------------------------------------------------------------------------- /tests/torchtune/modules/test_early_exit_loss.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/modules/test_early_exit_loss.py -------------------------------------------------------------------------------- /tests/torchtune/modules/test_embedding_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/modules/test_embedding_utils.py -------------------------------------------------------------------------------- /tests/torchtune/modules/test_feed_forward.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/modules/test_feed_forward.py -------------------------------------------------------------------------------- /tests/torchtune/modules/test_kv_cache.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/modules/test_kv_cache.py -------------------------------------------------------------------------------- /tests/torchtune/modules/test_layer_dropout.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/modules/test_layer_dropout.py -------------------------------------------------------------------------------- /tests/torchtune/modules/test_layernorm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/modules/test_layernorm.py -------------------------------------------------------------------------------- /tests/torchtune/modules/test_optim.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/modules/test_optim.py -------------------------------------------------------------------------------- /tests/torchtune/modules/test_position_embeddings.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/modules/test_position_embeddings.py -------------------------------------------------------------------------------- /tests/torchtune/modules/test_rms_norm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/modules/test_rms_norm.py -------------------------------------------------------------------------------- /tests/torchtune/modules/test_transformer_decoder.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/modules/test_transformer_decoder.py -------------------------------------------------------------------------------- /tests/torchtune/modules/test_vision_transformer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/modules/test_vision_transformer.py -------------------------------------------------------------------------------- /tests/torchtune/modules/test_vq_embeddings.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/modules/test_vq_embeddings.py -------------------------------------------------------------------------------- /tests/torchtune/modules/tokenizers/test_gpt2.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/modules/tokenizers/test_gpt2.py -------------------------------------------------------------------------------- /tests/torchtune/modules/transforms/test_pad_dim_to_size.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/modules/transforms/test_pad_dim_to_size.py -------------------------------------------------------------------------------- /tests/torchtune/modules/transforms/test_resize_with_pad.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/modules/transforms/test_resize_with_pad.py -------------------------------------------------------------------------------- /tests/torchtune/modules/transforms/test_tile_crop.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/modules/transforms/test_tile_crop.py -------------------------------------------------------------------------------- /tests/torchtune/modules/transforms/test_transforms.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/modules/transforms/test_transforms.py -------------------------------------------------------------------------------- /tests/torchtune/modules/transforms/tokenizers/test_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/modules/transforms/tokenizers/test_utils.py -------------------------------------------------------------------------------- /tests/torchtune/rl/test_parameter_server.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/rl/test_parameter_server.py -------------------------------------------------------------------------------- /tests/torchtune/rlhf/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/rlhf/__init__.py -------------------------------------------------------------------------------- /tests/torchtune/rlhf/loss/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/rlhf/loss/__init__.py -------------------------------------------------------------------------------- /tests/torchtune/rlhf/loss/test_dpo_loss.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/rlhf/loss/test_dpo_loss.py -------------------------------------------------------------------------------- /tests/torchtune/rlhf/loss/test_ppo_loss.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/rlhf/loss/test_ppo_loss.py -------------------------------------------------------------------------------- /tests/torchtune/rlhf/test_rewards.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/rlhf/test_rewards.py -------------------------------------------------------------------------------- /tests/torchtune/rlhf/test_sequence_processing.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/rlhf/test_sequence_processing.py -------------------------------------------------------------------------------- /tests/torchtune/training/checkpointing/test_checkpointer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/training/checkpointing/test_checkpointer.py -------------------------------------------------------------------------------- /tests/torchtune/training/test_activation_offloading.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/training/test_activation_offloading.py -------------------------------------------------------------------------------- /tests/torchtune/training/test_distributed.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/training/test_distributed.py -------------------------------------------------------------------------------- /tests/torchtune/training/test_memory.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/training/test_memory.py -------------------------------------------------------------------------------- /tests/torchtune/training/test_metric_logging.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/training/test_metric_logging.py -------------------------------------------------------------------------------- /tests/torchtune/training/test_model_util.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/training/test_model_util.py -------------------------------------------------------------------------------- /tests/torchtune/training/test_pooling.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/training/test_pooling.py -------------------------------------------------------------------------------- /tests/torchtune/training/test_precision.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/training/test_precision.py -------------------------------------------------------------------------------- /tests/torchtune/training/test_profiler.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/training/test_profiler.py -------------------------------------------------------------------------------- /tests/torchtune/training/test_quantization.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/training/test_quantization.py -------------------------------------------------------------------------------- /tests/torchtune/training/test_seed.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/training/test_seed.py -------------------------------------------------------------------------------- /tests/torchtune/utils/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/utils/__init__.py -------------------------------------------------------------------------------- /tests/torchtune/utils/test_device.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/utils/test_device.py -------------------------------------------------------------------------------- /tests/torchtune/utils/test_logging.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/tests/torchtune/utils/test_logging.py -------------------------------------------------------------------------------- /torchtune/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/__init__.py -------------------------------------------------------------------------------- /torchtune/_cli/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/_cli/__init__.py -------------------------------------------------------------------------------- /torchtune/_cli/cat.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/_cli/cat.py -------------------------------------------------------------------------------- /torchtune/_cli/cp.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/_cli/cp.py -------------------------------------------------------------------------------- /torchtune/_cli/download.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/_cli/download.py -------------------------------------------------------------------------------- /torchtune/_cli/ls.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/_cli/ls.py -------------------------------------------------------------------------------- /torchtune/_cli/run.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/_cli/run.py -------------------------------------------------------------------------------- /torchtune/_cli/subcommand.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/_cli/subcommand.py -------------------------------------------------------------------------------- /torchtune/_cli/tune.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/_cli/tune.py -------------------------------------------------------------------------------- /torchtune/_cli/validate.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/_cli/validate.py -------------------------------------------------------------------------------- /torchtune/_recipe_registry.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/_recipe_registry.py -------------------------------------------------------------------------------- /torchtune/config/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/config/__init__.py -------------------------------------------------------------------------------- /torchtune/config/_errors.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/config/_errors.py -------------------------------------------------------------------------------- /torchtune/config/_instantiate.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/config/_instantiate.py -------------------------------------------------------------------------------- /torchtune/config/_parse.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/config/_parse.py -------------------------------------------------------------------------------- /torchtune/config/_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/config/_utils.py -------------------------------------------------------------------------------- /torchtune/config/_validate.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/config/_validate.py -------------------------------------------------------------------------------- /torchtune/data/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/data/__init__.py -------------------------------------------------------------------------------- /torchtune/data/_collate.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/data/_collate.py -------------------------------------------------------------------------------- /torchtune/data/_common.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/data/_common.py -------------------------------------------------------------------------------- /torchtune/data/_messages.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/data/_messages.py -------------------------------------------------------------------------------- /torchtune/data/_prompt_templates.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/data/_prompt_templates.py -------------------------------------------------------------------------------- /torchtune/data/_torchdata.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/data/_torchdata.py -------------------------------------------------------------------------------- /torchtune/data/_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/data/_utils.py -------------------------------------------------------------------------------- /torchtune/datasets/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/datasets/__init__.py -------------------------------------------------------------------------------- /torchtune/datasets/_alpaca.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/datasets/_alpaca.py -------------------------------------------------------------------------------- /torchtune/datasets/_chat.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/datasets/_chat.py -------------------------------------------------------------------------------- /torchtune/datasets/_cnn_dailymail.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/datasets/_cnn_dailymail.py -------------------------------------------------------------------------------- /torchtune/datasets/_concat.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/datasets/_concat.py -------------------------------------------------------------------------------- /torchtune/datasets/_grammar.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/datasets/_grammar.py -------------------------------------------------------------------------------- /torchtune/datasets/_hh_rlhf_helpful.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/datasets/_hh_rlhf_helpful.py -------------------------------------------------------------------------------- /torchtune/datasets/_instruct.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/datasets/_instruct.py -------------------------------------------------------------------------------- /torchtune/datasets/_packed.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/datasets/_packed.py -------------------------------------------------------------------------------- /torchtune/datasets/_preference.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/datasets/_preference.py -------------------------------------------------------------------------------- /torchtune/datasets/_samsum.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/datasets/_samsum.py -------------------------------------------------------------------------------- /torchtune/datasets/_sft.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/datasets/_sft.py -------------------------------------------------------------------------------- /torchtune/datasets/_slimorca.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/datasets/_slimorca.py -------------------------------------------------------------------------------- /torchtune/datasets/_stack_exchange_paired.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/datasets/_stack_exchange_paired.py -------------------------------------------------------------------------------- /torchtune/datasets/_text_completion.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/datasets/_text_completion.py -------------------------------------------------------------------------------- /torchtune/datasets/_wikitext.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/datasets/_wikitext.py -------------------------------------------------------------------------------- /torchtune/datasets/multimodal/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/datasets/multimodal/__init__.py -------------------------------------------------------------------------------- /torchtune/datasets/multimodal/_llava_instruct.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/datasets/multimodal/_llava_instruct.py -------------------------------------------------------------------------------- /torchtune/datasets/multimodal/_multimodal.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/datasets/multimodal/_multimodal.py -------------------------------------------------------------------------------- /torchtune/datasets/multimodal/_the_cauldron.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/datasets/multimodal/_the_cauldron.py -------------------------------------------------------------------------------- /torchtune/datasets/multimodal/_vqa.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/datasets/multimodal/_vqa.py -------------------------------------------------------------------------------- /torchtune/dev/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/dev/README.md -------------------------------------------------------------------------------- /torchtune/dev/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/dev/__init__.py -------------------------------------------------------------------------------- /torchtune/dev/grpo/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/dev/grpo/__init__.py -------------------------------------------------------------------------------- /torchtune/dev/grpo/data.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/dev/grpo/data.py -------------------------------------------------------------------------------- /torchtune/dev/grpo/generation.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/dev/grpo/generation.py -------------------------------------------------------------------------------- /torchtune/dev/grpo/gsm8k.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/dev/grpo/gsm8k.py -------------------------------------------------------------------------------- /torchtune/dev/grpo/loss.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/dev/grpo/loss.py -------------------------------------------------------------------------------- /torchtune/dev/grpo/rewards.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/dev/grpo/rewards.py -------------------------------------------------------------------------------- /torchtune/dev/grpo/types.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/dev/grpo/types.py -------------------------------------------------------------------------------- /torchtune/dev/rl/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/dev/rl/__init__.py -------------------------------------------------------------------------------- /torchtune/dev/rl/data.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/dev/rl/data.py -------------------------------------------------------------------------------- /torchtune/dev/rl/datatypes/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/dev/rl/datatypes/__init__.py -------------------------------------------------------------------------------- /torchtune/dev/rl/datatypes/request_output.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/dev/rl/datatypes/request_output.py -------------------------------------------------------------------------------- /torchtune/dev/rl/datatypes/trajectory.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/dev/rl/datatypes/trajectory.py -------------------------------------------------------------------------------- /torchtune/dev/rl/datatypes/vllm_completion_output.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/dev/rl/datatypes/vllm_completion_output.py -------------------------------------------------------------------------------- /torchtune/dev/rl/generation.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/dev/rl/generation.py -------------------------------------------------------------------------------- /torchtune/dev/rl/gsm8k.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/dev/rl/gsm8k.py -------------------------------------------------------------------------------- /torchtune/dev/rl/linear_grpo_loss.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/dev/rl/linear_grpo_loss.py -------------------------------------------------------------------------------- /torchtune/dev/rl/rewards.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/dev/rl/rewards.py -------------------------------------------------------------------------------- /torchtune/dev/rl/types.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/dev/rl/types.py -------------------------------------------------------------------------------- /torchtune/dev/rl/utils/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/dev/rl/utils/__init__.py -------------------------------------------------------------------------------- /torchtune/dev/rl/utils/dist.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/dev/rl/utils/dist.py -------------------------------------------------------------------------------- /torchtune/dev/rl/workers/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/dev/rl/workers/__init__.py -------------------------------------------------------------------------------- /torchtune/dev/rl/workers/datacollectors/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/dev/rl/workers/datacollectors/__init__.py -------------------------------------------------------------------------------- /torchtune/dev/rl/workers/datacollectors/sync.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/dev/rl/workers/datacollectors/sync.py -------------------------------------------------------------------------------- /torchtune/dev/rl/workers/metric_logger.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/dev/rl/workers/metric_logger.py -------------------------------------------------------------------------------- /torchtune/dev/rl/workers/parameter_servers/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/dev/rl/workers/parameter_servers/__init__.py -------------------------------------------------------------------------------- /torchtune/dev/rl/workers/parameter_servers/vllm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/dev/rl/workers/parameter_servers/vllm.py -------------------------------------------------------------------------------- /torchtune/dev/rl/workers/postprocessing.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/dev/rl/workers/postprocessing.py -------------------------------------------------------------------------------- /torchtune/dev/rl/workers/trainers/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/dev/rl/workers/trainers/__init__.py -------------------------------------------------------------------------------- /torchtune/dev/rl/workers/trainers/training.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/dev/rl/workers/trainers/training.py -------------------------------------------------------------------------------- /torchtune/dev/rl/workers/weight_updaters/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/dev/rl/workers/weight_updaters/__init__.py -------------------------------------------------------------------------------- /torchtune/dev/rl/workers/weight_updaters/weight_updater.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/dev/rl/workers/weight_updaters/weight_updater.py -------------------------------------------------------------------------------- /torchtune/generation/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/generation/__init__.py -------------------------------------------------------------------------------- /torchtune/generation/_generation.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/generation/_generation.py -------------------------------------------------------------------------------- /torchtune/models/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/__init__.py -------------------------------------------------------------------------------- /torchtune/models/clip/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/clip/__init__.py -------------------------------------------------------------------------------- /torchtune/models/clip/_component_builders.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/clip/_component_builders.py -------------------------------------------------------------------------------- /torchtune/models/clip/_convert_weights.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/clip/_convert_weights.py -------------------------------------------------------------------------------- /torchtune/models/clip/_model_builders.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/clip/_model_builders.py -------------------------------------------------------------------------------- /torchtune/models/clip/_position_embeddings.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/clip/_position_embeddings.py -------------------------------------------------------------------------------- /torchtune/models/clip/_text_encoder.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/clip/_text_encoder.py -------------------------------------------------------------------------------- /torchtune/models/clip/_tokenizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/clip/_tokenizer.py -------------------------------------------------------------------------------- /torchtune/models/clip/_transform.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/clip/_transform.py -------------------------------------------------------------------------------- /torchtune/models/clip/inference/_transform.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/clip/inference/_transform.py -------------------------------------------------------------------------------- /torchtune/models/convert_weights.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/convert_weights.py -------------------------------------------------------------------------------- /torchtune/models/flux/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/flux/__init__.py -------------------------------------------------------------------------------- /torchtune/models/flux/_autoencoder.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/flux/_autoencoder.py -------------------------------------------------------------------------------- /torchtune/models/flux/_convert_weights.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/flux/_convert_weights.py -------------------------------------------------------------------------------- /torchtune/models/flux/_model_builders.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/flux/_model_builders.py -------------------------------------------------------------------------------- /torchtune/models/gemma/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/gemma/__init__.py -------------------------------------------------------------------------------- /torchtune/models/gemma/_component_builders.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/gemma/_component_builders.py -------------------------------------------------------------------------------- /torchtune/models/gemma/_model_builders.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/gemma/_model_builders.py -------------------------------------------------------------------------------- /torchtune/models/gemma/_tokenizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/gemma/_tokenizer.py -------------------------------------------------------------------------------- /torchtune/models/gemma/gemma_norm_embedding.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/gemma/gemma_norm_embedding.py -------------------------------------------------------------------------------- /torchtune/models/gemma/rms_norm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/gemma/rms_norm.py -------------------------------------------------------------------------------- /torchtune/models/gemma2/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/gemma2/__init__.py -------------------------------------------------------------------------------- /torchtune/models/gemma2/_attention.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/gemma2/_attention.py -------------------------------------------------------------------------------- /torchtune/models/gemma2/_component_builders.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/gemma2/_component_builders.py -------------------------------------------------------------------------------- /torchtune/models/gemma2/_convert_weights.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/gemma2/_convert_weights.py -------------------------------------------------------------------------------- /torchtune/models/gemma2/_model_builders.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/gemma2/_model_builders.py -------------------------------------------------------------------------------- /torchtune/models/llama2/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/llama2/__init__.py -------------------------------------------------------------------------------- /torchtune/models/llama2/_component_builders.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/llama2/_component_builders.py -------------------------------------------------------------------------------- /torchtune/models/llama2/_model_builders.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/llama2/_model_builders.py -------------------------------------------------------------------------------- /torchtune/models/llama2/_model_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/llama2/_model_utils.py -------------------------------------------------------------------------------- /torchtune/models/llama2/_prompt_template.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/llama2/_prompt_template.py -------------------------------------------------------------------------------- /torchtune/models/llama2/_tokenizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/llama2/_tokenizer.py -------------------------------------------------------------------------------- /torchtune/models/llama3/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/llama3/__init__.py -------------------------------------------------------------------------------- /torchtune/models/llama3/_component_builders.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/llama3/_component_builders.py -------------------------------------------------------------------------------- /torchtune/models/llama3/_model_builders.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/llama3/_model_builders.py -------------------------------------------------------------------------------- /torchtune/models/llama3/_model_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/llama3/_model_utils.py -------------------------------------------------------------------------------- /torchtune/models/llama3/_parallelism.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/llama3/_parallelism.py -------------------------------------------------------------------------------- /torchtune/models/llama3/_tokenizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/llama3/_tokenizer.py -------------------------------------------------------------------------------- /torchtune/models/llama3_1/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/llama3_1/__init__.py -------------------------------------------------------------------------------- /torchtune/models/llama3_1/_component_builders.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/llama3_1/_component_builders.py -------------------------------------------------------------------------------- /torchtune/models/llama3_1/_model_builders.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/llama3_1/_model_builders.py -------------------------------------------------------------------------------- /torchtune/models/llama3_1/_position_embeddings.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/llama3_1/_position_embeddings.py -------------------------------------------------------------------------------- /torchtune/models/llama3_2/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/llama3_2/__init__.py -------------------------------------------------------------------------------- /torchtune/models/llama3_2/_component_builders.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/llama3_2/_component_builders.py -------------------------------------------------------------------------------- /torchtune/models/llama3_2/_model_builders.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/llama3_2/_model_builders.py -------------------------------------------------------------------------------- /torchtune/models/llama3_2_vision/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/llama3_2_vision/__init__.py -------------------------------------------------------------------------------- /torchtune/models/llama3_2_vision/_component_builders.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/llama3_2_vision/_component_builders.py -------------------------------------------------------------------------------- /torchtune/models/llama3_2_vision/_convert_weights.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/llama3_2_vision/_convert_weights.py -------------------------------------------------------------------------------- /torchtune/models/llama3_2_vision/_encoder.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/llama3_2_vision/_encoder.py -------------------------------------------------------------------------------- /torchtune/models/llama3_2_vision/_model_builders.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/llama3_2_vision/_model_builders.py -------------------------------------------------------------------------------- /torchtune/models/llama3_2_vision/_transform.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/llama3_2_vision/_transform.py -------------------------------------------------------------------------------- /torchtune/models/llama3_3/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/llama3_3/__init__.py -------------------------------------------------------------------------------- /torchtune/models/llama3_3/_model_builders.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/llama3_3/_model_builders.py -------------------------------------------------------------------------------- /torchtune/models/llama4/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/llama4/__init__.py -------------------------------------------------------------------------------- /torchtune/models/llama4/_chunked_attention.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/llama4/_chunked_attention.py -------------------------------------------------------------------------------- /torchtune/models/llama4/_component_builders.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/llama4/_component_builders.py -------------------------------------------------------------------------------- /torchtune/models/llama4/_convert_weights.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/llama4/_convert_weights.py -------------------------------------------------------------------------------- /torchtune/models/llama4/_encoder.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/llama4/_encoder.py -------------------------------------------------------------------------------- /torchtune/models/llama4/_model_builders.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/llama4/_model_builders.py -------------------------------------------------------------------------------- /torchtune/models/llama4/_parallelism.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/llama4/_parallelism.py -------------------------------------------------------------------------------- /torchtune/models/llama4/_position_embeddings.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/llama4/_position_embeddings.py -------------------------------------------------------------------------------- /torchtune/models/llama4/_tokenizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/llama4/_tokenizer.py -------------------------------------------------------------------------------- /torchtune/models/llama4/_transform.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/llama4/_transform.py -------------------------------------------------------------------------------- /torchtune/models/mistral/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/mistral/__init__.py -------------------------------------------------------------------------------- /torchtune/models/mistral/_component_builders.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/mistral/_component_builders.py -------------------------------------------------------------------------------- /torchtune/models/mistral/_model_builders.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/mistral/_model_builders.py -------------------------------------------------------------------------------- /torchtune/models/mistral/_prompt_template.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/mistral/_prompt_template.py -------------------------------------------------------------------------------- /torchtune/models/mistral/_tokenizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/mistral/_tokenizer.py -------------------------------------------------------------------------------- /torchtune/models/phi3/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/phi3/__init__.py -------------------------------------------------------------------------------- /torchtune/models/phi3/_component_builders.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/phi3/_component_builders.py -------------------------------------------------------------------------------- /torchtune/models/phi3/_convert_weights.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/phi3/_convert_weights.py -------------------------------------------------------------------------------- /torchtune/models/phi3/_model_builders.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/phi3/_model_builders.py -------------------------------------------------------------------------------- /torchtune/models/phi3/_position_embeddings.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/phi3/_position_embeddings.py -------------------------------------------------------------------------------- /torchtune/models/phi3/_tokenizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/phi3/_tokenizer.py -------------------------------------------------------------------------------- /torchtune/models/phi4/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/phi4/__init__.py -------------------------------------------------------------------------------- /torchtune/models/phi4/_model_builders.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/phi4/_model_builders.py -------------------------------------------------------------------------------- /torchtune/models/phi4/_tokenizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/phi4/_tokenizer.py -------------------------------------------------------------------------------- /torchtune/models/qwen2/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/qwen2/__init__.py -------------------------------------------------------------------------------- /torchtune/models/qwen2/_component_builders.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/qwen2/_component_builders.py -------------------------------------------------------------------------------- /torchtune/models/qwen2/_convert_weights.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/qwen2/_convert_weights.py -------------------------------------------------------------------------------- /torchtune/models/qwen2/_model_builders.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/qwen2/_model_builders.py -------------------------------------------------------------------------------- /torchtune/models/qwen2/_positional_embeddings.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/qwen2/_positional_embeddings.py -------------------------------------------------------------------------------- /torchtune/models/qwen2/_tokenizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/qwen2/_tokenizer.py -------------------------------------------------------------------------------- /torchtune/models/qwen2_5/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/qwen2_5/__init__.py -------------------------------------------------------------------------------- /torchtune/models/qwen2_5/_model_builders.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/qwen2_5/_model_builders.py -------------------------------------------------------------------------------- /torchtune/models/qwen2_5/_tokenizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/qwen2_5/_tokenizer.py -------------------------------------------------------------------------------- /torchtune/models/qwen3/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/qwen3/__init__.py -------------------------------------------------------------------------------- /torchtune/models/qwen3/_attention.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/qwen3/_attention.py -------------------------------------------------------------------------------- /torchtune/models/qwen3/_component_builders.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/qwen3/_component_builders.py -------------------------------------------------------------------------------- /torchtune/models/qwen3/_convert_weights.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/qwen3/_convert_weights.py -------------------------------------------------------------------------------- /torchtune/models/qwen3/_model_builders.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/qwen3/_model_builders.py -------------------------------------------------------------------------------- /torchtune/models/qwen3/_tokenizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/qwen3/_tokenizer.py -------------------------------------------------------------------------------- /torchtune/models/smol/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/smol/__init__.py -------------------------------------------------------------------------------- /torchtune/models/smol/_component_builders.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/smol/_component_builders.py -------------------------------------------------------------------------------- /torchtune/models/smol/_model_builders.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/smol/_model_builders.py -------------------------------------------------------------------------------- /torchtune/models/t5/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/t5/__init__.py -------------------------------------------------------------------------------- /torchtune/models/t5/_component_builders.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/t5/_component_builders.py -------------------------------------------------------------------------------- /torchtune/models/t5/_convert_weights.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/t5/_convert_weights.py -------------------------------------------------------------------------------- /torchtune/models/t5/_encoder.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/t5/_encoder.py -------------------------------------------------------------------------------- /torchtune/models/t5/_model_builders.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/t5/_model_builders.py -------------------------------------------------------------------------------- /torchtune/models/t5/_tokenizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/models/t5/_tokenizer.py -------------------------------------------------------------------------------- /torchtune/modules/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/modules/__init__.py -------------------------------------------------------------------------------- /torchtune/modules/_export/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/modules/_export/README.md -------------------------------------------------------------------------------- /torchtune/modules/_export/_position_embeddings.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/modules/_export/_position_embeddings.py -------------------------------------------------------------------------------- /torchtune/modules/_export/attention.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/modules/_export/attention.py -------------------------------------------------------------------------------- /torchtune/modules/_export/install_requirements.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/modules/_export/install_requirements.sh -------------------------------------------------------------------------------- /torchtune/modules/_export/kv_cache.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/modules/_export/kv_cache.py -------------------------------------------------------------------------------- /torchtune/modules/attention.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/modules/attention.py -------------------------------------------------------------------------------- /torchtune/modules/attention_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/modules/attention_utils.py -------------------------------------------------------------------------------- /torchtune/modules/classifier.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/modules/classifier.py -------------------------------------------------------------------------------- /torchtune/modules/common_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/modules/common_utils.py -------------------------------------------------------------------------------- /torchtune/modules/early_exit_loss.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/modules/early_exit_loss.py -------------------------------------------------------------------------------- /torchtune/modules/embedding_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/modules/embedding_utils.py -------------------------------------------------------------------------------- /torchtune/modules/feed_forward.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/modules/feed_forward.py -------------------------------------------------------------------------------- /torchtune/modules/kv_cache.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/modules/kv_cache.py -------------------------------------------------------------------------------- /torchtune/modules/layer_dropout.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/modules/layer_dropout.py -------------------------------------------------------------------------------- /torchtune/modules/layer_norm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/modules/layer_norm.py -------------------------------------------------------------------------------- /torchtune/modules/loss/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/modules/loss/__init__.py -------------------------------------------------------------------------------- /torchtune/modules/loss/ce_chunked_output_loss.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/modules/loss/ce_chunked_output_loss.py -------------------------------------------------------------------------------- /torchtune/modules/loss/cross_entropy_loss.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/modules/loss/cross_entropy_loss.py -------------------------------------------------------------------------------- /torchtune/modules/loss/kd_losses.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/modules/loss/kd_losses.py -------------------------------------------------------------------------------- /torchtune/modules/loss/loss_types.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/modules/loss/loss_types.py -------------------------------------------------------------------------------- /torchtune/modules/low_precision/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/modules/low_precision/__init__.py -------------------------------------------------------------------------------- /torchtune/modules/low_precision/nf4_linear.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/modules/low_precision/nf4_linear.py -------------------------------------------------------------------------------- /torchtune/modules/model_fusion/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/modules/model_fusion/__init__.py -------------------------------------------------------------------------------- /torchtune/modules/model_fusion/_deep_fusion.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/modules/model_fusion/_deep_fusion.py -------------------------------------------------------------------------------- /torchtune/modules/model_fusion/_early_fusion.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/modules/model_fusion/_early_fusion.py -------------------------------------------------------------------------------- /torchtune/modules/model_fusion/_fusion_layers.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/modules/model_fusion/_fusion_layers.py -------------------------------------------------------------------------------- /torchtune/modules/model_fusion/_fusion_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/modules/model_fusion/_fusion_utils.py -------------------------------------------------------------------------------- /torchtune/modules/moe/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/modules/moe/__init__.py -------------------------------------------------------------------------------- /torchtune/modules/moe/_parallelism.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/modules/moe/_parallelism.py -------------------------------------------------------------------------------- /torchtune/modules/moe/experts.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/modules/moe/experts.py -------------------------------------------------------------------------------- /torchtune/modules/moe/indices.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/modules/moe/indices.py -------------------------------------------------------------------------------- /torchtune/modules/moe/moe.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/modules/moe/moe.py -------------------------------------------------------------------------------- /torchtune/modules/moe/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/modules/moe/utils.py -------------------------------------------------------------------------------- /torchtune/modules/optim.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/modules/optim.py -------------------------------------------------------------------------------- /torchtune/modules/peft/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/modules/peft/__init__.py -------------------------------------------------------------------------------- /torchtune/modules/peft/_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/modules/peft/_utils.py -------------------------------------------------------------------------------- /torchtune/modules/peft/dora.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/modules/peft/dora.py -------------------------------------------------------------------------------- /torchtune/modules/peft/lora.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/modules/peft/lora.py -------------------------------------------------------------------------------- /torchtune/modules/position_embeddings.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/modules/position_embeddings.py -------------------------------------------------------------------------------- /torchtune/modules/rms_norm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/modules/rms_norm.py -------------------------------------------------------------------------------- /torchtune/modules/tanh_gate.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/modules/tanh_gate.py -------------------------------------------------------------------------------- /torchtune/modules/tied_linear.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/modules/tied_linear.py -------------------------------------------------------------------------------- /torchtune/modules/tokenizers/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/modules/tokenizers/__init__.py -------------------------------------------------------------------------------- /torchtune/modules/transformer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/modules/transformer.py -------------------------------------------------------------------------------- /torchtune/modules/transforms/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/modules/transforms/__init__.py -------------------------------------------------------------------------------- /torchtune/modules/transforms/_transforms.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/modules/transforms/_transforms.py -------------------------------------------------------------------------------- /torchtune/modules/transforms/tokenizers/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/modules/transforms/tokenizers/__init__.py -------------------------------------------------------------------------------- /torchtune/modules/transforms/tokenizers/_gpt2.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/modules/transforms/tokenizers/_gpt2.py -------------------------------------------------------------------------------- /torchtune/modules/transforms/tokenizers/_hf_tokenizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/modules/transforms/tokenizers/_hf_tokenizer.py -------------------------------------------------------------------------------- /torchtune/modules/transforms/tokenizers/_sentencepiece.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/modules/transforms/tokenizers/_sentencepiece.py -------------------------------------------------------------------------------- /torchtune/modules/transforms/tokenizers/_tiktoken.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/modules/transforms/tokenizers/_tiktoken.py -------------------------------------------------------------------------------- /torchtune/modules/transforms/tokenizers/_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/modules/transforms/tokenizers/_utils.py -------------------------------------------------------------------------------- /torchtune/modules/transforms/vision_utils/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/modules/transforms/vision_utils/__init__.py -------------------------------------------------------------------------------- /torchtune/modules/transforms/vision_utils/pad_dim_to_size.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/modules/transforms/vision_utils/pad_dim_to_size.py -------------------------------------------------------------------------------- /torchtune/modules/transforms/vision_utils/resize_with_pad.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/modules/transforms/vision_utils/resize_with_pad.py -------------------------------------------------------------------------------- /torchtune/modules/transforms/vision_utils/tile_crop.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/modules/transforms/vision_utils/tile_crop.py -------------------------------------------------------------------------------- /torchtune/modules/vision_transformer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/modules/vision_transformer.py -------------------------------------------------------------------------------- /torchtune/modules/vq_embeddings.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/modules/vq_embeddings.py -------------------------------------------------------------------------------- /torchtune/recipe_interfaces.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/recipe_interfaces.py -------------------------------------------------------------------------------- /torchtune/rlhf/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/rlhf/__init__.py -------------------------------------------------------------------------------- /torchtune/rlhf/_types.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/rlhf/_types.py -------------------------------------------------------------------------------- /torchtune/rlhf/loss/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/rlhf/loss/__init__.py -------------------------------------------------------------------------------- /torchtune/rlhf/loss/dpo.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/rlhf/loss/dpo.py -------------------------------------------------------------------------------- /torchtune/rlhf/loss/ppo.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/rlhf/loss/ppo.py -------------------------------------------------------------------------------- /torchtune/rlhf/rewards.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/rlhf/rewards.py -------------------------------------------------------------------------------- /torchtune/rlhf/sequence_processing.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/rlhf/sequence_processing.py -------------------------------------------------------------------------------- /torchtune/rlhf/utils/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/rlhf/utils/__init__.py -------------------------------------------------------------------------------- /torchtune/rlhf/utils/_convert_weights.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/rlhf/utils/_convert_weights.py -------------------------------------------------------------------------------- /torchtune/training/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/training/__init__.py -------------------------------------------------------------------------------- /torchtune/training/_activation_offloading.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/training/_activation_offloading.py -------------------------------------------------------------------------------- /torchtune/training/_compile.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/training/_compile.py -------------------------------------------------------------------------------- /torchtune/training/_distributed.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/training/_distributed.py -------------------------------------------------------------------------------- /torchtune/training/_grad_scaler.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/training/_grad_scaler.py -------------------------------------------------------------------------------- /torchtune/training/_model_util.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/training/_model_util.py -------------------------------------------------------------------------------- /torchtune/training/_profiler.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/training/_profiler.py -------------------------------------------------------------------------------- /torchtune/training/activations.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/training/activations.py -------------------------------------------------------------------------------- /torchtune/training/checkpointing/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/training/checkpointing/__init__.py -------------------------------------------------------------------------------- /torchtune/training/checkpointing/_checkpoint_client.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/training/checkpointing/_checkpoint_client.py -------------------------------------------------------------------------------- /torchtune/training/checkpointing/_checkpointer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/training/checkpointing/_checkpointer.py -------------------------------------------------------------------------------- /torchtune/training/checkpointing/_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/training/checkpointing/_utils.py -------------------------------------------------------------------------------- /torchtune/training/lr_schedulers.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/training/lr_schedulers.py -------------------------------------------------------------------------------- /torchtune/training/memory.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/training/memory.py -------------------------------------------------------------------------------- /torchtune/training/metric_logging.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/training/metric_logging.py -------------------------------------------------------------------------------- /torchtune/training/pooling.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/training/pooling.py -------------------------------------------------------------------------------- /torchtune/training/precision.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/training/precision.py -------------------------------------------------------------------------------- /torchtune/training/quantization.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/training/quantization.py -------------------------------------------------------------------------------- /torchtune/training/seed.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/training/seed.py -------------------------------------------------------------------------------- /torchtune/utils/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/utils/__init__.py -------------------------------------------------------------------------------- /torchtune/utils/_device.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/utils/_device.py -------------------------------------------------------------------------------- /torchtune/utils/_import_guard.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/utils/_import_guard.py -------------------------------------------------------------------------------- /torchtune/utils/_logging.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/utils/_logging.py -------------------------------------------------------------------------------- /torchtune/utils/_version.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/meta-pytorch/torchtune/HEAD/torchtune/utils/_version.py -------------------------------------------------------------------------------- /version.txt: -------------------------------------------------------------------------------- 1 | 0.7.0 2 | --------------------------------------------------------------------------------