README.md

metadata

base_model:
  - bunnycore/Llama-3.2-3B-Pure-RP
  - huihui-ai/Llama-3.2-3B-Instruct-abliterated
  - Lyte/Llama-3.2-3B-Overthinker
library_name: transformers
tags:
  - mergekit
  - merge

Model Overview

The Llama-3.2-3B-All-Mix model is a merged language model that combines the strengths of multiple models using the TIES merge method. This model is designed to provide a balanced performance across various tasks and domains.

Capabilities

• The Llama-3.2-3B-All-Mix model is capable of:

• Generating human-like text
• Conversational dialogue
• Roleplay
• Long-form reasoning
• Answering questions
• Summarizing text

The following models were included in the merge:

• bunnycore/Llama-3.2-3B-Pure-RP: This model is particularly well-suited for roleplay tasks, allowing for more engaging and interactive conversations.
• Lyte/Llama-3.2-3B-Overthinker: This model excels at long-form reasoning and is capable of generating more in-depth and thoughtful responses.

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 huihui-ai/Llama-3.2-3B-Instruct-abliterated as a base.

Models Merged

The following models were included in the merge:

• bunnycore/Llama-3.2-3B-Pure-RP
• Lyte/Llama-3.2-3B-Overthinker

Configuration

The following YAML configuration was used to produce this model:

models:
  - model: Lyte/Llama-3.2-3B-Overthinker
    parameters:
      density: 0.5
      weight: 0.5
  - model: bunnycore/Llama-3.2-3B-Pure-RP
    parameters:
      density: 0.5
      weight: 0.5

merge_method: ties
base_model: huihui-ai/Llama-3.2-3B-Instruct-abliterated
parameters:
  normalize: false
  int8_mask: true
dtype: float16