Model Stock: All we need is just a few fine-tuned models
Paper • 2403.19522 • Published • 14
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 KidIkaros/Llama-3.2-1B-Instruct-abliterated as a base.
Models Merged
The following models were included in the merge:
Configuration
The following YAML configuration was used to produce this model:
merge_method: model_stock
models:
- model: Nexesenex/Llama_3.2_1b_Dolto_0.1
parameters:
weight: 1.0
- model: Nexesenex/Llama_3.2_1b_OpenTree_R1_0.1
parameters:
weight: 1.0
base_model: KidIkaros/Llama-3.2-1B-Instruct-abliterated
dtype: bfloat16
normalize: true
Open LLM Leaderboard Evaluation Results
Detailed results can be found here
| Metric | Value |
|---|---|
| Avg. | 12.81 |
| IFEval (0-Shot) | 55.43 |
| BBH (3-Shot) | 7.94 |
| MATH Lvl 5 (4-Shot) | 5.66 |
| GPQA (0-shot) | 0.00 |
| MuSR (0-shot) | 1.58 |
| MMLU-PRO (5-shot) | 6.26 |
- Downloads last month
- 5
Model tree for Nexesenex/Llama_3.2_1b_RandomLego_RP_R1_0.1
Merge model
this model
Paper for Nexesenex/Llama_3.2_1b_RandomLego_RP_R1_0.1
Evaluation results
- strict accuracy on IFEval (0-Shot)Open LLM Leaderboard55.430
- normalized accuracy on BBH (3-Shot)Open LLM Leaderboard7.940
- exact match on MATH Lvl 5 (4-Shot)Open LLM Leaderboard5.660
- acc_norm on GPQA (0-shot)Open LLM Leaderboard0.000
- acc_norm on MuSR (0-shot)Open LLM Leaderboard1.580
- accuracy on MMLU-PRO (5-shot)test set Open LLM Leaderboard6.260