lyraresearchlabs/affine-merged-v5

merged_v5_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 /models/qwen3-4b-instruct-2507 as a base.

Models Merged

The following models were included in the merge:

• /models/lgc-v2-top-train11
• /models/game-top-ana4-4
• /models/print-top-0101928

Configuration

The following YAML configuration was used to produce this model:

merge_method: dare_ties
base_model: /models/qwen3-4b-instruct-2507
models:
  - model: /models/game-top-ana4-4
    parameters:
      density: 0.1
      weight: 0.5
  - model: /models/print-top-0101928
    parameters:
      density: 0.1
      weight: 0.5
  - model: /models/lgc-v2-top-train11
    parameters:
      density: 0.2
      weight: 1.0
parameters:
  normalize: true
  int8_mask: true
dtype: bfloat16