Model Stock: All we need is just a few fine-tuned models
Paper
•
2403.19522
•
Published
•
13
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/Qwen2.5-3B-Instruct as a base.
Models Merged
The following models were included in the merge:
- bunnycore/Qwen2.5-3B-RP-Thinker-V2
- bunnycore/QwQen-3B-LCoT + bunnycore/Qwen-2.5-3b-R1-lora_model-v.1
- bunnycore/Qwen2.5-3B-Model-Stock-v2
Configuration
The following YAML configuration was used to produce this model:
models:
- model: bunnycore/Qwen2.5-3B-RP-Thinker-V2
- model: bunnycore/Qwen2.5-3B-Model-Stock-v2
- model: bunnycore/QwQen-3B-LCoT+bunnycore/Qwen-2.5-3b-R1-lora_model-v.1
base_model: Qwen/Qwen2.5-3B-Instruct
merge_method: model_stock
parameters:
normalize: true
dtype: bfloat16
tokenizer_source: Qwen/Qwen2.5-3B-Instruct
Open LLM Leaderboard Evaluation Results
Detailed results can be found here
| Metric | Value |
|---|---|
| Avg. | 27.97 |
| IFEval (0-Shot) | 64.81 |
| BBH (3-Shot) | 26.40 |
| MATH Lvl 5 (4-Shot) | 38.97 |
| GPQA (0-shot) | 4.59 |
| MuSR (0-shot) | 7.63 |
| MMLU-PRO (5-shot) | 25.44 |
- Downloads last month
- 13
Model tree for bunnycore/Qwen2.5-3B-Model-Stock-v3.1
Merge model
this model
Paper for bunnycore/Qwen2.5-3B-Model-Stock-v3.1
Evaluation results
- strict accuracy on IFEval (0-Shot)Open LLM Leaderboard64.810
- normalized accuracy on BBH (3-Shot)Open LLM Leaderboard26.400
- exact match on MATH Lvl 5 (4-Shot)Open LLM Leaderboard38.970
- acc_norm on GPQA (0-shot)Open LLM Leaderboard4.590
- acc_norm on MuSR (0-shot)Open LLM Leaderboard7.630
- accuracy on MMLU-PRO (5-shot)test set Open LLM Leaderboard25.440