更新0.12
Showing
megatron/core/transformer/attention.py
100755 → 100644
This diff is collapsed.
megatron/core/transformer/cuda_graphs.py
100755 → 100644
This diff is collapsed.
File mode changed from 100755 to 100644
File mode changed from 100755 to 100644
File mode changed from 100755 to 100644
megatron/core/transformer/enums.py
100755 → 100644
File mode changed from 100755 to 100644
megatron/core/transformer/identity_op.py
100755 → 100644
File mode changed from 100755 to 100644
megatron/core/transformer/mlp.py
100755 → 100644
This diff is collapsed.
megatron/core/transformer/module.py
100755 → 100644
File mode changed from 100755 to 100644
megatron/core/transformer/moe/README.md
100755 → 100644
This diff is collapsed.
megatron/core/transformer/moe/__init__.py
100755 → 100644
File mode changed from 100755 to 100644
megatron/core/transformer/moe/experts.py
100755 → 100644
This diff is collapsed.
This diff is collapsed.
File mode changed from 100755 to 100644
This diff is collapsed.
megatron/core/transformer/moe/moe_layer.py
100755 → 100644
This diff is collapsed.
megatron/core/transformer/moe/moe_utils.py
100755 → 100644
This diff is collapsed.
megatron/core/transformer/moe/router.py
100755 → 100644
This diff is collapsed.
megatron/core/transformer/moe/shared_experts.py
100755 → 100644
This diff is collapsed.
This diff is collapsed.
Please register or sign in to comment