2 년 전 · 6e9663ce87
--- a/ai/Megatron/English/Python/Start_Here.ipynb
+++ b/ai/Megatron/English/Python/Start_Here.ipynb
--- a/ai/Megatron/English/Python/jupyter_notebook/Lab1-2_EstimateComputeDaysNeeded.ipynb
+++ b/ai/Megatron/English/Python/jupyter_notebook/Lab1-2_EstimateComputeDaysNeeded.ipynb
--- a/ai/Megatron/English/Python/jupyter_notebook/Lab1-3_MegatronFundementals.ipynb
+++ b/ai/Megatron/English/Python/jupyter_notebook/Lab1-3_MegatronFundementals.ipynb
--- a/ai/Megatron/English/Python/jupyter_notebook/Lab1-4_GPT_vocab_merge_files.ipynb
+++ b/ai/Megatron/English/Python/jupyter_notebook/Lab1-4_GPT_vocab_merge_files.ipynb
--- a/ai/Megatron/English/Python/jupyter_notebook/Lab1-5_jsonfy_and_process2mmap.ipynb
+++ b/ai/Megatron/English/Python/jupyter_notebook/Lab1-5_jsonfy_and_process2mmap.ipynb
--- a/ai/Megatron/English/Python/jupyter_notebook/Lab1-6_Observe_GPT_runs_vs_performance.ipynb
+++ b/ai/Megatron/English/Python/jupyter_notebook/Lab1-6_Observe_GPT_runs_vs_performance.ipynb
--- a/ai/Megatron/English/Python/jupyter_notebook/Lab2-3_train_own_GPT2BPETokenizer.ipynb
+++ b/ai/Megatron/English/Python/jupyter_notebook/Lab2-3_train_own_GPT2BPETokenizer.ipynb
--- a/ai/Megatron/English/Python/jupyter_notebook/Lab2-4_customize_process2mmap.ipynb
+++ b/ai/Megatron/English/Python/jupyter_notebook/Lab2-4_customize_process2mmap.ipynb
--- a/ai/Megatron/English/Python/jupyter_notebook/Lab2-5_run_Megatron_with_varying_config.ipynb
+++ b/ai/Megatron/English/Python/jupyter_notebook/Lab2-5_run_Megatron_with_varying_config.ipynb
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/Dlprof_pretrain_gpt.py
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/Dlprof_pretrain_gpt.py
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/LICENSE
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/LICENSE
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/MANIFEST.in
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/MANIFEST.in
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/README.md
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/README.md
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/SV_preprocess_gpt.sh
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/SV_preprocess_gpt.sh
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/examples/evaluate_retriever_nq.sh
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/examples/evaluate_retriever_nq.sh
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/examples/pretrain_t5.sh
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/examples/pretrain_t5.sh
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/examples/pretrain_t5_distributed.sh
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/examples/pretrain_t5_distributed.sh
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/examples/pretrain_t5_distributed_with_mp.sh
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/examples/pretrain_t5_distributed_with_mp.sh
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/images/cases_april2021.png
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/images/cases_april2021.png
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/__init__.py
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/__init__.py
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/__pycache__/__init__.cpython-38.pyc
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/__pycache__/__init__.cpython-38.pyc
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/__pycache__/arguments.cpython-38.pyc
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/__pycache__/arguments.cpython-38.pyc
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/__pycache__/checkpointing.cpython-38.pyc
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/__pycache__/checkpointing.cpython-38.pyc
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/__pycache__/global_vars.cpython-38.pyc
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/__pycache__/global_vars.cpython-38.pyc
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/__pycache__/initialize.cpython-38.pyc
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/__pycache__/initialize.cpython-38.pyc
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/__pycache__/learning_rates.cpython-38.pyc
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/__pycache__/learning_rates.cpython-38.pyc
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/__pycache__/memory.cpython-38.pyc
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/__pycache__/memory.cpython-38.pyc
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/__pycache__/microbatches.cpython-38.pyc
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/__pycache__/microbatches.cpython-38.pyc
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/__pycache__/p2p_communication.cpython-38.pyc
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/__pycache__/p2p_communication.cpython-38.pyc
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/__pycache__/package_info.cpython-38.pyc
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/__pycache__/package_info.cpython-38.pyc
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/__pycache__/schedules.cpython-38.pyc
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/__pycache__/schedules.cpython-38.pyc
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/__pycache__/training.cpython-38.pyc
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/__pycache__/training.cpython-38.pyc
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/__pycache__/utils.cpython-38.pyc
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/__pycache__/utils.cpython-38.pyc
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/arguments.py
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/arguments.py
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/checkpointing.py
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/checkpointing.py
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/data/Makefile
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/data/Makefile
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/data/__init__.py
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/data/__init__.py
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/data/__pycache__/__init__.cpython-38.pyc
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/data/__pycache__/__init__.cpython-38.pyc
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/data/__pycache__/blendable_dataset.cpython-38.pyc
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/data/__pycache__/blendable_dataset.cpython-38.pyc
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/data/__pycache__/data_samplers.cpython-38.pyc
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/data/__pycache__/data_samplers.cpython-38.pyc
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/data/__pycache__/dataset_utils.cpython-38.pyc
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/data/__pycache__/dataset_utils.cpython-38.pyc
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/data/__pycache__/gpt_dataset.cpython-38.pyc
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/data/__pycache__/gpt_dataset.cpython-38.pyc
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/data/__pycache__/indexed_dataset.cpython-38.pyc
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/data/__pycache__/indexed_dataset.cpython-38.pyc
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/data/autoaugment.py
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/data/autoaugment.py
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/data/bert_dataset.py
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/data/bert_dataset.py
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/data/biencoder_dataset_utils.py
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/data/biencoder_dataset_utils.py
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/data/blendable_dataset.py
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/data/blendable_dataset.py
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/data/data_samplers.py
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/data/data_samplers.py
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/data/dataset_utils.py
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/data/dataset_utils.py
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/data/gpt_dataset.py
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/data/gpt_dataset.py
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/data/helpers.cpp
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/data/helpers.cpp
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/data/ict_dataset.py
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/data/ict_dataset.py
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/data/indexed_dataset.py
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/data/indexed_dataset.py
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/data/orqa_wiki_dataset.py
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/data/orqa_wiki_dataset.py
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/data/realm_dataset_utils.py
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/data/realm_dataset_utils.py
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/data/realm_index.py
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/data/realm_index.py
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/data/t5_dataset.py
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/data/t5_dataset.py
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/data/test/test_indexed_dataset.py
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/data/test/test_indexed_dataset.py
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/data/vit_dataset.py
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/data/vit_dataset.py
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/fused_kernels/__init__.py
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/fused_kernels/__init__.py
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/fused_kernels/__pycache__/__init__.cpython-38.pyc
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/fused_kernels/__pycache__/__init__.cpython-38.pyc
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/fused_kernels/build/.ninja_deps
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/fused_kernels/build/.ninja_deps
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/fused_kernels/build/.ninja_log
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/fused_kernels/build/.ninja_log
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/fused_kernels/build/build.ninja
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/fused_kernels/build/build.ninja
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/fused_kernels/compat.h
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/fused_kernels/compat.h
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/fused_kernels/layer_norm_cuda.cpp
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/fused_kernels/layer_norm_cuda.cpp
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/fused_kernels/layer_norm_cuda_kernel.cu
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/fused_kernels/layer_norm_cuda_kernel.cu
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/fused_kernels/scaled_masked_softmax.cpp
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/fused_kernels/scaled_masked_softmax.cpp
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/fused_kernels/scaled_masked_softmax.h
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/fused_kernels/scaled_masked_softmax.h
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/fused_kernels/scaled_masked_softmax_cuda.cu
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/fused_kernels/scaled_masked_softmax_cuda.cu
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/fused_kernels/scaled_upper_triang_masked_softmax.cpp
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/fused_kernels/scaled_upper_triang_masked_softmax.cpp
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/fused_kernels/scaled_upper_triang_masked_softmax.h
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/fused_kernels/scaled_upper_triang_masked_softmax.h
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/fused_kernels/scaled_upper_triang_masked_softmax_cuda.cu
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/fused_kernels/scaled_upper_triang_masked_softmax_cuda.cu
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/fused_kernels/type_shim.h
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/fused_kernels/type_shim.h
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/global_vars.py
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/global_vars.py
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/indexer.py
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/indexer.py
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/initialize.py
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/initialize.py
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/learning_rates.py
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/learning_rates.py
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/memory.py
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/memory.py
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/microbatches.py
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/microbatches.py
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/model/__init__.py
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/model/__init__.py
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/model/__pycache__/__init__.cpython-38.pyc
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/model/__pycache__/__init__.cpython-38.pyc
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/model/__pycache__/bert_model.cpython-38.pyc
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/model/__pycache__/bert_model.cpython-38.pyc
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/model/__pycache__/distributed.cpython-38.pyc
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/model/__pycache__/distributed.cpython-38.pyc
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/model/__pycache__/enums.cpython-38.pyc
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/model/__pycache__/enums.cpython-38.pyc
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/model/__pycache__/fused_bias_gelu.cpython-38.pyc
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/model/__pycache__/fused_bias_gelu.cpython-38.pyc
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/model/__pycache__/fused_layer_norm.cpython-38.pyc
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/model/__pycache__/fused_layer_norm.cpython-38.pyc
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/model/__pycache__/fused_softmax.cpython-38.pyc
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/model/__pycache__/fused_softmax.cpython-38.pyc
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/model/__pycache__/gpt_model.cpython-38.pyc
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/model/__pycache__/gpt_model.cpython-38.pyc
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/model/__pycache__/language_model.cpython-38.pyc
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/model/__pycache__/language_model.cpython-38.pyc
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/model/__pycache__/module.cpython-38.pyc
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/model/__pycache__/module.cpython-38.pyc
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/model/__pycache__/t5_model.cpython-38.pyc
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/model/__pycache__/t5_model.cpython-38.pyc
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/model/__pycache__/transformer.cpython-38.pyc
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/model/__pycache__/transformer.cpython-38.pyc
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/model/__pycache__/utils.cpython-38.pyc
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/model/__pycache__/utils.cpython-38.pyc
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/model/bert_model.py
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/model/bert_model.py
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/model/biencoder_model.py
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/model/biencoder_model.py
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/model/classification.py
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/model/classification.py
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/model/distributed.py
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/model/distributed.py
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/model/enums.py
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/model/enums.py
--- a/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/model/fused_bias_gelu.py
+++ b/ai/Megatron/English/Python/jupyter_notebook/Megatron-LM/megatron/model/fused_bias_gelu.py