Modelpublic

arkitekt-ai/qwen-coder-7b-specialists-dare-ties

sourceHugging Faceupdated 8mo agoView on Hugging Face

0likes7downloads

Model Card

qwen-coder-7b-specialists-dare-ties

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

Merge Details

Merge Method

This model was merged using the DARE TIES merge method using Qwen/Qwen2.5-Coder-7B-Instruct as a base.

Models Merged

The following models were included in the merge:

Configuration

The following YAML configuration was used to produce this model:

yaml

models:
  - model: Qwen/Qwen2.5-Coder-7B-Instruct
    # Base model - no parameters needed
  - model: arkitekt-ai/qwen-coder-7b-specialist-d1
    parameters:
      density: 0.5  # Keep 50% of parameters (drop 50%)
      weight: 0.5
  - model: arkitekt-ai/qwen-coder-7b-specialist-d2
    parameters:
      density: 0.5
      weight: 0.5

merge_method: dare_ties
base_model: Qwen/Qwen2.5-Coder-7B-Instruct
parameters:
  int8_mask: true  # More memory efficient
  normalize: true
dtype: bfloat16