Hamid Shojanazeri d51d2cce9c adding sdpa for flash attn 1 year ago
..
__init__.py 207d2f80e9 Make code-llama and hf-tgi inference runnable as module 1 year ago
chat_utils.py e554c1c8bf The tokenizer will not add eos_token by default 1 year ago
checkpoint_converter_fsdp_hf.py ce9501f22c remove relative imports 1 year ago
model_utils.py d51d2cce9c adding sdpa for flash attn 1 year ago
prompt_format_utils.py 3e710f71f8 renaming the prompt format file to conform to repo standards 1 year ago
safety_utils.py c0886a0a89 Fixing typo in self 1 year ago