baconnier/Napoleon_24B_V0.1

merge

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

Merge Details

Merge Method

This model was merged using the SLERP merge method.

Models Merged

The following models were included in the merge:

• baconnier/Napoleon_24B_V0.0
• cognitivecomputations/Dolphin3.0-Mistral-24B

Configuration

The following YAML configuration was used to produce this model:

slices:
  - sources:
    - model: cognitivecomputations/Dolphin3.0-Mistral-24B
      layer_range: [0, 39]  # 40 layers (0-39)
    - model: baconnier/Napoleon_24B_V0.0
      layer_range: [0, 39]

merge_method: slerp
base_model: cognitivecomputations/Dolphin3.0-Mistral-24B  # Use one of the source models

parameters:
  t:
    - filter: self_attn
      value: 0.5
    - filter: mlp
      value: 0.5
    - value: 0.5

dtype: float16  # Changed to float16 for wider compatibility
tokenizer_source: baconnier/Napoleon_24B_V0.0  # Use Napoleon's tokenizer