README.md

metadata

base_model:
  - DoumAI/Krx-Bench-2024-Qwen2.5-1.5B-Instruct-1106
  - Sakalti/model-4
library_name: transformers
tags:
  - mergekit
  - merge
inference: true
language:
  - ko

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 Sakalti/model-4 as a base.

Models Merged

The following models were included in the merge:

• DoumAI/Krx-Bench-2024-Qwen2.5-1.5B-Instruct-1106

Configuration

The following YAML configuration was used to produce this model:

models:
  - model: DoumAI/Krx-Bench-2024-Qwen2.5-1.5B-Instruct-1106
    parameters:
      weight: 1
      density: 1
merge_method: ties
base_model: Sakalti/model-4
parameters:
  weight: 1
  density: 1
  normalize: true
  int8_mask: true
dtype: bfloat16

注意点

韓国語メインですがシステムメッセージで日本語でしゃべってなど指定すると日本語でしゃべるのが多くなります。