OpenAssistant

- base model: OpenAssistant/pythia-12b-pre-v8-12.5k-steps - wandb: https://wandb.ai/open-assistant/supervised-finetuning/runs/pcw1ejda - sampling report

This model is a fine-tuning of TII's Falcon 40B LLM. It was trained with top-1 (high-quality) demonstrations of the OASST data set (exported on May 6, 2023) with an effective batch size of 144 for ~7.5 epochs with LIMA style dropout (p=0.3) and a context-length of 2048 tokens. - Finetuned from: tiiuae/falcon-40b - Model type: Causal decoder-only transformer language model - Language: English, German, Spanish, French (and limited capabilities in Italian, Portuguese, Polish, Dutch, Romanian, Czech, Swedish); - Demo: Continuations for 250 random prompts - Eval results: ilm-eval - Weights & Biases: Training log (Checkpoint: 560 steps) - License: Apache 2.0 - Contact: Open-Assistant Discord Two special tokens are used to mark the beginning of user and assistant turns: ` ` and ` `. Each turn ends with a ` ` token. The input ends with the ` ` token to signal that the model should start generating the assistant reply.

Due to the license attached to LLaMA models by Meta AI it is not possible to directly distribute LLaMA-based models. Instead we provide XOR weights for the OA models. Thanks to Mick for writing the `xorcodec.py` script which enables this process Note: This process applies to `oasst-sft-6-llama-30b` model. The same process can be applied to other models in future, but the checksums will be different.. This process is tested only on Linux (specifically Ubuntu). Some users have reported that the process does not work on Windows. We recommend using WSL if you only have a Windows machine. To use OpenAssistant LLaMA-Based Models, you should have a copy of the original LLaMA model weights and add them to a `llama` subdirectory here. If you cannot obtain the original LLaMA, see the note in italic below for a possible alternative. Ensure your LLaMA 30B checkpoint matches the correct md5sums: If you do not have a copy of the original LLaMA weights and cannot obtain one, you may still be able to complete this process. Some users have reported that this model can be used as a base for the XOR conversion. This will also allow you to skip to Step 7. However, we only support conversion starting from LLaMA original checkpoint and cannot provide support if you experience issues with this alternative approach. Important: Follow these exact steps to convert your original LLaMA checkpoint to a HuggingFace Transformers-compatible format. If you use the wrong versions of any dependency, you risk ending up with weights which are not compatible with the XOR files. 1. Create a clean Python 3.10 virtual environment & activate it: 2. Clone transformers repo and switch to tested version: 5. While in `transformers` repo root, run HF LLaMA conversion script: 6. Run `find . -type f -exec md5sum "{}" +` in the conversion target directory (`outputdir`). This should produce exactly the following checksums if your files are correct: Important: You should now have the correct LLaMA weights and be ready to apply the XORs. If the checksums above do not match yours, there is a problem. 7. Once you have LLaMA weights in the correct format, you can apply the XOR decoding: You should expect to see one warning message during execution: This is normal. If similar messages appear for other files, something has gone wrong. 8. Now run `find . -type f -exec md5sum "{}" +` in the output directory (here `oasst-sft-6-llama-30b`). You should get a file with exactly these checksums: If so you have successfully decoded the weights and should be able to use the model with HuggingFace Transformers. If your checksums do not match those above, there is a problem. - OASST dataset paper: https://arxiv.org/abs/2304.07327