README.md

metadata

license: apache-2.0
tags:
  - merge
  - mergekit
  - lazymergekit
  - unsloth/Llama-3.2-3B-Instruct
  - ValiantLabs/Llama3.2-3B-Enigma
  - Lyte/Llama-3.2-3B-Overthinker
  - huihui-ai/Llama-3.2-3B-Instruct-abliterated

Llama-3.2-16B-Instruct

Llama-3.2-16B-Instruct is a merge of the following models using mergekit:

• unsloth/Llama-3.2-3B-Instruct
• ValiantLabs/Llama3.2-3B-Enigma
• Lyte/Llama-3.2-3B-Overthinker
• huihui-ai/Llama-3.2-3B-Instruct-abliterated

🧩 Configuration

slices:
  - sources:
      - model: unsloth/Llama-3.2-3B-Instruct
        layer_range: [0, 28]
  - sources:
      - model: ValiantLabs/Llama3.2-3B-Enigma
        layer_range: [4, 24]
  - sources:
      - model: Lyte/Llama-3.2-3B-Overthinker
        layer_range: [4, 24]
  - sources:
      - model: huihui-ai/Llama-3.2-3B-Instruct-abliterated
        layer_range: [4, 28]
merge_method: passthrough
dtype: bfloat16