Remove AutoTokenizer from the config

Add bert-base-cased tokenizer with model_max_length: 8192

Update config.json

Update README.md

Model

Merge branch 'main' of https://huggingface.co/togethercomputer/m2-bert-80M-8k-retrieval into main

Rename

Update README.md

Update README.md

Update README.md

Automodel support

Update 8K model

Update model weights

Remove flashfft weights

Update model weights

8k retrieval

8k retrieval model

initial commit