README.md

metadata

base_model:
  - sarvamai/sarvam-m
  - LatitudeGames/Harbinger-24B
  - aixonlab/Eurydice-24b-v3.5
library_name: transformers
tags:
  - mergekit
  - merge

merge

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 LatitudeGames/Harbinger-24B as a base.

Models Merged

The following models were included in the merge:

• sarvamai/sarvam-m
• aixonlab/Eurydice-24b-v3.5

Configuration

The following YAML configuration was used to produce this model:

base_model: LatitudeGames/Harbinger-24B
merge_method: dare_ties
dtype: bfloat16

models:
  - model: LatitudeGames/Harbinger-24B
    parameters:
      weight: 0.45
      density: 0.92
  - model: sarvamai/sarvam-m
    parameters:
      weight: 0.30
      density: 0.86
  - model: aixonlab/Eurydice-24b-v3.5
    parameters:
      weight: 0.25
      density: 0.86

chat_template: auto

parameters:
  normalize: true
  int8_mask: true
  tolerance: 1e-7