hankook_llama

Untitled Model (1)

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 /home/metaai2/hyenjin/model/korean_model/Meta-Llama-3.1-8B-Instruct-abliterated/ as a base.

Models Merged

The following models were included in the merge:

• /home/metaai2/hyenjin/model/korean_model/Llama-3-Alpha-Ko-8B-Instruct/
• /home/metaai2/hyenjin/model/korean_model/Ko-Llama-3-8B-Instruct/
• /home/metaai2/hyenjin/model/korean_model/Llama3-Ko-Carrot-8B-it/
• /home/metaai2/hyenjin/model/korean_model/Llama-3-Lumimaid-8B-v0.1/
• /home/metaai2/hyenjin/model/korean_model/KONI-Llama3-8B-Instruct-20240729/
• /home/metaai2/hyenjin/model/korean_model/llama-3-Korean-Bllossom-8B/

Configuration

The following YAML configuration was used to produce this model:

models:
  - model: /home/metaai2/hyenjin/model/korean_model/Meta-Llama-3.1-8B-Instruct-abliterated/
    parameters:
      density: 0.8
      weight: 0.5
  - model: /home/metaai2/hyenjin/model/korean_model/llama-3-Korean-Bllossom-8B/
    parameters:
      density: 0.7
      weight: 0.4
  - model: /home/metaai2/hyenjin/model/korean_model/KONI-Llama3-8B-Instruct-20240729/
    parameters:
      density: 0.6
      weight: 0.3
  - model: /home/metaai2/hyenjin/model/korean_model/Llama3-Ko-Carrot-8B-it/
    parameters:
      density: 0.5
      weight: 0.25
  - model: /home/metaai2/hyenjin/model/korean_model/Ko-Llama-3-8B-Instruct/
    parameters:
      density: 0.4
      weight: 0.2
  - model: /home/metaai2/hyenjin/model/korean_model/Llama-3-Alpha-Ko-8B-Instruct/
    parameters:
      density: 0.4
      weight: 0.2
  - model: /home/metaai2/hyenjin/model/korean_model/Llama-3-Lumimaid-8B-v0.1/
    parameters:
      density: 0.4
      weight: 0.2
merge_method: dare_ties
base_model: /home/metaai2/hyenjin/model/korean_model/Meta-Llama-3.1-8B-Instruct-abliterated/
parameters:
  int8_mask: true
dtype: bfloat16