---
license: apache-2.0
---
# LeTI: Learning to Generate from Textual Interactions

Official model checkpoint for paper [LeTI: Learning to Generate from Textual Interactions](https://arxiv.org/abs/2305.10314).

The code associated with these checkpoints can be found [here](https://github.com/xingyaoww/LeTI).

Decompress the model checkpoint [`data.tar.gz`](https://huggingface.co./xingyaoww/LeTI/resolve/main/data.tar.gz) (using `tar -xzvf data.tar.gz`) will create a folder `data`.
After decompress the checkpoints, you can set `export GS_BUCKET_PREFIX=`pwd` ` and run evaluation of your choice following [this](https://github.com/xingyaoww/LeTI/blob/main/docs/EVAL.md).

We currently release the following checkpoints in the paper:

- **LeTI (350M)**: `mbpp-ft/actor-rw-conditioned/codegen-350M-mono/350M+rw_conditioned+mixpretrain+50x3+lr1e-5/checkpoint_31416`
- **LeTI (2B)**: `mbpp-ft/actor-rw-conditioned/codegen-2B-mono/2B+rw_conditioned+mixpretrain+10x3+lr5e-6/checkpoint_13465`