mergekit-community/mergekit-dare_ties-stjgmmc

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 TheDrummer/Anubis-70B-v1 as a base.

Models Merged

The following models were included in the merge:

• EVA-UNIT-01/EVA-LLaMA-3.33-70B-v0.1
• unsloth/Llama-3.3-70B-Instruct
• nvidia/Llama-3.1-Nemotron-70B-Instruct-HF

Configuration

The following YAML configuration was used to produce this model:

models:
  - model: unsloth/Llama-3.3-70B-Instruct
    parameters:
      density: [0.9, 0.1, 0.1]
      weight: [0.9, 0.3, 0.3]
  - model: nvidia/Llama-3.1-Nemotron-70B-Instruct-HF
    parameters:
      density: [0.1, 0.7, 0.9]
      weight: [0.2, 0.5, 0.7]
  - model: EVA-UNIT-01/EVA-LLaMA-3.33-70B-v0.1
    parameters:
      density: [0.1, 0.4, 0.2]
      weight: [0.1, 0.4, 0.2]
merge_method: dare_ties
base_model: TheDrummer/Anubis-70B-v1
parameters:
  normalize: true
  int8_mask: true
dtype: bfloat16