Model Stock: All we need is just a few fine-tuned models
Paper
•
2403.19522
•
Published
•
13
quantizations
GGUF static quantizations (Thanks Mradermacher!) :
https://huggingface.co/mradermacher/Llama_3.2_3b_Kermes_v2-GGUF
GGUF iMatrix quantizations (Thanks Mradermacher!) :
https://huggingface.co/mradermacher/Llama_3.2_3b_Kermes_v2-i1-GGUF
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 Nexesenex/Llama_3.2_3b_Kermes_0.2_bf16 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: MaziyarPanahi/calme-3.3-llamaloi-3b
parameters:
weight: 1.0
- model: cognitivecomputations/Dolphin3.0-Llama3.2-3B
parameters:
weight: 1.0
base_model: Nexesenex/Llama_3.2_3b_Kermes_0.2_bf16
dtype: float16
normalize: true
Open LLM Leaderboard Evaluation Results
Detailed results can be found here! Summarized results can be found here!
| Metric | Value (%) |
|---|---|
| Average | 18.50 |
| IFEval (0-Shot) | 57.54 |
| BBH (3-Shot) | 22.05 |
| MATH Lvl 5 (4-Shot) | 5.44 |
| GPQA (0-shot) | 2.01 |
| MuSR (0-shot) | 4.66 |
| MMLU-PRO (5-shot) | 19.27 |
- Downloads last month
- 8
Model tree for Nexesenex/Llama_3.2_3b_Kermes_v2
Merge model
this model
Paper for Nexesenex/Llama_3.2_3b_Kermes_v2
Evaluation results
- averaged accuracy on IFEval (0-Shot)Open LLM Leaderboard57.540
- normalized accuracy on BBH (3-Shot)test set Open LLM Leaderboard22.050
- exact match on MATH Lvl 5 (4-Shot)test set Open LLM Leaderboard5.440
- acc_norm on GPQA (0-shot)Open LLM Leaderboard2.010
- acc_norm on MuSR (0-shot)Open LLM Leaderboard4.660
- accuracy on MMLU-PRO (5-shot)test set Open LLM Leaderboard19.270