merge
This is a merge of pre-trained language models created using mergekit.
Merge Details
Merge Method
This model was merged using the della merge method using Qwen/Qwen2.5-7B as a base.
Models Merged
The following models were included in the merge:
Configuration
The following YAML configuration was used to produce this model:
models:
- model: Qwen/Qwen2.5-7B-Instruct
parameters:
weight: 1
density: 1
lambda: 0.9
merge_method: della
base_model: Qwen/Qwen2.5-7B
parameters:
weight: 1
density: 1
lambda: 0.9
int8_mask: true
dtype: bfloat16
Open LLM Leaderboard Evaluation Results
Detailed results can be found here
Metric | Value |
---|---|
Avg. | 31.83 |
IFEval (0-Shot) | 62.95 |
BBH (3-Shot) | 36.85 |
MATH Lvl 5 (4-Shot) | 30.59 |
GPQA (0-shot) | 8.95 |
MuSR (0-shot) | 12.89 |
MMLU-PRO (5-shot) | 38.75 |
- Downloads last month
- 28
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social
visibility and check back later, or deploy to Inference Endpoints (dedicated)
instead.
Model tree for Etherll/Qwen2.5-7B-della-test
Evaluation results
- strict accuracy on IFEval (0-Shot)Open LLM Leaderboard62.950
- normalized accuracy on BBH (3-Shot)Open LLM Leaderboard36.850
- exact match on MATH Lvl 5 (4-Shot)Open LLM Leaderboard30.590
- acc_norm on GPQA (0-shot)Open LLM Leaderboard8.950
- acc_norm on MuSR (0-shot)Open LLM Leaderboard12.890
- accuracy on MMLU-PRO (5-shot)test set Open LLM Leaderboard38.750