README.md

metadata

base_model:
  - occiglot/occiglot-7b-de-en-instruct
  - occiglot/occiglot-7b-it-en-instruct
  - occiglot/occiglot-7b-fr-en-instruct
  - occiglot/occiglot-7b-es-en-instruct
  - mistralai/Mistral-7B-v0.1
library_name: transformers
tags:
  - mergekit
  - merge

occitaskarithmetic

This is a merge of pre-trained language models created using mergekit.

Merge Details

Merge Method

This model was merged using the task arithmetic merge method using mistralai/Mistral-7B-v0.1 as a base.

Models Merged

The following models were included in the merge:

• occiglot/occiglot-7b-de-en-instruct
• occiglot/occiglot-7b-it-en-instruct
• occiglot/occiglot-7b-fr-en-instruct
• occiglot/occiglot-7b-es-en-instruct

Configuration

The following YAML configuration was used to produce this model:

models:
  - model: mistralai/Mistral-7B-v0.1
    # no parameters necessary for base model
  - model: occiglot/occiglot-7b-de-en-instruct
    parameters:
      density: 0.6
      weight: 0.25
  - model: occiglot/occiglot-7b-it-en-instruct
    parameters:
      density: 0.6
      weight: 0.25
  - model: occiglot/occiglot-7b-fr-en-instruct
    parameters:
      density: 0.6
      weight: 0.25
  - model: occiglot/occiglot-7b-es-en-instruct
    parameters:
      density: 0.6
      weight: 0.25
merge_method: task_arithmetic
base_model: mistralai/Mistral-7B-v0.1
parameters:
  int8_mask: true
dtype: bfloat16
random_seed: 0
tokenizer_source: model:occiglot/occiglot-7b-de-en-instruct