| --- |
| license: apache-2.0 |
| --- |
| # Pythia 1.4B Based Reward Model |
|
|
| - base model: [andreaskoepf/pythia-1.4b-gpt4all-pretrain](https://huggingface.co/andreaskoepf/pythia-1.4b-gpt4all-pretrain) |
| - wandb: https://wandb.ai/open-assistant/reward-model/runs/kadgqj65 |
| - checkpoint: 10k steps |
|
|
|
|
| Compute was generously provided by [Stability AI](https://stability.ai/) |
|
|
|
|
| ### How to use |
|
|
| ```python |
| # install open assistant model_training module (e.g. run `pip install -e .` in `model/` directory of open-assistant repository) |
| import model_training.models.reward_model # noqa: F401 (registers reward model for AutoModel loading) |
| |
| tokenizer = AutoTokenizer.from_pretrained(model_name) |
| model = AutoModelForSequenceClassification.from_pretrained(model_name) |
| input_text = "<|prompter|>Hi how are you?<|endoftext|><|assistant|>Hi, I am Open-Assistant a large open-source language model trained by LAION AI. How can I help you today?<|endoftext|>" |
| inputs = tokenizer(input_text, return_tensors="pt") |
| score = rm(**inputs).logits[0].cpu().detach() |
| print(score) |
| ``` |
|
|
| ### Datasets |
|
|
| ``` |
| datasets: |
| - oasst_export: |
| lang: "en,es,de,fr" |
| input_file_path: 2023-03-27_oasst_research_ready_synth.jsonl.gz |
| val_split: 0.1 |
| - augment_oasst: |
| input_file_path: augmented_latin_cyrillic_oasst_2023-03-27_v2.jsonl |
| - anthropic_rlhf: |
| fraction: 0.1 |
| max_val_set: 1000 |
| - shp: |
| max_val_set: 1000 |
| - hellaswag: |
| fraction: 0.5 |
| max_val_set: 1000 |
| - webgpt: |
| val_split: 0.05 |
| max_val_set: 1000 |
| - hf_summary_pairs: |
| fraction: 0.1 |
| max_val_set: 250 |
| ``` |
|
|
| (internal note: ignore (high) eval accuracy values of oasst_export, oasst-eval samples were part of training set) |
| |
