mistral-nemo-cc-12B
nbeerbower/mistral-nemo-gutenberg-12B-v3 finetuned on flammenai/casual-conversation-DPO.
This is an experimental finetune that formats the conversation data sequentially with ChatML.
Method
Finetuned using an A100 on Google Colab for 3 epochs.
Open LLM Leaderboard Evaluation Results
Detailed results can be found here
| Metric | Value |
|---|---|
| Avg. | 17.08 |
| IFEval (0-Shot) | 14.35 |
| BBH (3-Shot) | 34.45 |
| MATH Lvl 5 (4-Shot) | 1.81 |
| GPQA (0-shot) | 8.72 |
| MuSR (0-shot) | 14.26 |
| MMLU-PRO (5-shot) | 28.87 |
- Downloads last month
- 6
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social
visibility and check back later, or deploy to Inference Endpoints (dedicated)
instead.
Model tree for nbeerbower/mistral-nemo-cc-12B
Base model
intervitens/mini-magnum-12b-v1.1
Finetuned
nbeerbower/mistral-nemo-gutenberg-12B-v3
Dataset used to train nbeerbower/mistral-nemo-cc-12B
Evaluation results
- strict accuracy on IFEval (0-Shot)Open LLM Leaderboard14.350
- normalized accuracy on BBH (3-Shot)Open LLM Leaderboard34.450
- exact match on MATH Lvl 5 (4-Shot)Open LLM Leaderboard1.810
- acc_norm on GPQA (0-shot)Open LLM Leaderboard8.720
- acc_norm on MuSR (0-shot)Open LLM Leaderboard14.260
- accuracy on MMLU-PRO (5-shot)test set Open LLM Leaderboard28.870