Commit 7835249

authored

Merge branch 'AI-Hypercomputer:main' into main

2 parents d1a7c12 + 8e9f24a commit 7835249Copy full SHA for 7835249

225 files changed

.github/workflows
PREFLIGHT.md
README.md
RESTRUCTURE.md
benchmarks/api_server
- port_forward_xpk.sh
codecov.yml
dependencies
docs
- _static
  - css
    - custom.css
  - moe_strategy.png
- conf.py
- development.md
- explanations.md
- guides.md
- guides
  - checkpointing_solutions.md
  - checkpointing_solutions
    - multi_tier_checkpointing.md
  - data_input_pipeline.md
  - data_input_pipeline
  - install_maxtext.md
  - monitoring_and_debugging.md
  - monitoring_and_debugging
  - optimization.md
  - optimization
  - run_maxtext.md
  - run_python_notebook.md
- index.md
- install_maxtext.md
- reference.md
- reference
- release_notes.md
- run_maxtext.md
- run_maxtext
- tutorials.md
- tutorials
end_to_end/tpu
- deepseek
  - Run_DeepSeek.md
  - v2-16b
    - test_deepseek.sh
  - v3-671b
- gemma3
  - Run_Gemma3.md
- gemma
  - Run_Gemma.md
- gpt_oss/20b
  - test_gpt_oss.sh
- llama4
- mixtral/8x7b
  - 1_test_mixtral.sh
  - 2_test_mixtral.sh
- qwen/next
  - qwen3-next-80b-a3b
    - 1_test_qwen3_next_80b_a3b.sh
  - run_qwen3_next.md
preflight.sh
src/MaxText
- accelerator_to_spec_map.py
- checkpointing.py
- common_types.py
- configs
  - a3/llama_3.1_405b
    - 128vm.sh
  - base.yml
  - gpu_smoke_test.yml
  - models
  - rl.yml
  - types.py
  - vllm.yml
- elastic_train.py
- examples
  - rl_llama3_demo.ipynb
  - sft_llama3_demo.ipynb
  - sft_train_and_evaluate.py
- experimental
  - agent/ckpt_conversion_agent
    - README.md
  - rl
    - grpo_trainer.py
- inference
  - paged_attention.py
- input_pipeline
  - _grain_data_processing.py
  - _hf_data_processing.py
  - _input_pipeline_utils.py
- integration
  - tunix
    - tunix_adapter.py
    - utils.py
  - vllm/maxtext_vllm_adapter
    - adapter.py
- kernels
  - jax_flash_attention.py
- layers
  - attention_mla.py
  - attention_op.py
  - attentions.py
  - decoders.py
  - deepseek.py
  - deepseek_batchsplit.py
  - embeddings.py
  - encoders.py
  - gpt3.py
  - gpt_oss.py
  - llama2.py
  - llama4.py
  - mixtral.py
  - models.py
  - moe.py
  - multi_token_prediction.py
  - normalizations.py
  - pipeline.py
  - qwen3.py
  - simple_layer.py
- layerwise_quantization.py
- max_logging.py
- max_utils.py
- maxtext_utils.py
- metric_logger.py
- model_creation_utils.py
- multihost_dataloading.py
- muon_utils.py
- optimizers.py
- pyconfig.py
- pyconfig_deprecated.py
- rl
  - evaluate_rl.py
  - train_rl.py
- sft_trainer.py
- sft
  - sft_trainer.py
- sharding.py
- train.py
- train_compile.py
- train_utils.py
- utils
  - ckpt_conversion
  - ckpt_scripts
- vllm_decode.py
- vocabulary_tiling.py
tests
tools/setup
- setup.sh
- setup_post_training_requirements.sh

`‎.github/workflows/RunTests.yml‎`

This file was deleted.

Comments

(0)