Bagel DPO 57B
Model Details
- A result of interleaving layers of jondurbin/bagel-dpo-34b-v0.2 with itself.
- The resulting model has 100 layers and approximately 57 billion parameters.
- See mergekit-config.yml for details on the merge method used.
Warning: This model can produce NSFW content!
Results
Bigger version of original, uncensored like oryginal. All comments are greatly appreciated, download, test and if you appreciate my work, consider buying me my fuel:
Open LLM Leaderboard Evaluation Results
Detailed results can be found here
| Metric | Value |
|---|---|
| Avg. | 60.66 |
| AI2 Reasoning Challenge (25-Shot) | 65.27 |
| HellaSwag (10-Shot) | 79.35 |
| MMLU (5-Shot) | 73.64 |
| TruthfulQA (0-shot) | 67.15 |
| Winogrande (5-shot) | 76.40 |
| GSM8k (5-shot) | 2.12 |
- Downloads last month
- 3,051
Evaluation results
- normalized accuracy on AI2 Reasoning Challenge (25-Shot)test set Open LLM Leaderboard65.270
- normalized accuracy on HellaSwag (10-Shot)validation set Open LLM Leaderboard79.350
- accuracy on MMLU (5-Shot)test set Open LLM Leaderboard73.640
- mc2 on TruthfulQA (0-shot)validation set Open LLM Leaderboard67.150
- accuracy on Winogrande (5-shot)validation set Open LLM Leaderboard76.400
- accuracy on GSM8k (5-shot)test set Open LLM Leaderboard2.120