"vscode:/vscode.git/clone" did not exist on "dbb9a1aba9712ec93adcc37023b7a9f0730340a5"
-
Nicolas Patry authored
# 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
1da07e85
