README.md

metadata

base_model:
  - dfurman/CalmeRys-78B-Orpo-v0.1
  - MaziyarPanahi/calme-3.1-instruct-78b
  - MaziyarPanahi/calme-2.4-rys-78b
  - MaziyarPanahi/calme-3.2-instruct-78b
library_name: transformers
tags:
  - mergekit
  - merge

Linkbricks-Horizon-AI-Avengers-V2-78B

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 MaziyarPanahi/calme-3.2-instruct-78b as a base.

Models Merged

The following models were included in the merge:

• dfurman/CalmeRys-78B-Orpo-v0.1
• MaziyarPanahi/calme-3.1-instruct-78b
• MaziyarPanahi/calme-2.4-rys-78b

Configuration

The following YAML configuration was used to produce this model:

base_model: MaziyarPanahi/calme-3.2-instruct-78b
dtype: bfloat16
merge_method: ties
parameters:
  int8_mask: 1.0
  normalize: 1.0
slices:
- sources:
  - layer_range: [0, 86]
    model: MaziyarPanahi/calme-3.2-instruct-78b
  - layer_range: [0, 86]
    model: dfurman/CalmeRys-78B-Orpo-v0.1
    parameters:
      density: 0.53
      weight: 0.2
  - layer_range: [0, 86]
    model: MaziyarPanahi/calme-3.1-instruct-78b
    parameters:
      density: 0.53
      weight: 0.6
  - layer_range: [0, 86]
    model: MaziyarPanahi/calme-2.4-rys-78b
    parameters:
      density: 0.53
      weight: 0.2
tokenizer_source: union