Llama
Collection
Meta-based models
•
1004 items
•
Updated
•
1
Merge
This is a merge of pre-trained language models created using mergekit.
Merge Details
Llama 3.2 3b model combined with lora trained on reasoning datasets.
Merge Method
This model was merged using the Passthrough merge method using cognitivecomputations/Dolphin3.0-Llama3.2-3B + bunnycore/Llama-3.2-3B-R1-lora as a base.
Models Merged
The following models were included in the merge:
Configuration
The following YAML configuration was used to produce this model:
base_model: cognitivecomputations/Dolphin3.0-Llama3.2-3B+bunnycore/Llama-3.2-3B-R1-lora
dtype: bfloat16
merge_method: passthrough
models:
- model: cognitivecomputations/Dolphin3.0-Llama3.2-3B+bunnycore/Llama-3.2-3B-R1-lora
tokenizer_source: cognitivecomputations/Dolphin3.0-Llama3.2-3B
Open LLM Leaderboard Evaluation Results
Detailed results can be found here
| Metric | Value |
|---|---|
| Avg. | 13.63 |
| IFEval (0-Shot) | 43.52 |
| BBH (3-Shot) | 12.93 |
| MATH Lvl 5 (4-Shot) | 4.23 |
| GPQA (0-shot) | 2.24 |
| MuSR (0-shot) | 6.44 |
| MMLU-PRO (5-shot) | 12.41 |
- Downloads last month
- 5
Inference Providers
NEW
This model isn't deployed by any Inference Provider.
🙋
Ask for provider support
Model tree for Triangle104/Porpoise-R1-Llama3.2-3b
Merge model
this model
Collections including Triangle104/Porpoise-R1-Llama3.2-3b
Evaluation results
- strict accuracy on IFEval (0-Shot)Open LLM Leaderboard43.520
- normalized accuracy on BBH (3-Shot)Open LLM Leaderboard12.930
- exact match on MATH Lvl 5 (4-Shot)Open LLM Leaderboard4.230
- acc_norm on GPQA (0-shot)Open LLM Leaderboard2.240
- acc_norm on MuSR (0-shot)Open LLM Leaderboard6.440
- accuracy on MMLU-PRO (5-shot)test set Open LLM Leaderboard12.410