merge

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 FlagAlpha/Llama2-Chinese-7b-Chat as a base.

Models Merged

The following models were included in the merge:

• andreidima/Llama-2-7b-Romanian
• Rishabh02/Llama-2-7b-hindi_fine_fine
• BanglaLLM/bangla-llama-7b-instruct-v0.1

Configuration

The following YAML configuration was used to produce this model:

models:
  - model: "FlagAlpha/Llama2-Chinese-7b-Chat"
    parameters:
      density: 0.2
      weight: 0.2
  - model: "BanglaLLM/bangla-llama-7b-instruct-v0.1"
    parameters:
      density: 0.2
      weight: 0.2
  - model: "andreidima/Llama-2-7b-Romanian"
    parameters:
      density: 0.2
      weight: 0.2
  - model: "Rishabh02/Llama-2-7b-hindi_fine_fine"
    parameters:
      density: 0.4
      weight: 0.4

merge_method: "ties"
base_model: "FlagAlpha/Llama2-Chinese-7b-Chat"
parameters:
  normalize: false
  int8_mask: true
dtype: float16