├── .flake8 ├── .github ├── CODEOWNERS ├── repo_meta.yaml └── workflows │ ├── docs.yaml │ ├── formatting.yaml │ ├── gpu_unit_tests.yaml │ ├── semgroup.yaml │ └── unit_tests.yaml ├── .gitignore ├── .isort.cfg ├── .pre-commit-config.yaml ├── .readthedocs.yaml ├── CONTRIBUTING.md ├── LICENSE ├── Makefile ├── README.md ├── STYLE_GUIDE.md ├── arctic_training ├── __init__.py ├── callback │ ├── __init__.py │ ├── callback.py │ ├── logging.py │ └── mixin.py ├── checkpoint │ ├── __init__.py │ ├── ds_engine.py │ ├── engine.py │ └── hf_engine.py ├── config │ ├── __init__.py │ ├── base.py │ ├── checkpoint.py │ ├── data.py │ ├── enums.py │ ├── logger.py │ ├── model.py │ ├── optimizer.py │ ├── scheduler.py │ ├── tokenizer.py │ ├── trainer.py │ ├── utils.py │ └── wandb.py ├── data │ ├── __init__.py │ ├── causal_factory.py │ ├── dpo_factory.py │ ├── factory.py │ ├── hf_causal_source.py │ ├── hf_instruct_source.py │ ├── hf_source.py │ ├── sft_factory.py │ ├── source.py │ └── utils.py ├── debug.py ├── entrypoint.py ├── exceptions.py ├── launcher │ └── ray_launcher.py ├── logging.py ├── metrics.py ├── model │ ├── __init__.py │ ├── factory.py │ ├── hf_factory.py │ ├── liger_factory.py │ └── tiled_compute.py ├── monkey_patches.py ├── optimizer │ ├── __init__.py │ ├── adam_factory.py │ └── factory.py ├── registry.py ├── scheduler │ ├── __init__.py │ ├── factory.py │ ├── hf_factory.py │ └── wsd_factory.py ├── synth │ ├── README.md │ ├── __init__.py │ ├── base_caller.py │ ├── callers.py │ ├── cli.py │ ├── openai_callers.py │ ├── test_cortex_caller.py │ ├── test_multi_replica_vllm_caller.py │ ├── test_openai_callers.py │ ├── test_vllm_caller.py │ ├── utils.py │ ├── vllm_callers.py │ └── vllm_utils.py ├── testing_utils.py ├── tokenizer │ ├── __init__.py │ ├── factory.py │ └── hf_factory.py ├── trainer │ ├── __init__.py │ ├── causal_trainer.py │ ├── dpo_trainer.py │ ├── flops_counter.py │ ├── sft_trainer.py │ ├── trainer.py │ └── utils.py └── utils.py ├── arctic_training_cli.py ├── ci ├── __init__.py └── gpu_unit_tests.py ├── docs ├── Makefile ├── callbacks.rst ├── checkpoint.rst ├── conf.py ├── config.rst ├── data.rst ├── images │ └── arctic_training_logo.svg ├── index.rst ├── install.rst ├── make.bat ├── model.rst ├── optimizer.rst ├── quick-start.rst ├── scheduler.rst ├── synth.rst ├── tokenizer.rst ├── trainer.rst └── usage.rst ├── mypy.ini ├── projects ├── arctic_embed │ ├── README.md │ ├── examples │ │ └── finetune_models │ │ │ ├── README.md │ │ │ ├── data │ │ │ └── .gitignore │ │ │ ├── data_download_scripts │ │ │ ├── download_fever.py │ │ │ ├── download_hotpotqa.py │ │ │ ├── download_hotpotqa_from_bge_data.py │ │ │ ├── download_msmarco.py │ │ │ ├── download_nq.py │ │ │ └── download_stackexchange.py │ │ │ ├── data_processing_scripts │ │ │ ├── stage_1_embed.py │ │ │ ├── stage_2_score_and_retrieve.py │ │ │ ├── stage_3_mine.py │ │ │ ├── stage_4_combine.py │ │ │ ├── stage_5_tokenize_and_batch.py │ │ │ └── stage_6_get_dev_splits.py │ │ │ └── finetune_e5_base_unsupervised.py │ ├── pyproject.toml │ ├── src │ │ └── arctic_embed │ │ │ ├── __init__.py │ │ │ ├── biencoder_model_factory.py │ │ │ ├── checkpointing.py │ │ │ ├── contrastive_dataloader.py │ │ │ ├── core │ │ │ ├── __init__.py │ │ │ ├── biencoder_model.py │ │ │ ├── cuda_allocator_config.py │ │ │ ├── losses.py │ │ │ ├── pretokenized_batch_loader.py │ │ │ └── slice_sparse_coo_tensor.py │ │ │ ├── data_processing │ │ │ ├── __init__.py │ │ │ ├── dense_retrieval.py │ │ │ ├── many_parquet_dataset.py │ │ │ ├── pretokenize.py │ │ │ ├── teacher_models.py │ │ │ ├── typing_utils.py │ │ │ └── utils.py │ │ │ └── trainer.py │ └── tests │ │ ├── test_biencoder_model.py │ │ ├── test_losses.py │ │ └── test_slice_sparse_coo_tensor.py ├── arctic_lstm_speculator │ ├── README.md │ ├── llama3.1-70b.yaml │ ├── llama3.1-8b.yaml │ ├── llama3.3-70b.yaml │ ├── multigpu_runner.py │ ├── oss-120b.yaml │ ├── oss-20b.yaml │ ├── qwen2.5-32b.yaml │ ├── scripts │ │ ├── llama3.1-70b.sh │ │ ├── llama3.1-8b.sh │ │ ├── llama3.3-70b.sh │ │ ├── oss-120b.sh │ │ ├── oss-20b.sh │ │ └── qwen2.5-32b.sh │ ├── speculator │ │ ├── __init__.py │ │ ├── configs.py │ │ ├── data_generation │ │ │ ├── __init__.py │ │ │ ├── concat_generated_datasets.py │ │ │ ├── data_gen_script_maker.py │ │ │ └── vllm_data_generation.py │ │ └── speculator.py │ └── train.py ├── arctic_text2sql_r1 │ ├── README.md │ ├── assets │ │ └── leaderboard_plot_average_styled.png │ ├── bird_eval │ │ ├── auto_evaluation.py │ │ ├── eval_open_source_models.py │ │ ├── evaluate_bird.py │ │ ├── infer.py │ │ └── self_consistency.py │ ├── data_preprocessing │ │ ├── build_contents_index.py │ │ ├── nltk_downloader.py │ │ ├── process_dataset.py │ │ └── process_dataset.sh │ └── modelcard.md ├── causal │ ├── README.md │ └── run-causal.yml ├── excot_dpo │ ├── README.md │ ├── data_generation │ │ ├── configs │ │ │ ├── bird_config.yaml │ │ │ └── spider_config.yaml │ │ ├── data_generation.py │ │ ├── dpo_sample.py │ │ ├── local_verification.py │ │ ├── prompts │ │ │ └── divide_and_conquer.py │ │ └── sft_sample.py │ ├── dpo-llama-8b.yaml │ ├── environment.yaml │ ├── evalution │ │ ├── configs │ │ │ ├── bird_config.yaml │ │ │ └── spider_config.yaml │ │ ├── eval_w_arctic_syth.py │ │ ├── execute_utils.py │ │ ├── prompts │ │ │ └── divide_and_conquer.py │ │ └── sql_exec.py │ ├── requirements.txt │ ├── sft-llama-8b.yaml │ └── utils │ │ ├── __init__.py │ │ ├── execute_utils.py │ │ └── sql_exec.py ├── mlp_speculator │ ├── llama-8b.yaml │ ├── mlp_speculator │ │ ├── __init__.py │ │ ├── configs.py │ │ ├── data_generation │ │ │ └── vllm_multi_replica_gen_hf_datasets.py │ │ └── speculator.py │ └── train.py ├── moba_attention │ ├── moba-swiftkv-llama-8b.yaml │ ├── moba │ │ ├── __init__.py │ │ ├── config.py │ │ ├── llama.py │ │ ├── moba_efficient.py │ │ ├── moba_naive.py │ │ ├── moba_with_flash_interface.py │ │ └── wrapper.py │ └── train.py ├── sequence-parallelism │ ├── README.md │ ├── USAGE.md │ ├── images │ │ ├── correctness-plot.png │ │ └── loss-computation.png │ ├── run-h100-sp1-llama-8b-baseline.yml │ ├── run-h100-sp1-llama-8b.yml │ ├── run-h100-sp1-qwen3-32b.yml │ ├── run-h100-sp16-llama-70b.yml │ ├── run-h100-sp16-llama-8b.yml │ ├── run-h100-sp16-qwen3-32b.yml │ ├── run-h100-sp32-llama-70b.yml │ ├── run-h100-sp32-llama-8b-baseline.yml │ ├── run-h100-sp32-llama-8b.yml │ ├── run-h100-sp32-qwen3-32b.yml │ ├── run-h100-sp64-llama-70b.yml │ ├── run-h100-sp64-qwen3-32b.yml │ ├── run-h100-sp8-llama-70b.yml │ ├── run-h100-sp8-llama-8b-no-mlp-no-act-offload.yml │ ├── run-h100-sp8-llama-8b-no-ulysses-no-liger-no-extras.yml │ ├── run-h100-sp8-llama-8b-no-ulysses-yes-liger-no-extras.yml │ ├── run-h100-sp8-llama-8b-yes-act-offload.yml │ ├── run-h100-sp8-llama-8b-yes-mlp.yml │ ├── run-h100-sp8-llama-8b.yml │ ├── run-h100-sp8-qwen3-32b.yml │ ├── run-h200-sp1-llama-8b-baseline.yml │ ├── run-h200-sp1-llama-8b-liger-offload-tiled-mlp.yml │ ├── run-h200-sp1-llama-8b-liger-offload.yml │ └── run-h200-sp1-llama-8b-liger.yml └── swiftkv │ ├── README.md │ ├── configs │ ├── deepseek-v2-swiftkv-lite-chat.yaml │ ├── llama-3.1-swiftkv-405b-instruct.yaml │ ├── llama-3.1-swiftkv-70b-instruct.yaml │ ├── llama-3.1-swiftkv-8b-instruct-long.yaml │ ├── llama-3.1-swiftkv-8b-instruct.yaml │ ├── llama-3.3-swiftkv-70b-instruct-long.yaml │ ├── qwen2.5-swiftkv-14b-instruct.yaml │ ├── qwen2.5-swiftkv-32b-instruct.yaml │ └── qwen2.5-swiftkv-7b-instruct.yaml │ ├── eval.py │ ├── models │ ├── __init__.py │ ├── deepseek_v2 │ │ ├── __init__.py │ │ ├── configuration_deepseek.py │ │ ├── configuration_deepseek_swiftkv.py │ │ ├── modeling_deepseek.py │ │ └── modeling_deepseek_swiftkv.py │ ├── llama │ │ ├── __init__.py │ │ ├── configuration_llama_swiftkv.py │ │ └── modeling_llama_swiftkv.py │ └── qwen2 │ │ ├── __init__.py │ │ ├── configuration_qwen2_swiftkv.py │ │ └── modeling_qwen2_swiftkv.py │ └── train.py ├── pyproject.toml ├── scripts ├── check_license.py ├── download_data.py ├── install_deps.py ├── release.sh └── upgrade_user_code.py └── tests ├── __init__.py ├── callback ├── __init__.py └── test_mixin.py ├── checkpoint ├── __init__.py ├── test_ds_engine.py ├── test_hf_engine.py └── utils.py ├── config ├── test_config_custom_types.py ├── test_model_config.py └── test_trainer_config.py ├── conftest.py ├── data ├── __init__.py ├── test_factory.py ├── test_hf_instruct_source.py ├── test_hf_source.py ├── test_sft_factory.py ├── test_source.py └── utils.py ├── helpers.py ├── model └── test_hf_factory.py ├── pytest.ini ├── test_registry.py ├── trainer ├── test_causal_trainer.py ├── test_dpo_trainer.py ├── test_sft_trainer.py ├── test_ulysses_alst.py └── test_wandb.py └── utils.py /.flake8: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/.flake8 -------------------------------------------------------------------------------- /.github/CODEOWNERS: -------------------------------------------------------------------------------- 1 | * @sfc-gh-jrasley 2 | -------------------------------------------------------------------------------- /.github/repo_meta.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/.github/repo_meta.yaml -------------------------------------------------------------------------------- /.github/workflows/docs.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/.github/workflows/docs.yaml -------------------------------------------------------------------------------- /.github/workflows/formatting.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/.github/workflows/formatting.yaml -------------------------------------------------------------------------------- /.github/workflows/gpu_unit_tests.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/.github/workflows/gpu_unit_tests.yaml -------------------------------------------------------------------------------- /.github/workflows/semgroup.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/.github/workflows/semgroup.yaml -------------------------------------------------------------------------------- /.github/workflows/unit_tests.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/.github/workflows/unit_tests.yaml -------------------------------------------------------------------------------- /.gitignore: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/.gitignore -------------------------------------------------------------------------------- /.isort.cfg: -------------------------------------------------------------------------------- 1 | [settings] 2 | known_third_party = wandb 3 | -------------------------------------------------------------------------------- /.pre-commit-config.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/.pre-commit-config.yaml -------------------------------------------------------------------------------- /.readthedocs.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/.readthedocs.yaml -------------------------------------------------------------------------------- /CONTRIBUTING.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/CONTRIBUTING.md -------------------------------------------------------------------------------- /LICENSE: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/LICENSE -------------------------------------------------------------------------------- /Makefile: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/Makefile -------------------------------------------------------------------------------- /README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/README.md -------------------------------------------------------------------------------- /STYLE_GUIDE.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/STYLE_GUIDE.md -------------------------------------------------------------------------------- /arctic_training/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/__init__.py -------------------------------------------------------------------------------- /arctic_training/callback/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/callback/__init__.py -------------------------------------------------------------------------------- /arctic_training/callback/callback.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/callback/callback.py -------------------------------------------------------------------------------- /arctic_training/callback/logging.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/callback/logging.py -------------------------------------------------------------------------------- /arctic_training/callback/mixin.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/callback/mixin.py -------------------------------------------------------------------------------- /arctic_training/checkpoint/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/checkpoint/__init__.py -------------------------------------------------------------------------------- /arctic_training/checkpoint/ds_engine.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/checkpoint/ds_engine.py -------------------------------------------------------------------------------- /arctic_training/checkpoint/engine.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/checkpoint/engine.py -------------------------------------------------------------------------------- /arctic_training/checkpoint/hf_engine.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/checkpoint/hf_engine.py -------------------------------------------------------------------------------- /arctic_training/config/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/config/__init__.py -------------------------------------------------------------------------------- /arctic_training/config/base.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/config/base.py -------------------------------------------------------------------------------- /arctic_training/config/checkpoint.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/config/checkpoint.py -------------------------------------------------------------------------------- /arctic_training/config/data.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/config/data.py -------------------------------------------------------------------------------- /arctic_training/config/enums.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/config/enums.py -------------------------------------------------------------------------------- /arctic_training/config/logger.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/config/logger.py -------------------------------------------------------------------------------- /arctic_training/config/model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/config/model.py -------------------------------------------------------------------------------- /arctic_training/config/optimizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/config/optimizer.py -------------------------------------------------------------------------------- /arctic_training/config/scheduler.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/config/scheduler.py -------------------------------------------------------------------------------- /arctic_training/config/tokenizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/config/tokenizer.py -------------------------------------------------------------------------------- /arctic_training/config/trainer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/config/trainer.py -------------------------------------------------------------------------------- /arctic_training/config/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/config/utils.py -------------------------------------------------------------------------------- /arctic_training/config/wandb.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/config/wandb.py -------------------------------------------------------------------------------- /arctic_training/data/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/data/__init__.py -------------------------------------------------------------------------------- /arctic_training/data/causal_factory.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/data/causal_factory.py -------------------------------------------------------------------------------- /arctic_training/data/dpo_factory.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/data/dpo_factory.py -------------------------------------------------------------------------------- /arctic_training/data/factory.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/data/factory.py -------------------------------------------------------------------------------- /arctic_training/data/hf_causal_source.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/data/hf_causal_source.py -------------------------------------------------------------------------------- /arctic_training/data/hf_instruct_source.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/data/hf_instruct_source.py -------------------------------------------------------------------------------- /arctic_training/data/hf_source.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/data/hf_source.py -------------------------------------------------------------------------------- /arctic_training/data/sft_factory.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/data/sft_factory.py -------------------------------------------------------------------------------- /arctic_training/data/source.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/data/source.py -------------------------------------------------------------------------------- /arctic_training/data/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/data/utils.py -------------------------------------------------------------------------------- /arctic_training/debug.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/debug.py -------------------------------------------------------------------------------- /arctic_training/entrypoint.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/entrypoint.py -------------------------------------------------------------------------------- /arctic_training/exceptions.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/exceptions.py -------------------------------------------------------------------------------- /arctic_training/launcher/ray_launcher.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/launcher/ray_launcher.py -------------------------------------------------------------------------------- /arctic_training/logging.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/logging.py -------------------------------------------------------------------------------- /arctic_training/metrics.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/metrics.py -------------------------------------------------------------------------------- /arctic_training/model/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/model/__init__.py -------------------------------------------------------------------------------- /arctic_training/model/factory.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/model/factory.py -------------------------------------------------------------------------------- /arctic_training/model/hf_factory.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/model/hf_factory.py -------------------------------------------------------------------------------- /arctic_training/model/liger_factory.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/model/liger_factory.py -------------------------------------------------------------------------------- /arctic_training/model/tiled_compute.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/model/tiled_compute.py -------------------------------------------------------------------------------- /arctic_training/monkey_patches.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/monkey_patches.py -------------------------------------------------------------------------------- /arctic_training/optimizer/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/optimizer/__init__.py -------------------------------------------------------------------------------- /arctic_training/optimizer/adam_factory.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/optimizer/adam_factory.py -------------------------------------------------------------------------------- /arctic_training/optimizer/factory.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/optimizer/factory.py -------------------------------------------------------------------------------- /arctic_training/registry.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/registry.py -------------------------------------------------------------------------------- /arctic_training/scheduler/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/scheduler/__init__.py -------------------------------------------------------------------------------- /arctic_training/scheduler/factory.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/scheduler/factory.py -------------------------------------------------------------------------------- /arctic_training/scheduler/hf_factory.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/scheduler/hf_factory.py -------------------------------------------------------------------------------- /arctic_training/scheduler/wsd_factory.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/scheduler/wsd_factory.py -------------------------------------------------------------------------------- /arctic_training/synth/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/synth/README.md -------------------------------------------------------------------------------- /arctic_training/synth/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/synth/__init__.py -------------------------------------------------------------------------------- /arctic_training/synth/base_caller.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/synth/base_caller.py -------------------------------------------------------------------------------- /arctic_training/synth/callers.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/synth/callers.py -------------------------------------------------------------------------------- /arctic_training/synth/cli.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/synth/cli.py -------------------------------------------------------------------------------- /arctic_training/synth/openai_callers.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/synth/openai_callers.py -------------------------------------------------------------------------------- /arctic_training/synth/test_cortex_caller.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/synth/test_cortex_caller.py -------------------------------------------------------------------------------- /arctic_training/synth/test_multi_replica_vllm_caller.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/synth/test_multi_replica_vllm_caller.py -------------------------------------------------------------------------------- /arctic_training/synth/test_openai_callers.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/synth/test_openai_callers.py -------------------------------------------------------------------------------- /arctic_training/synth/test_vllm_caller.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/synth/test_vllm_caller.py -------------------------------------------------------------------------------- /arctic_training/synth/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/synth/utils.py -------------------------------------------------------------------------------- /arctic_training/synth/vllm_callers.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/synth/vllm_callers.py -------------------------------------------------------------------------------- /arctic_training/synth/vllm_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/synth/vllm_utils.py -------------------------------------------------------------------------------- /arctic_training/testing_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/testing_utils.py -------------------------------------------------------------------------------- /arctic_training/tokenizer/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/tokenizer/__init__.py -------------------------------------------------------------------------------- /arctic_training/tokenizer/factory.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/tokenizer/factory.py -------------------------------------------------------------------------------- /arctic_training/tokenizer/hf_factory.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/tokenizer/hf_factory.py -------------------------------------------------------------------------------- /arctic_training/trainer/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/trainer/__init__.py -------------------------------------------------------------------------------- /arctic_training/trainer/causal_trainer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/trainer/causal_trainer.py -------------------------------------------------------------------------------- /arctic_training/trainer/dpo_trainer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/trainer/dpo_trainer.py -------------------------------------------------------------------------------- /arctic_training/trainer/flops_counter.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/trainer/flops_counter.py -------------------------------------------------------------------------------- /arctic_training/trainer/sft_trainer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/trainer/sft_trainer.py -------------------------------------------------------------------------------- /arctic_training/trainer/trainer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/trainer/trainer.py -------------------------------------------------------------------------------- /arctic_training/trainer/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/trainer/utils.py -------------------------------------------------------------------------------- /arctic_training/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training/utils.py -------------------------------------------------------------------------------- /arctic_training_cli.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/arctic_training_cli.py -------------------------------------------------------------------------------- /ci/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/ci/__init__.py -------------------------------------------------------------------------------- /ci/gpu_unit_tests.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/ci/gpu_unit_tests.py -------------------------------------------------------------------------------- /docs/Makefile: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/docs/Makefile -------------------------------------------------------------------------------- /docs/callbacks.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/docs/callbacks.rst -------------------------------------------------------------------------------- /docs/checkpoint.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/docs/checkpoint.rst -------------------------------------------------------------------------------- /docs/conf.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/docs/conf.py -------------------------------------------------------------------------------- /docs/config.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/docs/config.rst -------------------------------------------------------------------------------- /docs/data.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/docs/data.rst -------------------------------------------------------------------------------- /docs/images/arctic_training_logo.svg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/docs/images/arctic_training_logo.svg -------------------------------------------------------------------------------- /docs/index.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/docs/index.rst -------------------------------------------------------------------------------- /docs/install.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/docs/install.rst -------------------------------------------------------------------------------- /docs/make.bat: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/docs/make.bat -------------------------------------------------------------------------------- /docs/model.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/docs/model.rst -------------------------------------------------------------------------------- /docs/optimizer.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/docs/optimizer.rst -------------------------------------------------------------------------------- /docs/quick-start.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/docs/quick-start.rst -------------------------------------------------------------------------------- /docs/scheduler.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/docs/scheduler.rst -------------------------------------------------------------------------------- /docs/synth.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/docs/synth.rst -------------------------------------------------------------------------------- /docs/tokenizer.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/docs/tokenizer.rst -------------------------------------------------------------------------------- /docs/trainer.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/docs/trainer.rst -------------------------------------------------------------------------------- /docs/usage.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/docs/usage.rst -------------------------------------------------------------------------------- /mypy.ini: -------------------------------------------------------------------------------- 1 | [mypy] 2 | plugins = pydantic.mypy 3 | -------------------------------------------------------------------------------- /projects/arctic_embed/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/arctic_embed/README.md -------------------------------------------------------------------------------- /projects/arctic_embed/examples/finetune_models/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/arctic_embed/examples/finetune_models/README.md -------------------------------------------------------------------------------- /projects/arctic_embed/examples/finetune_models/data/.gitignore: -------------------------------------------------------------------------------- 1 | *.parquet 2 | -------------------------------------------------------------------------------- /projects/arctic_embed/examples/finetune_models/data_download_scripts/download_fever.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/arctic_embed/examples/finetune_models/data_download_scripts/download_fever.py -------------------------------------------------------------------------------- /projects/arctic_embed/examples/finetune_models/data_download_scripts/download_hotpotqa.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/arctic_embed/examples/finetune_models/data_download_scripts/download_hotpotqa.py -------------------------------------------------------------------------------- /projects/arctic_embed/examples/finetune_models/data_download_scripts/download_hotpotqa_from_bge_data.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/arctic_embed/examples/finetune_models/data_download_scripts/download_hotpotqa_from_bge_data.py -------------------------------------------------------------------------------- /projects/arctic_embed/examples/finetune_models/data_download_scripts/download_msmarco.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/arctic_embed/examples/finetune_models/data_download_scripts/download_msmarco.py -------------------------------------------------------------------------------- /projects/arctic_embed/examples/finetune_models/data_download_scripts/download_nq.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/arctic_embed/examples/finetune_models/data_download_scripts/download_nq.py -------------------------------------------------------------------------------- /projects/arctic_embed/examples/finetune_models/data_download_scripts/download_stackexchange.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/arctic_embed/examples/finetune_models/data_download_scripts/download_stackexchange.py -------------------------------------------------------------------------------- /projects/arctic_embed/examples/finetune_models/data_processing_scripts/stage_1_embed.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/arctic_embed/examples/finetune_models/data_processing_scripts/stage_1_embed.py -------------------------------------------------------------------------------- /projects/arctic_embed/examples/finetune_models/data_processing_scripts/stage_2_score_and_retrieve.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/arctic_embed/examples/finetune_models/data_processing_scripts/stage_2_score_and_retrieve.py -------------------------------------------------------------------------------- /projects/arctic_embed/examples/finetune_models/data_processing_scripts/stage_3_mine.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/arctic_embed/examples/finetune_models/data_processing_scripts/stage_3_mine.py -------------------------------------------------------------------------------- /projects/arctic_embed/examples/finetune_models/data_processing_scripts/stage_4_combine.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/arctic_embed/examples/finetune_models/data_processing_scripts/stage_4_combine.py -------------------------------------------------------------------------------- /projects/arctic_embed/examples/finetune_models/data_processing_scripts/stage_5_tokenize_and_batch.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/arctic_embed/examples/finetune_models/data_processing_scripts/stage_5_tokenize_and_batch.py -------------------------------------------------------------------------------- /projects/arctic_embed/examples/finetune_models/data_processing_scripts/stage_6_get_dev_splits.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/arctic_embed/examples/finetune_models/data_processing_scripts/stage_6_get_dev_splits.py -------------------------------------------------------------------------------- /projects/arctic_embed/examples/finetune_models/finetune_e5_base_unsupervised.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/arctic_embed/examples/finetune_models/finetune_e5_base_unsupervised.py -------------------------------------------------------------------------------- /projects/arctic_embed/pyproject.toml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/arctic_embed/pyproject.toml -------------------------------------------------------------------------------- /projects/arctic_embed/src/arctic_embed/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/arctic_embed/src/arctic_embed/__init__.py -------------------------------------------------------------------------------- /projects/arctic_embed/src/arctic_embed/biencoder_model_factory.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/arctic_embed/src/arctic_embed/biencoder_model_factory.py -------------------------------------------------------------------------------- /projects/arctic_embed/src/arctic_embed/checkpointing.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/arctic_embed/src/arctic_embed/checkpointing.py -------------------------------------------------------------------------------- /projects/arctic_embed/src/arctic_embed/contrastive_dataloader.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/arctic_embed/src/arctic_embed/contrastive_dataloader.py -------------------------------------------------------------------------------- /projects/arctic_embed/src/arctic_embed/core/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/arctic_embed/src/arctic_embed/core/__init__.py -------------------------------------------------------------------------------- /projects/arctic_embed/src/arctic_embed/core/biencoder_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/arctic_embed/src/arctic_embed/core/biencoder_model.py -------------------------------------------------------------------------------- /projects/arctic_embed/src/arctic_embed/core/cuda_allocator_config.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/arctic_embed/src/arctic_embed/core/cuda_allocator_config.py -------------------------------------------------------------------------------- /projects/arctic_embed/src/arctic_embed/core/losses.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/arctic_embed/src/arctic_embed/core/losses.py -------------------------------------------------------------------------------- /projects/arctic_embed/src/arctic_embed/core/pretokenized_batch_loader.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/arctic_embed/src/arctic_embed/core/pretokenized_batch_loader.py -------------------------------------------------------------------------------- /projects/arctic_embed/src/arctic_embed/core/slice_sparse_coo_tensor.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/arctic_embed/src/arctic_embed/core/slice_sparse_coo_tensor.py -------------------------------------------------------------------------------- /projects/arctic_embed/src/arctic_embed/data_processing/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/arctic_embed/src/arctic_embed/data_processing/__init__.py -------------------------------------------------------------------------------- /projects/arctic_embed/src/arctic_embed/data_processing/dense_retrieval.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/arctic_embed/src/arctic_embed/data_processing/dense_retrieval.py -------------------------------------------------------------------------------- /projects/arctic_embed/src/arctic_embed/data_processing/many_parquet_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/arctic_embed/src/arctic_embed/data_processing/many_parquet_dataset.py -------------------------------------------------------------------------------- /projects/arctic_embed/src/arctic_embed/data_processing/pretokenize.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/arctic_embed/src/arctic_embed/data_processing/pretokenize.py -------------------------------------------------------------------------------- /projects/arctic_embed/src/arctic_embed/data_processing/teacher_models.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/arctic_embed/src/arctic_embed/data_processing/teacher_models.py -------------------------------------------------------------------------------- /projects/arctic_embed/src/arctic_embed/data_processing/typing_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/arctic_embed/src/arctic_embed/data_processing/typing_utils.py -------------------------------------------------------------------------------- /projects/arctic_embed/src/arctic_embed/data_processing/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/arctic_embed/src/arctic_embed/data_processing/utils.py -------------------------------------------------------------------------------- /projects/arctic_embed/src/arctic_embed/trainer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/arctic_embed/src/arctic_embed/trainer.py -------------------------------------------------------------------------------- /projects/arctic_embed/tests/test_biencoder_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/arctic_embed/tests/test_biencoder_model.py -------------------------------------------------------------------------------- /projects/arctic_embed/tests/test_losses.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/arctic_embed/tests/test_losses.py -------------------------------------------------------------------------------- /projects/arctic_embed/tests/test_slice_sparse_coo_tensor.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/arctic_embed/tests/test_slice_sparse_coo_tensor.py -------------------------------------------------------------------------------- /projects/arctic_lstm_speculator/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/arctic_lstm_speculator/README.md -------------------------------------------------------------------------------- /projects/arctic_lstm_speculator/llama3.1-70b.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/arctic_lstm_speculator/llama3.1-70b.yaml -------------------------------------------------------------------------------- /projects/arctic_lstm_speculator/llama3.1-8b.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/arctic_lstm_speculator/llama3.1-8b.yaml -------------------------------------------------------------------------------- /projects/arctic_lstm_speculator/llama3.3-70b.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/arctic_lstm_speculator/llama3.3-70b.yaml -------------------------------------------------------------------------------- /projects/arctic_lstm_speculator/multigpu_runner.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/arctic_lstm_speculator/multigpu_runner.py -------------------------------------------------------------------------------- /projects/arctic_lstm_speculator/oss-120b.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/arctic_lstm_speculator/oss-120b.yaml -------------------------------------------------------------------------------- /projects/arctic_lstm_speculator/oss-20b.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/arctic_lstm_speculator/oss-20b.yaml -------------------------------------------------------------------------------- /projects/arctic_lstm_speculator/qwen2.5-32b.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/arctic_lstm_speculator/qwen2.5-32b.yaml -------------------------------------------------------------------------------- /projects/arctic_lstm_speculator/scripts/llama3.1-70b.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/arctic_lstm_speculator/scripts/llama3.1-70b.sh -------------------------------------------------------------------------------- /projects/arctic_lstm_speculator/scripts/llama3.1-8b.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/arctic_lstm_speculator/scripts/llama3.1-8b.sh -------------------------------------------------------------------------------- /projects/arctic_lstm_speculator/scripts/llama3.3-70b.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/arctic_lstm_speculator/scripts/llama3.3-70b.sh -------------------------------------------------------------------------------- /projects/arctic_lstm_speculator/scripts/oss-120b.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/arctic_lstm_speculator/scripts/oss-120b.sh -------------------------------------------------------------------------------- /projects/arctic_lstm_speculator/scripts/oss-20b.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/arctic_lstm_speculator/scripts/oss-20b.sh -------------------------------------------------------------------------------- /projects/arctic_lstm_speculator/scripts/qwen2.5-32b.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/arctic_lstm_speculator/scripts/qwen2.5-32b.sh -------------------------------------------------------------------------------- /projects/arctic_lstm_speculator/speculator/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/arctic_lstm_speculator/speculator/__init__.py -------------------------------------------------------------------------------- /projects/arctic_lstm_speculator/speculator/configs.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/arctic_lstm_speculator/speculator/configs.py -------------------------------------------------------------------------------- /projects/arctic_lstm_speculator/speculator/data_generation/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/arctic_lstm_speculator/speculator/data_generation/__init__.py -------------------------------------------------------------------------------- /projects/arctic_lstm_speculator/speculator/data_generation/concat_generated_datasets.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/arctic_lstm_speculator/speculator/data_generation/concat_generated_datasets.py -------------------------------------------------------------------------------- /projects/arctic_lstm_speculator/speculator/data_generation/data_gen_script_maker.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/arctic_lstm_speculator/speculator/data_generation/data_gen_script_maker.py -------------------------------------------------------------------------------- /projects/arctic_lstm_speculator/speculator/data_generation/vllm_data_generation.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/arctic_lstm_speculator/speculator/data_generation/vllm_data_generation.py -------------------------------------------------------------------------------- /projects/arctic_lstm_speculator/speculator/speculator.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/arctic_lstm_speculator/speculator/speculator.py -------------------------------------------------------------------------------- /projects/arctic_lstm_speculator/train.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/arctic_lstm_speculator/train.py -------------------------------------------------------------------------------- /projects/arctic_text2sql_r1/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/arctic_text2sql_r1/README.md -------------------------------------------------------------------------------- /projects/arctic_text2sql_r1/assets/leaderboard_plot_average_styled.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/arctic_text2sql_r1/assets/leaderboard_plot_average_styled.png -------------------------------------------------------------------------------- /projects/arctic_text2sql_r1/bird_eval/auto_evaluation.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/arctic_text2sql_r1/bird_eval/auto_evaluation.py -------------------------------------------------------------------------------- /projects/arctic_text2sql_r1/bird_eval/eval_open_source_models.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/arctic_text2sql_r1/bird_eval/eval_open_source_models.py -------------------------------------------------------------------------------- /projects/arctic_text2sql_r1/bird_eval/evaluate_bird.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/arctic_text2sql_r1/bird_eval/evaluate_bird.py -------------------------------------------------------------------------------- /projects/arctic_text2sql_r1/bird_eval/infer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/arctic_text2sql_r1/bird_eval/infer.py -------------------------------------------------------------------------------- /projects/arctic_text2sql_r1/bird_eval/self_consistency.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/arctic_text2sql_r1/bird_eval/self_consistency.py -------------------------------------------------------------------------------- /projects/arctic_text2sql_r1/data_preprocessing/build_contents_index.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/arctic_text2sql_r1/data_preprocessing/build_contents_index.py -------------------------------------------------------------------------------- /projects/arctic_text2sql_r1/data_preprocessing/nltk_downloader.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/arctic_text2sql_r1/data_preprocessing/nltk_downloader.py -------------------------------------------------------------------------------- /projects/arctic_text2sql_r1/data_preprocessing/process_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/arctic_text2sql_r1/data_preprocessing/process_dataset.py -------------------------------------------------------------------------------- /projects/arctic_text2sql_r1/data_preprocessing/process_dataset.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/arctic_text2sql_r1/data_preprocessing/process_dataset.sh -------------------------------------------------------------------------------- /projects/arctic_text2sql_r1/modelcard.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/arctic_text2sql_r1/modelcard.md -------------------------------------------------------------------------------- /projects/causal/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/causal/README.md -------------------------------------------------------------------------------- /projects/causal/run-causal.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/causal/run-causal.yml -------------------------------------------------------------------------------- /projects/excot_dpo/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/excot_dpo/README.md -------------------------------------------------------------------------------- /projects/excot_dpo/data_generation/configs/bird_config.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/excot_dpo/data_generation/configs/bird_config.yaml -------------------------------------------------------------------------------- /projects/excot_dpo/data_generation/configs/spider_config.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/excot_dpo/data_generation/configs/spider_config.yaml -------------------------------------------------------------------------------- /projects/excot_dpo/data_generation/data_generation.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/excot_dpo/data_generation/data_generation.py -------------------------------------------------------------------------------- /projects/excot_dpo/data_generation/dpo_sample.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/excot_dpo/data_generation/dpo_sample.py -------------------------------------------------------------------------------- /projects/excot_dpo/data_generation/local_verification.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/excot_dpo/data_generation/local_verification.py -------------------------------------------------------------------------------- /projects/excot_dpo/data_generation/prompts/divide_and_conquer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/excot_dpo/data_generation/prompts/divide_and_conquer.py -------------------------------------------------------------------------------- /projects/excot_dpo/data_generation/sft_sample.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/excot_dpo/data_generation/sft_sample.py -------------------------------------------------------------------------------- /projects/excot_dpo/dpo-llama-8b.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/excot_dpo/dpo-llama-8b.yaml -------------------------------------------------------------------------------- /projects/excot_dpo/environment.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/excot_dpo/environment.yaml -------------------------------------------------------------------------------- /projects/excot_dpo/evalution/configs/bird_config.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/excot_dpo/evalution/configs/bird_config.yaml -------------------------------------------------------------------------------- /projects/excot_dpo/evalution/configs/spider_config.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/excot_dpo/evalution/configs/spider_config.yaml -------------------------------------------------------------------------------- /projects/excot_dpo/evalution/eval_w_arctic_syth.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/excot_dpo/evalution/eval_w_arctic_syth.py -------------------------------------------------------------------------------- /projects/excot_dpo/evalution/execute_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/excot_dpo/evalution/execute_utils.py -------------------------------------------------------------------------------- /projects/excot_dpo/evalution/prompts/divide_and_conquer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/excot_dpo/evalution/prompts/divide_and_conquer.py -------------------------------------------------------------------------------- /projects/excot_dpo/evalution/sql_exec.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/excot_dpo/evalution/sql_exec.py -------------------------------------------------------------------------------- /projects/excot_dpo/requirements.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/excot_dpo/requirements.txt -------------------------------------------------------------------------------- /projects/excot_dpo/sft-llama-8b.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/excot_dpo/sft-llama-8b.yaml -------------------------------------------------------------------------------- /projects/excot_dpo/utils/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/excot_dpo/utils/__init__.py -------------------------------------------------------------------------------- /projects/excot_dpo/utils/execute_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/excot_dpo/utils/execute_utils.py -------------------------------------------------------------------------------- /projects/excot_dpo/utils/sql_exec.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/excot_dpo/utils/sql_exec.py -------------------------------------------------------------------------------- /projects/mlp_speculator/llama-8b.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/mlp_speculator/llama-8b.yaml -------------------------------------------------------------------------------- /projects/mlp_speculator/mlp_speculator/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/mlp_speculator/mlp_speculator/__init__.py -------------------------------------------------------------------------------- /projects/mlp_speculator/mlp_speculator/configs.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/mlp_speculator/mlp_speculator/configs.py -------------------------------------------------------------------------------- /projects/mlp_speculator/mlp_speculator/data_generation/vllm_multi_replica_gen_hf_datasets.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/mlp_speculator/mlp_speculator/data_generation/vllm_multi_replica_gen_hf_datasets.py -------------------------------------------------------------------------------- /projects/mlp_speculator/mlp_speculator/speculator.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/mlp_speculator/mlp_speculator/speculator.py -------------------------------------------------------------------------------- /projects/mlp_speculator/train.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/mlp_speculator/train.py -------------------------------------------------------------------------------- /projects/moba_attention/moba-swiftkv-llama-8b.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/moba_attention/moba-swiftkv-llama-8b.yaml -------------------------------------------------------------------------------- /projects/moba_attention/moba/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/moba_attention/moba/__init__.py -------------------------------------------------------------------------------- /projects/moba_attention/moba/config.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/moba_attention/moba/config.py -------------------------------------------------------------------------------- /projects/moba_attention/moba/llama.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/moba_attention/moba/llama.py -------------------------------------------------------------------------------- /projects/moba_attention/moba/moba_efficient.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/moba_attention/moba/moba_efficient.py -------------------------------------------------------------------------------- /projects/moba_attention/moba/moba_naive.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/moba_attention/moba/moba_naive.py -------------------------------------------------------------------------------- /projects/moba_attention/moba/moba_with_flash_interface.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/moba_attention/moba/moba_with_flash_interface.py -------------------------------------------------------------------------------- /projects/moba_attention/moba/wrapper.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/moba_attention/moba/wrapper.py -------------------------------------------------------------------------------- /projects/moba_attention/train.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/moba_attention/train.py -------------------------------------------------------------------------------- /projects/sequence-parallelism/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/sequence-parallelism/README.md -------------------------------------------------------------------------------- /projects/sequence-parallelism/USAGE.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/sequence-parallelism/USAGE.md -------------------------------------------------------------------------------- /projects/sequence-parallelism/images/correctness-plot.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/sequence-parallelism/images/correctness-plot.png -------------------------------------------------------------------------------- /projects/sequence-parallelism/images/loss-computation.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/sequence-parallelism/images/loss-computation.png -------------------------------------------------------------------------------- /projects/sequence-parallelism/run-h100-sp1-llama-8b-baseline.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/sequence-parallelism/run-h100-sp1-llama-8b-baseline.yml -------------------------------------------------------------------------------- /projects/sequence-parallelism/run-h100-sp1-llama-8b.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/sequence-parallelism/run-h100-sp1-llama-8b.yml -------------------------------------------------------------------------------- /projects/sequence-parallelism/run-h100-sp1-qwen3-32b.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/sequence-parallelism/run-h100-sp1-qwen3-32b.yml -------------------------------------------------------------------------------- /projects/sequence-parallelism/run-h100-sp16-llama-70b.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/sequence-parallelism/run-h100-sp16-llama-70b.yml -------------------------------------------------------------------------------- /projects/sequence-parallelism/run-h100-sp16-llama-8b.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/sequence-parallelism/run-h100-sp16-llama-8b.yml -------------------------------------------------------------------------------- /projects/sequence-parallelism/run-h100-sp16-qwen3-32b.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/sequence-parallelism/run-h100-sp16-qwen3-32b.yml -------------------------------------------------------------------------------- /projects/sequence-parallelism/run-h100-sp32-llama-70b.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/sequence-parallelism/run-h100-sp32-llama-70b.yml -------------------------------------------------------------------------------- /projects/sequence-parallelism/run-h100-sp32-llama-8b-baseline.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/sequence-parallelism/run-h100-sp32-llama-8b-baseline.yml -------------------------------------------------------------------------------- /projects/sequence-parallelism/run-h100-sp32-llama-8b.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/sequence-parallelism/run-h100-sp32-llama-8b.yml -------------------------------------------------------------------------------- /projects/sequence-parallelism/run-h100-sp32-qwen3-32b.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/sequence-parallelism/run-h100-sp32-qwen3-32b.yml -------------------------------------------------------------------------------- /projects/sequence-parallelism/run-h100-sp64-llama-70b.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/sequence-parallelism/run-h100-sp64-llama-70b.yml -------------------------------------------------------------------------------- /projects/sequence-parallelism/run-h100-sp64-qwen3-32b.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/sequence-parallelism/run-h100-sp64-qwen3-32b.yml -------------------------------------------------------------------------------- /projects/sequence-parallelism/run-h100-sp8-llama-70b.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/sequence-parallelism/run-h100-sp8-llama-70b.yml -------------------------------------------------------------------------------- /projects/sequence-parallelism/run-h100-sp8-llama-8b-no-mlp-no-act-offload.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/sequence-parallelism/run-h100-sp8-llama-8b-no-mlp-no-act-offload.yml -------------------------------------------------------------------------------- /projects/sequence-parallelism/run-h100-sp8-llama-8b-no-ulysses-no-liger-no-extras.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/sequence-parallelism/run-h100-sp8-llama-8b-no-ulysses-no-liger-no-extras.yml -------------------------------------------------------------------------------- /projects/sequence-parallelism/run-h100-sp8-llama-8b-no-ulysses-yes-liger-no-extras.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/sequence-parallelism/run-h100-sp8-llama-8b-no-ulysses-yes-liger-no-extras.yml -------------------------------------------------------------------------------- /projects/sequence-parallelism/run-h100-sp8-llama-8b-yes-act-offload.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/sequence-parallelism/run-h100-sp8-llama-8b-yes-act-offload.yml -------------------------------------------------------------------------------- /projects/sequence-parallelism/run-h100-sp8-llama-8b-yes-mlp.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/sequence-parallelism/run-h100-sp8-llama-8b-yes-mlp.yml -------------------------------------------------------------------------------- /projects/sequence-parallelism/run-h100-sp8-llama-8b.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/sequence-parallelism/run-h100-sp8-llama-8b.yml -------------------------------------------------------------------------------- /projects/sequence-parallelism/run-h100-sp8-qwen3-32b.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/sequence-parallelism/run-h100-sp8-qwen3-32b.yml -------------------------------------------------------------------------------- /projects/sequence-parallelism/run-h200-sp1-llama-8b-baseline.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/sequence-parallelism/run-h200-sp1-llama-8b-baseline.yml -------------------------------------------------------------------------------- /projects/sequence-parallelism/run-h200-sp1-llama-8b-liger-offload-tiled-mlp.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/sequence-parallelism/run-h200-sp1-llama-8b-liger-offload-tiled-mlp.yml -------------------------------------------------------------------------------- /projects/sequence-parallelism/run-h200-sp1-llama-8b-liger-offload.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/sequence-parallelism/run-h200-sp1-llama-8b-liger-offload.yml -------------------------------------------------------------------------------- /projects/sequence-parallelism/run-h200-sp1-llama-8b-liger.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/sequence-parallelism/run-h200-sp1-llama-8b-liger.yml -------------------------------------------------------------------------------- /projects/swiftkv/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/swiftkv/README.md -------------------------------------------------------------------------------- /projects/swiftkv/configs/deepseek-v2-swiftkv-lite-chat.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/swiftkv/configs/deepseek-v2-swiftkv-lite-chat.yaml -------------------------------------------------------------------------------- /projects/swiftkv/configs/llama-3.1-swiftkv-405b-instruct.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/swiftkv/configs/llama-3.1-swiftkv-405b-instruct.yaml -------------------------------------------------------------------------------- /projects/swiftkv/configs/llama-3.1-swiftkv-70b-instruct.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/swiftkv/configs/llama-3.1-swiftkv-70b-instruct.yaml -------------------------------------------------------------------------------- /projects/swiftkv/configs/llama-3.1-swiftkv-8b-instruct-long.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/swiftkv/configs/llama-3.1-swiftkv-8b-instruct-long.yaml -------------------------------------------------------------------------------- /projects/swiftkv/configs/llama-3.1-swiftkv-8b-instruct.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/swiftkv/configs/llama-3.1-swiftkv-8b-instruct.yaml -------------------------------------------------------------------------------- /projects/swiftkv/configs/llama-3.3-swiftkv-70b-instruct-long.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/swiftkv/configs/llama-3.3-swiftkv-70b-instruct-long.yaml -------------------------------------------------------------------------------- /projects/swiftkv/configs/qwen2.5-swiftkv-14b-instruct.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/swiftkv/configs/qwen2.5-swiftkv-14b-instruct.yaml -------------------------------------------------------------------------------- /projects/swiftkv/configs/qwen2.5-swiftkv-32b-instruct.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/swiftkv/configs/qwen2.5-swiftkv-32b-instruct.yaml -------------------------------------------------------------------------------- /projects/swiftkv/configs/qwen2.5-swiftkv-7b-instruct.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/swiftkv/configs/qwen2.5-swiftkv-7b-instruct.yaml -------------------------------------------------------------------------------- /projects/swiftkv/eval.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/swiftkv/eval.py -------------------------------------------------------------------------------- /projects/swiftkv/models/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/swiftkv/models/__init__.py -------------------------------------------------------------------------------- /projects/swiftkv/models/deepseek_v2/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/swiftkv/models/deepseek_v2/__init__.py -------------------------------------------------------------------------------- /projects/swiftkv/models/deepseek_v2/configuration_deepseek.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/swiftkv/models/deepseek_v2/configuration_deepseek.py -------------------------------------------------------------------------------- /projects/swiftkv/models/deepseek_v2/configuration_deepseek_swiftkv.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/swiftkv/models/deepseek_v2/configuration_deepseek_swiftkv.py -------------------------------------------------------------------------------- /projects/swiftkv/models/deepseek_v2/modeling_deepseek.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/swiftkv/models/deepseek_v2/modeling_deepseek.py -------------------------------------------------------------------------------- /projects/swiftkv/models/deepseek_v2/modeling_deepseek_swiftkv.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/swiftkv/models/deepseek_v2/modeling_deepseek_swiftkv.py -------------------------------------------------------------------------------- /projects/swiftkv/models/llama/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/swiftkv/models/llama/__init__.py -------------------------------------------------------------------------------- /projects/swiftkv/models/llama/configuration_llama_swiftkv.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/swiftkv/models/llama/configuration_llama_swiftkv.py -------------------------------------------------------------------------------- /projects/swiftkv/models/llama/modeling_llama_swiftkv.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/swiftkv/models/llama/modeling_llama_swiftkv.py -------------------------------------------------------------------------------- /projects/swiftkv/models/qwen2/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/swiftkv/models/qwen2/__init__.py -------------------------------------------------------------------------------- /projects/swiftkv/models/qwen2/configuration_qwen2_swiftkv.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/swiftkv/models/qwen2/configuration_qwen2_swiftkv.py -------------------------------------------------------------------------------- /projects/swiftkv/models/qwen2/modeling_qwen2_swiftkv.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/swiftkv/models/qwen2/modeling_qwen2_swiftkv.py -------------------------------------------------------------------------------- /projects/swiftkv/train.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/projects/swiftkv/train.py -------------------------------------------------------------------------------- /pyproject.toml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/pyproject.toml -------------------------------------------------------------------------------- /scripts/check_license.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/scripts/check_license.py -------------------------------------------------------------------------------- /scripts/download_data.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/scripts/download_data.py -------------------------------------------------------------------------------- /scripts/install_deps.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/scripts/install_deps.py -------------------------------------------------------------------------------- /scripts/release.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/scripts/release.sh -------------------------------------------------------------------------------- /scripts/upgrade_user_code.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/scripts/upgrade_user_code.py -------------------------------------------------------------------------------- /tests/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/tests/__init__.py -------------------------------------------------------------------------------- /tests/callback/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/tests/callback/__init__.py -------------------------------------------------------------------------------- /tests/callback/test_mixin.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/tests/callback/test_mixin.py -------------------------------------------------------------------------------- /tests/checkpoint/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/tests/checkpoint/__init__.py -------------------------------------------------------------------------------- /tests/checkpoint/test_ds_engine.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/tests/checkpoint/test_ds_engine.py -------------------------------------------------------------------------------- /tests/checkpoint/test_hf_engine.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/tests/checkpoint/test_hf_engine.py -------------------------------------------------------------------------------- /tests/checkpoint/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/tests/checkpoint/utils.py -------------------------------------------------------------------------------- /tests/config/test_config_custom_types.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/tests/config/test_config_custom_types.py -------------------------------------------------------------------------------- /tests/config/test_model_config.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/tests/config/test_model_config.py -------------------------------------------------------------------------------- /tests/config/test_trainer_config.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/tests/config/test_trainer_config.py -------------------------------------------------------------------------------- /tests/conftest.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/tests/conftest.py -------------------------------------------------------------------------------- /tests/data/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/tests/data/__init__.py -------------------------------------------------------------------------------- /tests/data/test_factory.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/tests/data/test_factory.py -------------------------------------------------------------------------------- /tests/data/test_hf_instruct_source.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/tests/data/test_hf_instruct_source.py -------------------------------------------------------------------------------- /tests/data/test_hf_source.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/tests/data/test_hf_source.py -------------------------------------------------------------------------------- /tests/data/test_sft_factory.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/tests/data/test_sft_factory.py -------------------------------------------------------------------------------- /tests/data/test_source.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/tests/data/test_source.py -------------------------------------------------------------------------------- /tests/data/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/tests/data/utils.py -------------------------------------------------------------------------------- /tests/helpers.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/tests/helpers.py -------------------------------------------------------------------------------- /tests/model/test_hf_factory.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/tests/model/test_hf_factory.py -------------------------------------------------------------------------------- /tests/pytest.ini: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/tests/pytest.ini -------------------------------------------------------------------------------- /tests/test_registry.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/tests/test_registry.py -------------------------------------------------------------------------------- /tests/trainer/test_causal_trainer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/tests/trainer/test_causal_trainer.py -------------------------------------------------------------------------------- /tests/trainer/test_dpo_trainer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/tests/trainer/test_dpo_trainer.py -------------------------------------------------------------------------------- /tests/trainer/test_sft_trainer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/tests/trainer/test_sft_trainer.py -------------------------------------------------------------------------------- /tests/trainer/test_ulysses_alst.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/tests/trainer/test_ulysses_alst.py -------------------------------------------------------------------------------- /tests/trainer/test_wandb.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/tests/trainer/test_wandb.py -------------------------------------------------------------------------------- /tests/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/snowflakedb/ArcticTraining/HEAD/tests/utils.py --------------------------------------------------------------------------------