mergekit-uploader's picture
Upload folder using huggingface_hub
12d0733 verified
metadata
base_model:
  - Lambent/Gilded-Arsenic-12B
  - mergekit-community/MN-Sappho-j-12B
  - mergekit-community/MN-Hekate-Noctiluca-12B-v2
  - nbeerbower/mistral-nemo-gutenberg-12B-v4
  - nbeerbower/mistral-nemo-bophades-12B
  - mistralai/Mistral-Nemo-Base-2407
  - mergekit-community/MN-Hekate-Limenoskopos-17B
  - nbeerbower/Mistral-Nemo-Gutenberg-Doppel-12B-v2
library_name: transformers
tags:
  - mergekit
  - merge

merge

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

Merge Details

Merge Method

This model was merged using the Model Stock merge method using mergekit-community/MN-Hekate-Noctiluca-12B-v2 as a base.

Models Merged

The following models were included in the merge:

Configuration

The following YAML configuration was used to produce this model:

out_dtype: bfloat16
merge_method: model_stock
base_model: mergekit-community/MN-Hekate-Noctiluca-12B-v2
slices:
  - sources:
    - model: mergekit-community/MN-Hekate-Noctiluca-12B-v2
      layer_range: [0, 12]
      
  - sources:
    - model: mergekit-community/MN-Hekate-Noctiluca-12B-v2
      layer_range: [12, 16]
    - model: mergekit-community/MN-Sappho-j-12B
      layer_range: [12, 16]
    - model: mistralai/Mistral-Nemo-Base-2407
      layer_range: [12, 16]
      parameters:
        weight: 0.5
    - model: Lambent/Gilded-Arsenic-12B
      layer_range: [12, 16]
    - model: nbeerbower/Mistral-Nemo-Gutenberg-Doppel-12B-v2
      layer_range: [12, 16]
    - model: mergekit-community/MN-Hekate-Limenoskopos-17B
      layer_range: [12, 16]

  - sources:
    - model: mergekit-community/MN-Hekate-Noctiluca-12B-v2
      layer_range: [16, 20]
    - model: mergekit-community/MN-Sappho-j-12B
      layer_range: [16, 20]
    - model: mistralai/Mistral-Nemo-Base-2407
      layer_range: [16, 20]
      parameters:
        weight: 0.5
    - model: Lambent/Gilded-Arsenic-12B
      layer_range: [16, 20]
    - model: nbeerbower/Mistral-Nemo-Gutenberg-Doppel-12B-v2
      layer_range: [16, 20]
    - model: mergekit-community/MN-Hekate-Limenoskopos-17B
      layer_range: [16, 20]
    - model: mergekit-community/MN-Hekate-Limenoskopos-17B
      layer_range: [20, 24]

  - sources:
    - model: mergekit-community/MN-Hekate-Noctiluca-12B-v2
      layer_range: [20, 28]
    - model: mergekit-community/MN-Sappho-j-12B
      layer_range: [20, 28]
    - model: mistralai/Mistral-Nemo-Base-2407
      layer_range: [20, 28]
      parameters:
        weight: 0.5
    - model: Lambent/Gilded-Arsenic-12B
      layer_range: [20, 28]
    - model: nbeerbower/Mistral-Nemo-Gutenberg-Doppel-12B-v2
      layer_range: [20, 28]
    - model: mergekit-community/MN-Hekate-Limenoskopos-17B
      layer_range: [24, 32]
    - model: mergekit-community/MN-Hekate-Limenoskopos-17B
      layer_range: [36, 44]

  - sources:
    - model: mergekit-community/MN-Hekate-Noctiluca-12B-v2
      layer_range: [28, 32]
    - model: mergekit-community/MN-Sappho-j-12B
      layer_range: [28, 32]
    - model: mistralai/Mistral-Nemo-Base-2407
      layer_range: [28, 32]
      parameters:
        weight: 0.5
    - model: Lambent/Gilded-Arsenic-12B
      layer_range: [28, 32]
    - model: nbeerbower/mistral-nemo-bophades-12B
      layer_range: [28, 32]
    - model: nbeerbower/Mistral-Nemo-Gutenberg-Doppel-12B-v2
      layer_range: [28, 32]
    - model: mergekit-community/MN-Hekate-Limenoskopos-17B
      layer_range: [32, 36]
    - model: mergekit-community/MN-Hekate-Limenoskopos-17B
      layer_range: [44, 48]

  - sources:
    - model: mergekit-community/MN-Hekate-Noctiluca-12B-v2
      layer_range: [32, 40]
      parameters:
        weight: 2
    - model: mergekit-community/MN-Sappho-j-12B
      layer_range: [32, 40]
    - model: mistralai/Mistral-Nemo-Base-2407
      layer_range: [32, 40]
      parameters:
        weight: 0.5
    - model: Lambent/Gilded-Arsenic-12B
      layer_range: [32, 40]
    - model: nbeerbower/mistral-nemo-bophades-12B
      layer_range: [32, 40]
    - model: nbeerbower/mistral-nemo-gutenberg-12B-v4
      layer_range: [32, 40]
    - model: mergekit-community/MN-Hekate-Limenoskopos-17B
      layer_range: [48, 56]
      parameters:
        weight: 3

tokenizer:
  source: mergekit-community/MN-Hekate-Noctiluca-12B-v2