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 artificialguybr/LLAMA3.2-1B-Synthia-II-Redmond 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: hyunseoki/llama3.2-1b-Open-R1-GRPO-test0
parameters:
weight: 1.0
- model: artificialguybr/LLAMA-3.2-1B-OpenHermes2.5
parameters:
weight: 1.0
base_model: artificialguybr/LLAMA3.2-1B-Synthia-II-Redmond
dtype: bfloat16
normalize: true
chat_template: auto
tokenizer:
source: union
Open LLM Leaderboard Evaluation Results
Detailed results can be found here
| Metric | Value |
|---|---|
| Avg. | 6.56 |
| IFEval (0-Shot) | 28.42 |
| BBH (3-Shot) | 3.44 |
| MATH Lvl 5 (4-Shot) | 1.28 |
| GPQA (0-shot) | 1.68 |
| MuSR (0-shot) | 3.15 |
| MMLU-PRO (5-shot) | 1.36 |
- Downloads last month
- 27
Inference Providers
NEW
This model is not currently available via any of the supported Inference Providers.
Model tree for Nexesenex/Llama_3.2_1b_Synopsys_0.11
Merge model
this model
Evaluation results
- strict accuracy on IFEval (0-Shot)Open LLM Leaderboard28.420
- normalized accuracy on BBH (3-Shot)Open LLM Leaderboard3.440
- exact match on MATH Lvl 5 (4-Shot)Open LLM Leaderboard1.280
- acc_norm on GPQA (0-shot)Open LLM Leaderboard1.680
- acc_norm on MuSR (0-shot)Open LLM Leaderboard3.150
- accuracy on MMLU-PRO (5-shot)test set Open LLM Leaderboard1.360