├── .dockerignore ├── .flake8 ├── .github ├── dependabot.yml └── workflows │ ├── run_entry_tests.yaml │ ├── run_pre_commit.yaml │ ├── run_ray_tests.yaml │ └── run_tests.yaml ├── .gitignore ├── .idea └── .gitignore ├── .pre-commit-config.yaml ├── .readthedocs.yaml ├── CONTRIBUTING.md ├── LICENSE ├── README.md ├── config ├── backpack.yaml ├── backpack_nano.yaml ├── data │ ├── openwebtext_source.yaml │ ├── pile_mixture.yaml │ ├── pile_source_old.yaml │ ├── pubmed_source.yaml │ ├── redpajama_1b_source.yaml │ ├── redpajama_1t_source.yaml │ ├── rpv1_llama.yaml │ └── wikitext_source.yaml ├── doremi │ └── doremi_nano.yaml ├── gpt2_1536.yaml ├── gpt2_1536_sophiah.yaml ├── gpt2_20b.yaml ├── gpt2_7b.yaml ├── gpt2_large.yaml ├── gpt2_large_sophia_h.yaml ├── gpt2_medium.yaml ├── gpt2_micro.yaml ├── gpt2_nano.yaml ├── gpt2_nano_fp8.yaml ├── gpt2_nano_mixture.yaml ├── gpt2_nano_tb.yaml ├── gpt2_small.yaml ├── gpt2_small_fast.yaml ├── gpt2_small_fast_fp8.yaml ├── gpt2_small_fast_mix.yaml ├── gpt2_small_fast_pile.yaml ├── gpt2_small_fast_public.yaml ├── gpt2_small_fast_sophia_h.yaml ├── gpt2_small_fast_sophiah.yaml ├── gpt2_small_fast_wiki.yaml ├── gpt2_small_pile.yaml ├── gpt2_small_pile_mixture.yaml ├── gpt2_small_sophiah.yaml ├── gpt2_xl.yaml ├── llama2_3b_pretrain.yaml ├── llama2_7b.yaml ├── llama2_7b_continued.yaml ├── llama2_nano.yaml ├── llama_small_fast.yaml ├── lora │ └── mpt_biomed.yaml ├── lora_llama2.yaml ├── mistral_7b.yaml ├── optim │ ├── sophia-h_large.yaml │ ├── sophia-h_medium.yaml │ ├── sophia-h_small.yaml │ └── sophia-h_xl.yaml └── whisper_tiny_librispeech.yaml ├── docker └── nvidia │ └── Dockerfile ├── docs ├── Configuration-Guide.md ├── Fine-Tuning.md ├── Getting-Started-GPU.md ├── Getting-Started-TPU-VM.md ├── Getting-Started-Training.md ├── Hardware-Agnostic-Training.md ├── Installation.md ├── Levanter-1.0-Release.md ├── LoRA.md ├── Performance-Guide.md ├── Training-On-Your-Data.md ├── css │ ├── custom.css │ ├── friendly.css │ └── mkdocstrings.css ├── design │ ├── Data-Loader-Design.md │ └── Multiple-Data-Mixture.md ├── dev │ ├── GPU-Docker-Dev.md │ ├── Port-Models.md │ └── Trackers.md ├── faq.md ├── figures │ ├── bitwise_repro_curve.png │ ├── data_parallel_mesh.png │ ├── data_parallel_mesh_replicated.png │ ├── device_mesh_1d.png │ ├── device_mesh_1d_zero.png │ ├── device_mesh_2d.png │ ├── device_mesh_2d_batch_partitioned.png │ ├── device_mesh_2d_data_replicated.png │ ├── device_mesh_2d_data_replicated_mlp_partitioned.png │ ├── device_mesh_2d_intermediate_fully_partitioned.png │ ├── device_mesh_2d_zero.png │ ├── finetune_func_cm_full_weight.png │ ├── finetune_func_cm_lora.png │ ├── helm-gsm8k-results.png │ ├── helm-instance-example.png │ ├── lora-diagram.png │ ├── palm_mfu_table.png │ ├── resumed_curve.png │ ├── stopped_curve.png │ └── token_probabilities.mov ├── index.md ├── javascripts │ └── mathjax.js ├── requirements.txt └── tutorials │ ├── Fine-Tuning-Semantic-Parsing.md │ └── Training-On-Audio-Data.md ├── examples ├── alpaca-lora │ ├── alpaca-lora-llama2.yaml │ ├── alpaca-lora.yaml │ ├── alpaca_lora.py │ ├── code-alpaca-lora.yaml │ ├── hf_lora_inference.py │ └── peft_inference.ipynb ├── alpaca │ ├── alpaca-llama2.yaml │ ├── alpaca.py │ └── alpaca.yaml └── gsm8k-lora │ ├── gsm8k-llama2.yaml │ └── gsm8k_lora.py ├── infra ├── babysit-tpu-vm ├── babysit-tpu-vm.sh ├── helpers │ ├── gen-id.sh │ ├── parse-tpu-creation-args.sh │ ├── setup-tpu-vm-nfs.sh │ └── setup-tpu-vm.sh ├── launch.sh ├── run-slurm.sh ├── run.sh └── spin-up-vm.sh ├── mkdocs.yml ├── pyproject.toml ├── scripts ├── clean_old_checkpoints.py ├── launch_gpt2_small_fast_gpu.sh ├── launch_gpt2_small_fast_tpu.sh ├── loss_history.py ├── preproc │ └── split-pile-shards.py ├── repair_cache.py └── train_tokenizer.py ├── src └── levanter │ ├── __init__.py │ ├── callbacks.py │ ├── checkpoint.py │ ├── compat │ ├── __init__.py │ ├── hf_checkpoints.py │ └── torch_serialization.py │ ├── config.py │ ├── data │ ├── __init__.py │ ├── _preprocessor.py │ ├── audio.py │ ├── dataset.py │ ├── loader.py │ ├── mixture.py │ ├── shard_cache.py │ ├── sharded_dataset.py │ ├── text.py │ └── utils.py │ ├── distributed.py │ ├── doremi.py │ ├── eval.py │ ├── grad_accum.py │ ├── logging.py │ ├── lora.py │ ├── main │ ├── cache_dataset.py │ ├── doremi_lm.py │ ├── eval_lm.py │ ├── export_lm_to_hf.py │ ├── lora_lm.py │ ├── train_asr.py │ ├── train_lm.py │ └── viz_logprobs.py │ ├── mesh.py │ ├── models │ ├── __init__.py │ ├── asr_model.py │ ├── attention.py │ ├── backpack.py │ ├── flash_attention.py │ ├── gpt2.py │ ├── llama.py │ ├── lm_model.py │ ├── longformer.py │ ├── loss.py │ ├── mistral.py │ ├── mpt.py │ └── whisper.py │ ├── optim │ ├── __init__.py │ ├── config.py │ ├── sophia.py │ └── util.py │ ├── shapes.py │ ├── tensorstore_serialization.py │ ├── tracker │ ├── __init__.py │ ├── helpers.py │ ├── tensorboard.py │ ├── tracker.py │ ├── tracker_fns.py │ └── wandb.py │ ├── trainer.py │ ├── trainer_state.py │ ├── types.py │ ├── utils │ ├── __init__.py │ ├── background_iterable.py │ ├── cloud_utils.py │ ├── datetime_utils.py │ ├── fsspec_utils.py │ ├── hf_utils.py │ ├── jax_utils.py │ ├── py_utils.py │ ├── ray_utils.py │ ├── stat_utils.py │ └── tree_utils.py │ └── visualization.py └── tests ├── data └── hero_data.npy ├── gpt2_test.py ├── gpt2_tokenizer_config.json ├── test_attention.py ├── test_audio.py ├── test_background_iterable.py ├── test_backpack.py ├── test_checkpoint.py ├── test_config.py ├── test_data_mixture.py ├── test_datetime_utils.py ├── test_distributed.py ├── test_doremi.py ├── test_eval_lm.py ├── test_export_to_hf.py ├── test_flash_attention.py ├── test_grad_accum.py ├── test_hf_checkpoints.py ├── test_hf_gpt2_serialize.py ├── test_hf_utils.py ├── test_jax_utils.py ├── test_llama.py ├── test_logging.py ├── test_longformer.py ├── test_lora.py ├── test_mistral.py ├── test_mpt.py ├── test_py_utils.py ├── test_replicated_loader.py ├── test_shard_cache.py ├── test_sharded_dataset.py ├── test_sharded_loader.py ├── test_shuffle_dataset.py ├── test_sophia.py ├── test_tensorstore_serialization.py ├── test_text.py ├── test_tokenized_document_cache.py ├── test_torch_serialization.py ├── test_tracker.py ├── test_train_asr.py ├── test_train_lm.py ├── test_utils.py ├── test_viz_lm.py ├── test_weight_decay_mask.py ├── tiny_test_corpus.py └── whisper_test.py /.dockerignore: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/.dockerignore -------------------------------------------------------------------------------- /.flake8: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/.flake8 -------------------------------------------------------------------------------- /.github/dependabot.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/.github/dependabot.yml -------------------------------------------------------------------------------- /.github/workflows/run_entry_tests.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/.github/workflows/run_entry_tests.yaml -------------------------------------------------------------------------------- /.github/workflows/run_pre_commit.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/.github/workflows/run_pre_commit.yaml -------------------------------------------------------------------------------- /.github/workflows/run_ray_tests.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/.github/workflows/run_ray_tests.yaml -------------------------------------------------------------------------------- /.github/workflows/run_tests.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/.github/workflows/run_tests.yaml -------------------------------------------------------------------------------- /.gitignore: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/.gitignore -------------------------------------------------------------------------------- /.idea/.gitignore: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/.idea/.gitignore -------------------------------------------------------------------------------- /.pre-commit-config.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/.pre-commit-config.yaml -------------------------------------------------------------------------------- /.readthedocs.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/.readthedocs.yaml -------------------------------------------------------------------------------- /CONTRIBUTING.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/CONTRIBUTING.md -------------------------------------------------------------------------------- /LICENSE: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/LICENSE -------------------------------------------------------------------------------- /README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/README.md -------------------------------------------------------------------------------- /config/backpack.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/config/backpack.yaml -------------------------------------------------------------------------------- /config/backpack_nano.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/config/backpack_nano.yaml -------------------------------------------------------------------------------- /config/data/openwebtext_source.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/config/data/openwebtext_source.yaml -------------------------------------------------------------------------------- /config/data/pile_mixture.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/config/data/pile_mixture.yaml -------------------------------------------------------------------------------- /config/data/pile_source_old.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/config/data/pile_source_old.yaml -------------------------------------------------------------------------------- /config/data/pubmed_source.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/config/data/pubmed_source.yaml -------------------------------------------------------------------------------- /config/data/redpajama_1b_source.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/config/data/redpajama_1b_source.yaml -------------------------------------------------------------------------------- /config/data/redpajama_1t_source.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/config/data/redpajama_1t_source.yaml -------------------------------------------------------------------------------- /config/data/rpv1_llama.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/config/data/rpv1_llama.yaml -------------------------------------------------------------------------------- /config/data/wikitext_source.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/config/data/wikitext_source.yaml -------------------------------------------------------------------------------- /config/doremi/doremi_nano.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/config/doremi/doremi_nano.yaml -------------------------------------------------------------------------------- /config/gpt2_1536.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/config/gpt2_1536.yaml -------------------------------------------------------------------------------- /config/gpt2_1536_sophiah.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/config/gpt2_1536_sophiah.yaml -------------------------------------------------------------------------------- /config/gpt2_20b.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/config/gpt2_20b.yaml -------------------------------------------------------------------------------- /config/gpt2_7b.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/config/gpt2_7b.yaml -------------------------------------------------------------------------------- /config/gpt2_large.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/config/gpt2_large.yaml -------------------------------------------------------------------------------- /config/gpt2_large_sophia_h.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/config/gpt2_large_sophia_h.yaml -------------------------------------------------------------------------------- /config/gpt2_medium.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/config/gpt2_medium.yaml -------------------------------------------------------------------------------- /config/gpt2_micro.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/config/gpt2_micro.yaml -------------------------------------------------------------------------------- /config/gpt2_nano.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/config/gpt2_nano.yaml -------------------------------------------------------------------------------- /config/gpt2_nano_fp8.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/config/gpt2_nano_fp8.yaml -------------------------------------------------------------------------------- /config/gpt2_nano_mixture.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/config/gpt2_nano_mixture.yaml -------------------------------------------------------------------------------- /config/gpt2_nano_tb.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/config/gpt2_nano_tb.yaml -------------------------------------------------------------------------------- /config/gpt2_small.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/config/gpt2_small.yaml -------------------------------------------------------------------------------- /config/gpt2_small_fast.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/config/gpt2_small_fast.yaml -------------------------------------------------------------------------------- /config/gpt2_small_fast_fp8.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/config/gpt2_small_fast_fp8.yaml -------------------------------------------------------------------------------- /config/gpt2_small_fast_mix.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/config/gpt2_small_fast_mix.yaml -------------------------------------------------------------------------------- /config/gpt2_small_fast_pile.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/config/gpt2_small_fast_pile.yaml -------------------------------------------------------------------------------- /config/gpt2_small_fast_public.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/config/gpt2_small_fast_public.yaml -------------------------------------------------------------------------------- /config/gpt2_small_fast_sophia_h.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/config/gpt2_small_fast_sophia_h.yaml -------------------------------------------------------------------------------- /config/gpt2_small_fast_sophiah.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/config/gpt2_small_fast_sophiah.yaml -------------------------------------------------------------------------------- /config/gpt2_small_fast_wiki.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/config/gpt2_small_fast_wiki.yaml -------------------------------------------------------------------------------- /config/gpt2_small_pile.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/config/gpt2_small_pile.yaml -------------------------------------------------------------------------------- /config/gpt2_small_pile_mixture.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/config/gpt2_small_pile_mixture.yaml -------------------------------------------------------------------------------- /config/gpt2_small_sophiah.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/config/gpt2_small_sophiah.yaml -------------------------------------------------------------------------------- /config/gpt2_xl.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/config/gpt2_xl.yaml -------------------------------------------------------------------------------- /config/llama2_3b_pretrain.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/config/llama2_3b_pretrain.yaml -------------------------------------------------------------------------------- /config/llama2_7b.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/config/llama2_7b.yaml -------------------------------------------------------------------------------- /config/llama2_7b_continued.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/config/llama2_7b_continued.yaml -------------------------------------------------------------------------------- /config/llama2_nano.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/config/llama2_nano.yaml -------------------------------------------------------------------------------- /config/llama_small_fast.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/config/llama_small_fast.yaml -------------------------------------------------------------------------------- /config/lora/mpt_biomed.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/config/lora/mpt_biomed.yaml -------------------------------------------------------------------------------- /config/lora_llama2.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/config/lora_llama2.yaml -------------------------------------------------------------------------------- /config/mistral_7b.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/config/mistral_7b.yaml -------------------------------------------------------------------------------- /config/optim/sophia-h_large.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/config/optim/sophia-h_large.yaml -------------------------------------------------------------------------------- /config/optim/sophia-h_medium.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/config/optim/sophia-h_medium.yaml -------------------------------------------------------------------------------- /config/optim/sophia-h_small.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/config/optim/sophia-h_small.yaml -------------------------------------------------------------------------------- /config/optim/sophia-h_xl.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/config/optim/sophia-h_xl.yaml -------------------------------------------------------------------------------- /config/whisper_tiny_librispeech.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/config/whisper_tiny_librispeech.yaml -------------------------------------------------------------------------------- /docker/nvidia/Dockerfile: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/docker/nvidia/Dockerfile -------------------------------------------------------------------------------- /docs/Configuration-Guide.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/docs/Configuration-Guide.md -------------------------------------------------------------------------------- /docs/Fine-Tuning.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/docs/Fine-Tuning.md -------------------------------------------------------------------------------- /docs/Getting-Started-GPU.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/docs/Getting-Started-GPU.md -------------------------------------------------------------------------------- /docs/Getting-Started-TPU-VM.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/docs/Getting-Started-TPU-VM.md -------------------------------------------------------------------------------- /docs/Getting-Started-Training.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/docs/Getting-Started-Training.md -------------------------------------------------------------------------------- /docs/Hardware-Agnostic-Training.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/docs/Hardware-Agnostic-Training.md -------------------------------------------------------------------------------- /docs/Installation.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/docs/Installation.md -------------------------------------------------------------------------------- /docs/Levanter-1.0-Release.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/docs/Levanter-1.0-Release.md -------------------------------------------------------------------------------- /docs/LoRA.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/docs/LoRA.md -------------------------------------------------------------------------------- /docs/Performance-Guide.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/docs/Performance-Guide.md -------------------------------------------------------------------------------- /docs/Training-On-Your-Data.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/docs/Training-On-Your-Data.md -------------------------------------------------------------------------------- /docs/css/custom.css: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/docs/css/custom.css -------------------------------------------------------------------------------- /docs/css/friendly.css: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/docs/css/friendly.css -------------------------------------------------------------------------------- /docs/css/mkdocstrings.css: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/docs/css/mkdocstrings.css -------------------------------------------------------------------------------- /docs/design/Data-Loader-Design.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/docs/design/Data-Loader-Design.md -------------------------------------------------------------------------------- /docs/design/Multiple-Data-Mixture.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/docs/design/Multiple-Data-Mixture.md -------------------------------------------------------------------------------- /docs/dev/GPU-Docker-Dev.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/docs/dev/GPU-Docker-Dev.md -------------------------------------------------------------------------------- /docs/dev/Port-Models.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/docs/dev/Port-Models.md -------------------------------------------------------------------------------- /docs/dev/Trackers.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/docs/dev/Trackers.md -------------------------------------------------------------------------------- /docs/faq.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/docs/faq.md -------------------------------------------------------------------------------- /docs/figures/bitwise_repro_curve.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/docs/figures/bitwise_repro_curve.png -------------------------------------------------------------------------------- /docs/figures/data_parallel_mesh.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/docs/figures/data_parallel_mesh.png -------------------------------------------------------------------------------- /docs/figures/data_parallel_mesh_replicated.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/docs/figures/data_parallel_mesh_replicated.png -------------------------------------------------------------------------------- /docs/figures/device_mesh_1d.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/docs/figures/device_mesh_1d.png -------------------------------------------------------------------------------- /docs/figures/device_mesh_1d_zero.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/docs/figures/device_mesh_1d_zero.png -------------------------------------------------------------------------------- /docs/figures/device_mesh_2d.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/docs/figures/device_mesh_2d.png -------------------------------------------------------------------------------- /docs/figures/device_mesh_2d_batch_partitioned.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/docs/figures/device_mesh_2d_batch_partitioned.png -------------------------------------------------------------------------------- /docs/figures/device_mesh_2d_data_replicated.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/docs/figures/device_mesh_2d_data_replicated.png -------------------------------------------------------------------------------- /docs/figures/device_mesh_2d_data_replicated_mlp_partitioned.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/docs/figures/device_mesh_2d_data_replicated_mlp_partitioned.png -------------------------------------------------------------------------------- /docs/figures/device_mesh_2d_intermediate_fully_partitioned.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/docs/figures/device_mesh_2d_intermediate_fully_partitioned.png -------------------------------------------------------------------------------- /docs/figures/device_mesh_2d_zero.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/docs/figures/device_mesh_2d_zero.png -------------------------------------------------------------------------------- /docs/figures/finetune_func_cm_full_weight.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/docs/figures/finetune_func_cm_full_weight.png -------------------------------------------------------------------------------- /docs/figures/finetune_func_cm_lora.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/docs/figures/finetune_func_cm_lora.png -------------------------------------------------------------------------------- /docs/figures/helm-gsm8k-results.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/docs/figures/helm-gsm8k-results.png -------------------------------------------------------------------------------- /docs/figures/helm-instance-example.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/docs/figures/helm-instance-example.png -------------------------------------------------------------------------------- /docs/figures/lora-diagram.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/docs/figures/lora-diagram.png -------------------------------------------------------------------------------- /docs/figures/palm_mfu_table.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/docs/figures/palm_mfu_table.png -------------------------------------------------------------------------------- /docs/figures/resumed_curve.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/docs/figures/resumed_curve.png -------------------------------------------------------------------------------- /docs/figures/stopped_curve.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/docs/figures/stopped_curve.png -------------------------------------------------------------------------------- /docs/figures/token_probabilities.mov: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/docs/figures/token_probabilities.mov -------------------------------------------------------------------------------- /docs/index.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/docs/index.md -------------------------------------------------------------------------------- /docs/javascripts/mathjax.js: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/docs/javascripts/mathjax.js -------------------------------------------------------------------------------- /docs/requirements.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/docs/requirements.txt -------------------------------------------------------------------------------- /docs/tutorials/Fine-Tuning-Semantic-Parsing.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/docs/tutorials/Fine-Tuning-Semantic-Parsing.md -------------------------------------------------------------------------------- /docs/tutorials/Training-On-Audio-Data.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/docs/tutorials/Training-On-Audio-Data.md -------------------------------------------------------------------------------- /examples/alpaca-lora/alpaca-lora-llama2.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/examples/alpaca-lora/alpaca-lora-llama2.yaml -------------------------------------------------------------------------------- /examples/alpaca-lora/alpaca-lora.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/examples/alpaca-lora/alpaca-lora.yaml -------------------------------------------------------------------------------- /examples/alpaca-lora/alpaca_lora.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/examples/alpaca-lora/alpaca_lora.py -------------------------------------------------------------------------------- /examples/alpaca-lora/code-alpaca-lora.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/examples/alpaca-lora/code-alpaca-lora.yaml -------------------------------------------------------------------------------- /examples/alpaca-lora/hf_lora_inference.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/examples/alpaca-lora/hf_lora_inference.py -------------------------------------------------------------------------------- /examples/alpaca-lora/peft_inference.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/examples/alpaca-lora/peft_inference.ipynb -------------------------------------------------------------------------------- /examples/alpaca/alpaca-llama2.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/examples/alpaca/alpaca-llama2.yaml -------------------------------------------------------------------------------- /examples/alpaca/alpaca.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/examples/alpaca/alpaca.py -------------------------------------------------------------------------------- /examples/alpaca/alpaca.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/examples/alpaca/alpaca.yaml -------------------------------------------------------------------------------- /examples/gsm8k-lora/gsm8k-llama2.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/examples/gsm8k-lora/gsm8k-llama2.yaml -------------------------------------------------------------------------------- /examples/gsm8k-lora/gsm8k_lora.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/examples/gsm8k-lora/gsm8k_lora.py -------------------------------------------------------------------------------- /infra/babysit-tpu-vm: -------------------------------------------------------------------------------- 1 | babysit-tpu-vm.sh -------------------------------------------------------------------------------- /infra/babysit-tpu-vm.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/infra/babysit-tpu-vm.sh -------------------------------------------------------------------------------- /infra/helpers/gen-id.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/infra/helpers/gen-id.sh -------------------------------------------------------------------------------- /infra/helpers/parse-tpu-creation-args.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/infra/helpers/parse-tpu-creation-args.sh -------------------------------------------------------------------------------- /infra/helpers/setup-tpu-vm-nfs.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/infra/helpers/setup-tpu-vm-nfs.sh -------------------------------------------------------------------------------- /infra/helpers/setup-tpu-vm.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/infra/helpers/setup-tpu-vm.sh -------------------------------------------------------------------------------- /infra/launch.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/infra/launch.sh -------------------------------------------------------------------------------- /infra/run-slurm.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/infra/run-slurm.sh -------------------------------------------------------------------------------- /infra/run.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/infra/run.sh -------------------------------------------------------------------------------- /infra/spin-up-vm.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/infra/spin-up-vm.sh -------------------------------------------------------------------------------- /mkdocs.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/mkdocs.yml -------------------------------------------------------------------------------- /pyproject.toml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/pyproject.toml -------------------------------------------------------------------------------- /scripts/clean_old_checkpoints.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/scripts/clean_old_checkpoints.py -------------------------------------------------------------------------------- /scripts/launch_gpt2_small_fast_gpu.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/scripts/launch_gpt2_small_fast_gpu.sh -------------------------------------------------------------------------------- /scripts/launch_gpt2_small_fast_tpu.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/scripts/launch_gpt2_small_fast_tpu.sh -------------------------------------------------------------------------------- /scripts/loss_history.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/scripts/loss_history.py -------------------------------------------------------------------------------- /scripts/preproc/split-pile-shards.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/scripts/preproc/split-pile-shards.py -------------------------------------------------------------------------------- /scripts/repair_cache.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/scripts/repair_cache.py -------------------------------------------------------------------------------- /scripts/train_tokenizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/scripts/train_tokenizer.py -------------------------------------------------------------------------------- /src/levanter/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/src/levanter/__init__.py -------------------------------------------------------------------------------- /src/levanter/callbacks.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/src/levanter/callbacks.py -------------------------------------------------------------------------------- /src/levanter/checkpoint.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/src/levanter/checkpoint.py -------------------------------------------------------------------------------- /src/levanter/compat/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /src/levanter/compat/hf_checkpoints.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/src/levanter/compat/hf_checkpoints.py -------------------------------------------------------------------------------- /src/levanter/compat/torch_serialization.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/src/levanter/compat/torch_serialization.py -------------------------------------------------------------------------------- /src/levanter/config.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/src/levanter/config.py -------------------------------------------------------------------------------- /src/levanter/data/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/src/levanter/data/__init__.py -------------------------------------------------------------------------------- /src/levanter/data/_preprocessor.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/src/levanter/data/_preprocessor.py -------------------------------------------------------------------------------- /src/levanter/data/audio.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/src/levanter/data/audio.py -------------------------------------------------------------------------------- /src/levanter/data/dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/src/levanter/data/dataset.py -------------------------------------------------------------------------------- /src/levanter/data/loader.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/src/levanter/data/loader.py -------------------------------------------------------------------------------- /src/levanter/data/mixture.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/src/levanter/data/mixture.py -------------------------------------------------------------------------------- /src/levanter/data/shard_cache.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/src/levanter/data/shard_cache.py -------------------------------------------------------------------------------- /src/levanter/data/sharded_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/src/levanter/data/sharded_dataset.py -------------------------------------------------------------------------------- /src/levanter/data/text.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/src/levanter/data/text.py -------------------------------------------------------------------------------- /src/levanter/data/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/src/levanter/data/utils.py -------------------------------------------------------------------------------- /src/levanter/distributed.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/src/levanter/distributed.py -------------------------------------------------------------------------------- /src/levanter/doremi.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/src/levanter/doremi.py -------------------------------------------------------------------------------- /src/levanter/eval.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/src/levanter/eval.py -------------------------------------------------------------------------------- /src/levanter/grad_accum.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/src/levanter/grad_accum.py -------------------------------------------------------------------------------- /src/levanter/logging.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/src/levanter/logging.py -------------------------------------------------------------------------------- /src/levanter/lora.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/src/levanter/lora.py -------------------------------------------------------------------------------- /src/levanter/main/cache_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/src/levanter/main/cache_dataset.py -------------------------------------------------------------------------------- /src/levanter/main/doremi_lm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/src/levanter/main/doremi_lm.py -------------------------------------------------------------------------------- /src/levanter/main/eval_lm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/src/levanter/main/eval_lm.py -------------------------------------------------------------------------------- /src/levanter/main/export_lm_to_hf.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/src/levanter/main/export_lm_to_hf.py -------------------------------------------------------------------------------- /src/levanter/main/lora_lm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/src/levanter/main/lora_lm.py -------------------------------------------------------------------------------- /src/levanter/main/train_asr.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/src/levanter/main/train_asr.py -------------------------------------------------------------------------------- /src/levanter/main/train_lm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/src/levanter/main/train_lm.py -------------------------------------------------------------------------------- /src/levanter/main/viz_logprobs.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/src/levanter/main/viz_logprobs.py -------------------------------------------------------------------------------- /src/levanter/mesh.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/src/levanter/mesh.py -------------------------------------------------------------------------------- /src/levanter/models/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /src/levanter/models/asr_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/src/levanter/models/asr_model.py -------------------------------------------------------------------------------- /src/levanter/models/attention.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/src/levanter/models/attention.py -------------------------------------------------------------------------------- /src/levanter/models/backpack.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/src/levanter/models/backpack.py -------------------------------------------------------------------------------- /src/levanter/models/flash_attention.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/src/levanter/models/flash_attention.py -------------------------------------------------------------------------------- /src/levanter/models/gpt2.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/src/levanter/models/gpt2.py -------------------------------------------------------------------------------- /src/levanter/models/llama.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/src/levanter/models/llama.py -------------------------------------------------------------------------------- /src/levanter/models/lm_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/src/levanter/models/lm_model.py -------------------------------------------------------------------------------- /src/levanter/models/longformer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/src/levanter/models/longformer.py -------------------------------------------------------------------------------- /src/levanter/models/loss.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/src/levanter/models/loss.py -------------------------------------------------------------------------------- /src/levanter/models/mistral.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/src/levanter/models/mistral.py -------------------------------------------------------------------------------- /src/levanter/models/mpt.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/src/levanter/models/mpt.py -------------------------------------------------------------------------------- /src/levanter/models/whisper.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/src/levanter/models/whisper.py -------------------------------------------------------------------------------- /src/levanter/optim/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/src/levanter/optim/__init__.py -------------------------------------------------------------------------------- /src/levanter/optim/config.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/src/levanter/optim/config.py -------------------------------------------------------------------------------- /src/levanter/optim/sophia.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/src/levanter/optim/sophia.py -------------------------------------------------------------------------------- /src/levanter/optim/util.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/src/levanter/optim/util.py -------------------------------------------------------------------------------- /src/levanter/shapes.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/src/levanter/shapes.py -------------------------------------------------------------------------------- /src/levanter/tensorstore_serialization.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/src/levanter/tensorstore_serialization.py -------------------------------------------------------------------------------- /src/levanter/tracker/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/src/levanter/tracker/__init__.py -------------------------------------------------------------------------------- /src/levanter/tracker/helpers.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/src/levanter/tracker/helpers.py -------------------------------------------------------------------------------- /src/levanter/tracker/tensorboard.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/src/levanter/tracker/tensorboard.py -------------------------------------------------------------------------------- /src/levanter/tracker/tracker.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/src/levanter/tracker/tracker.py -------------------------------------------------------------------------------- /src/levanter/tracker/tracker_fns.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/src/levanter/tracker/tracker_fns.py -------------------------------------------------------------------------------- /src/levanter/tracker/wandb.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/src/levanter/tracker/wandb.py -------------------------------------------------------------------------------- /src/levanter/trainer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/src/levanter/trainer.py -------------------------------------------------------------------------------- /src/levanter/trainer_state.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/src/levanter/trainer_state.py -------------------------------------------------------------------------------- /src/levanter/types.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/src/levanter/types.py -------------------------------------------------------------------------------- /src/levanter/utils/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /src/levanter/utils/background_iterable.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/src/levanter/utils/background_iterable.py -------------------------------------------------------------------------------- /src/levanter/utils/cloud_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/src/levanter/utils/cloud_utils.py -------------------------------------------------------------------------------- /src/levanter/utils/datetime_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/src/levanter/utils/datetime_utils.py -------------------------------------------------------------------------------- /src/levanter/utils/fsspec_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/src/levanter/utils/fsspec_utils.py -------------------------------------------------------------------------------- /src/levanter/utils/hf_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/src/levanter/utils/hf_utils.py -------------------------------------------------------------------------------- /src/levanter/utils/jax_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/src/levanter/utils/jax_utils.py -------------------------------------------------------------------------------- /src/levanter/utils/py_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/src/levanter/utils/py_utils.py -------------------------------------------------------------------------------- /src/levanter/utils/ray_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/src/levanter/utils/ray_utils.py -------------------------------------------------------------------------------- /src/levanter/utils/stat_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/src/levanter/utils/stat_utils.py -------------------------------------------------------------------------------- /src/levanter/utils/tree_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/src/levanter/utils/tree_utils.py -------------------------------------------------------------------------------- /src/levanter/visualization.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/src/levanter/visualization.py -------------------------------------------------------------------------------- /tests/data/hero_data.npy: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/tests/data/hero_data.npy -------------------------------------------------------------------------------- /tests/gpt2_test.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/tests/gpt2_test.py -------------------------------------------------------------------------------- /tests/gpt2_tokenizer_config.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/tests/gpt2_tokenizer_config.json -------------------------------------------------------------------------------- /tests/test_attention.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/tests/test_attention.py -------------------------------------------------------------------------------- /tests/test_audio.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/tests/test_audio.py -------------------------------------------------------------------------------- /tests/test_background_iterable.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/tests/test_background_iterable.py -------------------------------------------------------------------------------- /tests/test_backpack.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/tests/test_backpack.py -------------------------------------------------------------------------------- /tests/test_checkpoint.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/tests/test_checkpoint.py -------------------------------------------------------------------------------- /tests/test_config.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/tests/test_config.py -------------------------------------------------------------------------------- /tests/test_data_mixture.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/tests/test_data_mixture.py -------------------------------------------------------------------------------- /tests/test_datetime_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/tests/test_datetime_utils.py -------------------------------------------------------------------------------- /tests/test_distributed.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/tests/test_distributed.py -------------------------------------------------------------------------------- /tests/test_doremi.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/tests/test_doremi.py -------------------------------------------------------------------------------- /tests/test_eval_lm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/tests/test_eval_lm.py -------------------------------------------------------------------------------- /tests/test_export_to_hf.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/tests/test_export_to_hf.py -------------------------------------------------------------------------------- /tests/test_flash_attention.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/tests/test_flash_attention.py -------------------------------------------------------------------------------- /tests/test_grad_accum.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/tests/test_grad_accum.py -------------------------------------------------------------------------------- /tests/test_hf_checkpoints.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/tests/test_hf_checkpoints.py -------------------------------------------------------------------------------- /tests/test_hf_gpt2_serialize.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/tests/test_hf_gpt2_serialize.py -------------------------------------------------------------------------------- /tests/test_hf_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/tests/test_hf_utils.py -------------------------------------------------------------------------------- /tests/test_jax_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/tests/test_jax_utils.py -------------------------------------------------------------------------------- /tests/test_llama.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/tests/test_llama.py -------------------------------------------------------------------------------- /tests/test_logging.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/tests/test_logging.py -------------------------------------------------------------------------------- /tests/test_longformer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/tests/test_longformer.py -------------------------------------------------------------------------------- /tests/test_lora.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/tests/test_lora.py -------------------------------------------------------------------------------- /tests/test_mistral.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/tests/test_mistral.py -------------------------------------------------------------------------------- /tests/test_mpt.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/tests/test_mpt.py -------------------------------------------------------------------------------- /tests/test_py_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/tests/test_py_utils.py -------------------------------------------------------------------------------- /tests/test_replicated_loader.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/tests/test_replicated_loader.py -------------------------------------------------------------------------------- /tests/test_shard_cache.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/tests/test_shard_cache.py -------------------------------------------------------------------------------- /tests/test_sharded_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/tests/test_sharded_dataset.py -------------------------------------------------------------------------------- /tests/test_sharded_loader.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/tests/test_sharded_loader.py -------------------------------------------------------------------------------- /tests/test_shuffle_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/tests/test_shuffle_dataset.py -------------------------------------------------------------------------------- /tests/test_sophia.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/tests/test_sophia.py -------------------------------------------------------------------------------- /tests/test_tensorstore_serialization.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/tests/test_tensorstore_serialization.py -------------------------------------------------------------------------------- /tests/test_text.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/tests/test_text.py -------------------------------------------------------------------------------- /tests/test_tokenized_document_cache.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/tests/test_tokenized_document_cache.py -------------------------------------------------------------------------------- /tests/test_torch_serialization.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/tests/test_torch_serialization.py -------------------------------------------------------------------------------- /tests/test_tracker.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/tests/test_tracker.py -------------------------------------------------------------------------------- /tests/test_train_asr.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/tests/test_train_asr.py -------------------------------------------------------------------------------- /tests/test_train_lm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/tests/test_train_lm.py -------------------------------------------------------------------------------- /tests/test_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/tests/test_utils.py -------------------------------------------------------------------------------- /tests/test_viz_lm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/tests/test_viz_lm.py -------------------------------------------------------------------------------- /tests/test_weight_decay_mask.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/tests/test_weight_decay_mask.py -------------------------------------------------------------------------------- /tests/tiny_test_corpus.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/tests/tiny_test_corpus.py -------------------------------------------------------------------------------- /tests/whisper_test.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Helw150/levanter/HEAD/tests/whisper_test.py --------------------------------------------------------------------------------