Fix RuntimeError: pad attn scores back to original query sequence length, instead of unpadded sequence length (i.e. no change).

Correct the output dtype of rmsnorm_func (#13)

remove torch.jit

Update README.md

Update tokenizer_config.json

Update README.md

Update README.md

Update README.md

Update README.md

fix missing inv freq

Update README.md

Update README.md

Update README.md

Update README.md

Update README.md

Update README.md

Merge branch 'main' of https://huggingface.co/togethercomputer/Llama-2-7B-32KCtx-v0.1

Update README.md

init

initial commit