20 lines
833 B
Markdown
20 lines
833 B
Markdown
|
---
|
||
|
|
language:
|
||
|
|
- en
|
||
|
|
tags:
|
||
|
|
- pytorch
|
||
|
|
- causal-lm
|
||
|
|
- pythia
|
||
|
|
license: apache-2.0
|
||
|
|
datasets:
|
||
|
|
- Anthropic/hh-rlhf
|
||
|
|
---
|
||
|
|
[Pythia-6.9b](https://huggingface.co/EleutherAI/pythia-6.9b) supervised finetuned with [Anthropic-hh-rlhf dataset](https://huggingface.co/datasets/Anthropic/hh-rlhf) for 1 epoch [(sft-model)](https://huggingface.co/lomahony/eleuther-pythia6.9b-hh-sft),
|
||
|
|
before DPO [(paper)](https://arxiv.org/abs/2305.18290) with same dataset for 1 epoch.
|
||
|
|
|
||
|
|
[wandb log](https://wandb.ai/pythia_dpo/Pythia_LOM/runs/i53fbcih?workspace=user-lauraomahony)
|
||
|
|
|
||
|
|
Benchmark evaluations included in repo done using [lm-evaluation-harness](https://github.com/EleutherAI/lm-evaluation-harness/tree/big-refactor).
|
||
|
|
|
||
|
|
See [Pythia-6.9b](https://huggingface.co/EleutherAI/pythia-6.9b) for original model details [(paper)](https://arxiv.org/abs/2101.00027).
|