merge
This is a merge of pre-trained language models created using mergekit.
Merge Details
Merge Method
This model was merged using the Model Stock merge method using Qwen/Qwen2.5-14B as a base.
Models Merged
The following models were included in the merge:
- v000000/Qwen2.5-Lumen-14B
- arcee-ai/SuperNova-Medius
- rombodawg/Rombos-LLM-V2.6-Qwen-14b
- Qwen/Qwen2.5-14B-Instruct
- EVA-UNIT-01/EVA-Qwen2.5-14B-v0.0
Configuration
The following YAML configuration was used to produce this model:
models:
- model: EVA-UNIT-01/EVA-Qwen2.5-14B-v0.0
- model: Qwen/Qwen2.5-14B-Instruct
- model: arcee-ai/SuperNova-Medius
- model: rombodawg/Rombos-LLM-V2.6-Qwen-14b
- model: v000000/Qwen2.5-Lumen-14B
base_model: Qwen/Qwen2.5-14B
merge_method: model_stock
dtype: bfloat16
Open LLM Leaderboard Evaluation Results
Detailed results can be found here
| Metric | Value |
|---|---|
| Avg. | 36.60 |
| IFEval (0-Shot) | 52.35 |
| BBH (3-Shot) | 50.64 |
| MATH Lvl 5 (4-Shot) | 30.06 |
| GPQA (0-shot) | 19.13 |
| MuSR (0-shot) | 18.25 |
| MMLU-PRO (5-shot) | 49.15 |
Open LLM Leaderboard Evaluation Results
Detailed results can be found here
| Metric | Value |
|---|---|
| Avg. | 33.52 |
| IFEval (0-Shot) | 49.37 |
| BBH (3-Shot) | 49.33 |
| MATH Lvl 5 (4-Shot) | 35.80 |
| GPQA (0-shot) | 13.87 |
| MuSR (0-shot) | 7.55 |
| MMLU-PRO (5-shot) | 45.22 |
- Downloads last month
- 73
Model tree for CultriX/Qwen2.5-14B-Wernicke-SFT
Datasets used to train CultriX/Qwen2.5-14B-Wernicke-SFT
Evaluation results
- strict accuracy on IFEval (0-Shot)Open LLM Leaderboard52.350
- strict accuracy on IFEval (0-Shot)Open LLM Leaderboard49.370
- normalized accuracy on BBH (3-Shot)Open LLM Leaderboard50.640
- normalized accuracy on BBH (3-Shot)Open LLM Leaderboard49.330
- exact match on MATH Lvl 5 (4-Shot)Open LLM Leaderboard30.060
- exact match on MATH Lvl 5 (4-Shot)Open LLM Leaderboard35.800
- acc_norm on GPQA (0-shot)Open LLM Leaderboard19.130
- acc_norm on GPQA (0-shot)Open LLM Leaderboard13.870
- acc_norm on MuSR (0-shot)Open LLM Leaderboard18.250
- acc_norm on MuSR (0-shot)Open LLM Leaderboard7.550