huggingface/text-generation-inference
5
0
Fork 0
mirror of https://github.com/huggingface/text-generation-inference.git synced 2025-04-26 12:32:10 +00:00
Code Issues Packages Projects Releases Wiki Activity

Enables Flash Attention in TGI for gemma models (#235)

Browse Source
This commit is contained in:
Thanaji Rao Thakkalapelli 2024-10-18 09:20:42 -07:00 committed by GitHub
parent 9ae5ad5057
commit c5e3881051
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 4 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
server/text_generation_server/models/causal_lm.py
View File

@ -694,10 +694,12 @@ class CausalLM(Model):
"return_dict": True,
}
if model.config.model_type in ["llama", "mistral", "starcoder2", "qwen2", "falcon"]:
if model.config.model_type in ["llama", "mistral", "starcoder2", "qwen2", "falcon", "gemma"]:
if model.config.model_type not in ["falcon"]:
kwargs["attn_softmax_bf16"] = True
if model.config.model_type not in ["gemma"]:
kwargs["trim_logits"] = True
if os.getenv("USE_FLASH_ATTENTION", "false").lower() == "true":