github/ColossalAI
1
0
Fork 0
mirror of https://github.com/hpcaitech/ColossalAI.git synced 2025-09-07 03:52:01 +00:00
Code Issues Packages Projects Releases Wiki Activity

[devops] remove post commit ci (#5566)

Browse Source 
* [devops] remove post commit ci

* [misc] run pre-commit on all files

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

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

---------

Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
This commit is contained in:
Hongxin Liu
2024-04-08 15:09:40 +08:00
committed by GitHub
parent 341263df48
commit 641b1ee71a
82 changed files with 849 additions and 962 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
colossalai/legacy/inference/tensor_parallel/modeling/llama.py
View File

@@ -8,7 +8,9 @@ from transformers.models.llama.modeling_llama import LlamaAttention, LlamaDecode
from colossalai.inference.tensor_parallel.batch_infer_state import BatchInferState
from colossalai.kernel.triton import llama_context_attn_fwd, token_attention_fwd
from colossalai.kernel.triton.token_attention_kernel import Llama2TokenAttentionForwards
from ._utils import copy_kv_to_mem_cache
try:
from lightllm.models.llama.triton_kernel.context_flashattention_nopad import (
context_attention_fwd as lightllm_llama_context_attention_fwd,
@@ -90,7 +92,7 @@ def llama_triton_token_attention(query_states, attn_output, infer_state, num_key
# infer_state.cache_manager.past_key_values_length,
infer_state.max_len_in_batch,
)
else:
Llama2TokenAttentionForwards.token_attn(
query_states,