1da07e85aa
# What does this PR do? This adds a non flash version of MPT. Flash is harder because we need to create a bias ready cuda kernel of flash attention. Fixes https://github.com/huggingface/text-generation-inference/issues/361 Fixes https://github.com/huggingface/text-generation-inference/issues/491 Fixes https://github.com/huggingface/text-generation-inference/issues/290 |
||
|---|---|---|
| .. | ||
| models | ||
| conftest.py | ||
| pytest.ini | ||
| requirements.txt | ||