[Feature] Zigzag Ring attention (#5905)

* halfway

* fix cross-PP-stage position id length diff bug

* fix typo

* fix typo

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* unified cross entropy func for all shardformer models

* remove redundant lines

* add basic ring attn; debug cross entropy

* fwd bwd logic complete

* fwd bwd logic complete; add experimental triton rescale

* precision tests passed

* precision tests passed

* fix typos and remove misc files

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* add sp_mode to benchmark; fix varlen interface

* update softmax_lse shape by new interface

* change tester name

* remove buffer clone; support packed seq layout

* add varlen tests

* fix typo

* all tests passed

* add dkv_group; fix mask

* remove debug statements

---------

Co-authored-by: Edenzzzz <wtan45@wisc.edu>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
This commit is contained in:
Edenzzzz
2024-08-16 13:56:38 +08:00
committed by GitHub
parent 887d2d579b
commit f5c84af0b0
50 changed files with 1870 additions and 326 deletions

View File

@@ -18,7 +18,7 @@ from tests.kit.model_zoo import model_zoo
@clear_cache_before_run()
@parameterize("model_name", ["transformers_llama_for_casual_lm"])
@parameterize("model_name", ["transformers_llama_for_causal_lm"])
@parameterize("plugin_type", ["ddp", "zero", "gemini"])
def exam_from_pretrained(plugin_type: str, model_name: str, shard=True, size_per_shard=32):
(model_fn, data_gen_fn, output_transform_fn, loss_fn, _) = next(