github/ColossalAI
1
0
Fork 0
mirror of https://github.com/hpcaitech/ColossalAI.git synced 2025-09-10 21:40:02 +00:00
Code Issues Packages Projects Releases Wiki Activity

[shardformer, pipeline] add gradient_checkpointing_ratio and heterogenous shard policy for llama (#5508)

Browse Source 
* feat: add `GradientCheckpointConfig` and `PipelineGradientCheckpointConfig`

* feat: apply `GradientCheckpointConfig` to policy and llama_forward

* feat: move `distribute_layer` and `get_stage_index` to PipelineStageManager

* fix: add optional args for `distribute_layer` and `get_stage_index`

* fix: fix changed API calls

* test: update llama tests

* style: polish `GradientCheckpointConfig`

* fix: fix pipeline utils tests
This commit is contained in:
Wenhao Chen
2024-04-01 11:34:58 +08:00
committed by GitHub
parent df5e9c53cf
commit e614aa34f3
28 changed files with 396 additions and 213 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
tests/kit/model_zoo/transformers/llama.py
View File

@@ -49,9 +49,9 @@ if HAS_LLAMA:
loss_fn_for_seq_classification = lambda output: output["logits"].mean()
config = LlamaConfig(
num_hidden_layers=4,
hidden_size=128,
intermediate_size=256,
num_hidden_layers=8,
hidden_size=32,
intermediate_size=64,
num_attention_heads=4,
max_position_embeddings=128,
num_labels=16,