README.md

metadata

base_model:
  - grimjim/Magot-v1-Gemma2-8k-9B
  - ehristoforu/Gemma2-9b-it-train6
library_name: transformers
tags:
  - mergekit
  - merge
license: apache-2.0

4PRYMMAL-GEMMA2-9B-SLERP

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:

• grimjim/Magot-v1-Gemma2-8k-9B
• ehristoforu/Gemma2-9b-it-train6

Configuration

The following YAML configuration was used to produce this model:

 
slices:
  - sources:
      - model: grimjim/Magot-v1-Gemma2-8k-9B
        layer_range: [0, 42]
      - model: ehristoforu/Gemma2-9b-it-train6
        layer_range: [0, 42]    
merge_method: slerp
base_model: ehristoforu/Gemma2-9b-it-train6
parameters:
  t:
    - filter: self_attn
      value: [0, 0.25, 0.5, 0.75, 1]
    - filter: mlp
      value: [1, 0.75, 0.5, 0.25, 0]
    - value: 0.5
dtype: bfloat16