qwen-2.5-7B-Instruct-Resta-lr5e-5-scale0.1

resta_recovered_0.5

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

Merge Details

Merge Method

This model was merged using the linear merge method.

Models Merged

The following models were included in the merge:

• wvnvwn/gemma-2-9b-it-lr5e-5-gsm8k-lr5e-5
• google/gemma-2-9b-it
• wvnvwn/gemma-2-9b-it-ssft-lr5e-5

Configuration

The following YAML configuration was used to produce this model:

dtype: float16
merge_method: linear
slices:
- sources:
  - layer_range: [0, 42]
    model:
      model:
        path: wvnvwn/gemma-2-9b-it-lr5e-5-gsm8k-lr5e-5
    parameters:
      weight: 1.0
  - layer_range: [0, 42]
    model:
      model:
        path: wvnvwn/gemma-2-9b-it-ssft-lr5e-5
    parameters:
      weight: 0.5
  - layer_range: [0, 42]
    model:
      model:
        path: google/gemma-2-9b-it
    parameters:
      weight: -0.5