kaizen9/a4b-dupe

Untitled Model (1)

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.

Models Merged

The following models were included in the merge:

• RedMod/aspen4b-prev-q

Configuration

The following YAML configuration was used to produce this model:

slices:
  # Duplicate all layers - create a 72-layer Frankenstein model
  - sources:
      - model: RedMod/aspen4b-prev-q
        layer_range: [0, 36]  # Original 36 layers
  - sources:
      - model: RedMod/aspen4b-prev-q
        layer_range: [0, 36]  # Duplicate all 36 layers again

merge_method: passthrough
dtype: bfloat16