README.md

metadata

base_model:
  - Gryphe/Pantheon-RP-1.0-8b-Llama-3
  - Locutusque/Llama-3-Hercules-5.0-8B
  - aaditya/Llama3-OpenBioLLM-8B
  - Blackroot/Llama-3-LongStory
library_name: transformers
tags:
  - mergekit
  - merge

Sun-v0.1-8B

This is a merge of pre-trained language models created using mergekit. This is one of my merge parts for EtherealRainbow-v0.2-8B. It's pretty unusable by itself (lots of bugs/issues) but merged into a different base it seems to increase response length and quality of prose.

Uploading just in case others find it useful.

Merge Details

Merge Method

This model was merged using the DARE TIES merge method using Gryphe/Pantheon-RP-1.0-8b-Llama-3 as a base.

Models Merged

The following models were included in the merge:

• Locutusque/Llama-3-Hercules-5.0-8B
• aaditya/Llama3-OpenBioLLM-8B
• Blackroot/Llama-3-LongStory

Configuration

The following YAML configuration was used to produce this model:

models:
  - model: Gryphe/Pantheon-RP-1.0-8b-Llama-3
  - model: aaditya/Llama3-OpenBioLLM-8B
    parameters:
      density: 0.36
      weight: 0.2
  - model: Blackroot/Llama-3-LongStory
    parameters:
      density: 0.40
      weight: 0.3
  - model: Locutusque/Llama-3-Hercules-5.0-8B
    parameters:
      density: 0.49
      weight: 0.5
merge_method: dare_ties
base_model: Gryphe/Pantheon-RP-1.0-8b-Llama-3
parameters:
  int8_mask: true
dtype: bfloat16