Upload README.md with huggingface_hub
Browse files
README.md
ADDED
|
@@ -0,0 +1,31 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
{}
|
| 3 |
+
---
|
| 4 |
+
# allenai/open_instruct: Rejection Sampling Dataset
|
| 5 |
+
|
| 6 |
+
See https://github.com/allenai/open-instruct/blob/main/docs/algorithms/rejection_sampling.md for more detail
|
| 7 |
+
|
| 8 |
+
## Configs
|
| 9 |
+
|
| 10 |
+
```
|
| 11 |
+
args:
|
| 12 |
+
{'add_timestamp': True,
|
| 13 |
+
'hf_entity': 'vwxyzjn',
|
| 14 |
+
'hf_repo_id': 'rejection_sampling',
|
| 15 |
+
'hf_repo_id_scores': 'rejection_sampling_scores',
|
| 16 |
+
'include_reference_completion_for_rejection_sampling': True,
|
| 17 |
+
'input_filename': 'output/completions.jsonl',
|
| 18 |
+
'max_forward_batch_size': 64,
|
| 19 |
+
'mode': 'judgement',
|
| 20 |
+
'model_names_or_paths': ['allenai/llama-3-tulu-2-8b-uf-mean-rm'],
|
| 21 |
+
'num_completions': 3,
|
| 22 |
+
'num_gpus': 1,
|
| 23 |
+
'push_to_hub': True,
|
| 24 |
+
'save_filename': 'output/rejection_sampled_completions.jsonl',
|
| 25 |
+
'save_filename_scores': 'output/completions_scores.jsonl',
|
| 26 |
+
'skill': 'chat'}
|
| 27 |
+
```
|
| 28 |
+
|
| 29 |
+
## Additional Information
|
| 30 |
+
|
| 31 |
+
1. Command used to run `python open_instruct/rejection_sampling/rejection_sampling.py --input_filename output/completions.jsonl --model_names_or_paths allenai/llama-3-tulu-2-8b-uf-mean-rm --save_filename_scores output/completions_scores.jsonl --save_filename output/rejection_sampled_completions.jsonl --num_completions 3 --push_to_hub --num_gpus 1`
|