license: apache-2.0 | |
- wandb: https://wandb.ai/open-assistant/rlhf/runs/csz0ickm | |
- base-model: [andreaskoepf/oasst-sft-4-pythia-12b-epoch-3.5](https://huggingface.co/andreaskoepf/oasst-sft-4-pythia-12b-epoch-3.5) | |
- reward-model: [andreaskoepf/oasst-rm-2-pythia-1.4b-10000](https://huggingface.co/andreaskoepf/oasst-rm-2-pythia-1.4b-10000) | |
- checkpoint: 2000 steps | |