fix flash attention import

small update

update for new version

fix eos & update README for tech report

fix `eos_token` (#10)

fix no white space when using stream_chat with fast tokenizer

update model weights

fix no white space when using stream_chat with fast tokenizer

Add special tokens for chat

small

fast tokenizer and stream_chat fix (#7)

update rope_scaling

remove unnecessary attention_drop

Update special tokens (#3)

Update license in README.md

fix errir links

Update README.md

update readme

Update README.md

fix import error

support flash attn 2

Create README.md

fix: add eoa into eos_token_id in chat to accelerate chat interface

use bin instead of safetensors with max shard of 2GB

fix(modeling): fix inference code

initial commit internlm2-chat-20b model

update gitattributes

initial commit