README.md

metadata

base_model:
  - happzy2633/qwen2.5-7b-ins-v3
  - AIDC-AI/Marco-o1
library_name: transformers
tags:
  - mergekit
  - merge

merge

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

Merge Details

Merge Method

This model was merged using the TIES merge method using happzy2633/qwen2.5-7b-ins-v3 as a base.

Models Merged

The following models were included in the merge:

• AIDC-AI/Marco-o1

Configuration

The following YAML configuration was used to produce this model:

models:
  - model: AIDC-AI/Marco-o1
    parameters:
      density: [1, 0.7, 0.1] # density gradient
      weight: 0.8 # slightly reduced weight
  - model: happzy2633/qwen2.5-7b-ins-v3
    parameters:
      density: 0.5
      weight: [0.2, 0.4, 0.6, 0.8] # adjusted weight gradient
  - model: AIDC-AI/Marco-o1
    parameters:
      density: 0.33
      weight:
        - filter: mlp
          value: 0.6 # slightly increased weight for mlp
        - value: 0.1 # slight weight for other layers
merge_method: ties
base_model: happzy2633/qwen2.5-7b-ins-v3
parameters:
  normalize: true
  int8_mask: true
dtype: bfloat16 # trying bfloat16 for potential performance improvement