README.md

metadata

base_model:
  - AdaptLLM/finance-LLM-13B
  - starmpcc/Asclepius-Llama2-13B
library_name: transformers
tags:
  - mergekit
  - merge

llama3s-merged

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 starmpcc/Asclepius-Llama2-13B as a base.

Models Merged

The following models were included in the merge:

• AdaptLLM/finance-LLM-13B

Configuration

The following YAML configuration was used to produce this model:

base_model: starmpcc/Asclepius-Llama2-13B
dtype: bfloat16
merge_method: dare_ties
modules:
  default:
    slices:
    - sources:
      - layer_range: [0, 40]
        model: AdaptLLM/finance-LLM-13B
        parameters:
          density: 0.53
          weight: 0.6
      - layer_range: [0, 40]
        model: starmpcc/Asclepius-Llama2-13B
        parameters:
          density: 0.5
          weight: 0.4
parameters:
  int8_mask: 1.0