GML-Mistral-merged-v1
Quick attempt to reproduce zyh3826/GML-Mistral-merged-v1.
This model is a merge of the following models made with mergekit:
🧩 Configuration
slices:
- sources:
- model: quantumaikr/quantum-v0.01
layer_range: [0, 32]
- sources:
- model: mncai/mistral-7b-dpo-v5
layer_range: [24, 32]
merge_method: passthrough
dtype: bfloat16
Open LLM Leaderboard Evaluation Results
Detailed results can be found here
| Metric | Value |
|---|---|
| Avg. | 48.54 |
| AI2 Reasoning Challenge (25-Shot) | 43.77 |
| HellaSwag (10-Shot) | 57.89 |
| MMLU (5-Shot) | 64.13 |
| TruthfulQA (0-shot) | 51.58 |
| Winogrande (5-shot) | 73.88 |
| GSM8k (5-shot) | 0.00 |
- Downloads last month
- 712
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.
Evaluation results
- normalized accuracy on AI2 Reasoning Challenge (25-Shot)test set Open LLM Leaderboard43.770
- normalized accuracy on HellaSwag (10-Shot)validation set Open LLM Leaderboard57.890
- accuracy on MMLU (5-Shot)test set Open LLM Leaderboard64.130
- mc2 on TruthfulQA (0-shot)validation set Open LLM Leaderboard51.580
- accuracy on Winogrande (5-shot)validation set Open LLM Leaderboard73.880
- accuracy on GSM8k (5-shot)test set Open LLM Leaderboard0.000