README.md

metadata

base_model:
  - Qwen/Qwen2.5-7B
  - pittawat/qwen2.5-7b-med-acc-without-prior
  - pittawat/qwen2.5-7b-med-acc-without-prior-alphamed
  - pittawat/qwen2.5-7b-med-acc-mcq-rank-without-prior
library_name: transformers
tags:
  - mergekit
  - merge

dare-linear-3-models

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

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:

• pittawat/qwen2.5-7b-med-acc-without-prior
• pittawat/qwen2.5-7b-med-acc-without-prior-alphamed
• pittawat/qwen2.5-7b-med-acc-mcq-rank-without-prior

Configuration

The following YAML configuration was used to produce this model:

models:
  - model: Qwen/Qwen2.5-7B
  - model: pittawat/qwen2.5-7b-med-acc-without-prior
    parameters:
      density: 0.53
      weight: 0.5
  - model: pittawat/qwen2.5-7b-med-acc-without-prior-alphamed
    parameters:
      density: 0.53
      weight: 0.3
  - model: pittawat/qwen2.5-7b-med-acc-mcq-rank-without-prior
    parameters:
      density: 0.53
      weight: 0.2
merge_method: dare_linear
base_model: Qwen/Qwen2.5-7B
parameters:
  normalize: true
dtype: bfloat16

tokenizer:
  source: Qwen/Qwen2.5-7B-Instruct