mergekit-community/Llama-3-8B-666

merge

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

Merge Details

Merge Method

This model was merged using the passthrough merge method using meta-llama/Meta-Llama-3-8B-Instruct as a base.

Models Merged

The following models were included in the merge:

• NousResearch/Hermes-2-Pro-Llama-3-8B
• hfl/llama-3-chinese-8b-instruct-v2
• shenzhi-wang/Llama3-8B-Chinese-Chat

Configuration

The following YAML configuration was used to produce this model:

slices:
  - sources:
      - model: "shenzhi-wang/Llama3-8B-Chinese-Chat"
        layer_range: [0, 10]
  - sources:
      - model: "hfl/llama-3-chinese-8b-instruct-v2"
        layer_range: [7, 17]
  - sources:
      - model: "NousResearch/Hermes-2-Pro-Llama-3-8B"
        layer_range: [13, 23]
  - sources:
      - model: "NousResearch/Hermes-2-Pro-Llama-3-8B"
        layer_range: [18, 28]
  - sources:
      - model: "NousResearch/Hermes-2-Pro-Llama-3-8B"
        layer_range: [22, 32]

merge_method: passthrough
base_model: "meta-llama/Meta-Llama-3-8B-Instruct"
dtype: bfloat16