nthehai01/Qwen2.5-7B-Instruct-Math-Code-dare-linear

This is a merge of pre-trained language models created using mergekit.

Performance

| Metric |Value|

|---------------------------------|----:|

|GSM8k (zero-shot) |87.79|

|HellaSwag (zero-Shot) |34.29|

|MBPP (zero-shot) |60.41|

Merge Details

Merge Method

This model was merged using the Linear DARE 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:


base_model: Qwen/Qwen2.5-7B

dtype: bfloat16

merge_method: dare_linear

parameters:

  lambda: 0.690661354021995

  normalize: 1.0

slices:

- sources:

  - layer_range: [0, 28]

    model: Qwen/Qwen2.5-7B

  - layer_range: [0, 28]

    model: Qwen/Qwen2.5-Math-7B

    parameters:

      density: 0.9593725853706829

      weight: 0.11472446469404357

  - layer_range: [0, 28]

    model: Qwen/Qwen2.5-Coder-7B

    parameters:

      density: 0.768281938201547

      weight: 0.11350094855547865

  - layer_range: [0, 28]

    model: Qwen/Qwen2.5-7B-Instruct

    parameters:

      density: 0.48528478746069637

      weight: 0.6453505470133651