CoolFace
Modelpublic

miraiminds/evolutionary-mergingLLM-Hermes-Dans-zephyr

sourceHugging Faceupdated 2y agoView on Hugging Face
0likes10downloads
Model Card

final_merge

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

Merge Details

Merge Method

This model was merged using the task arithmetic merge method using /home/evolmergestorage/inputmodels/Mistral-7B-v0.18133861 as a base.

Models Merged

The following models were included in the merge:

  • —/home/evolmergestorage/inputmodels/zephyr-7b-beta2449712360
  • —/home/evolmergestorage/inputmodels/Hermes-2-Pro-Mistral-7B2793206805
  • —/home/evolmergestorage/inputmodels/Dans-AdventurousWinds-Mk2-7b1152917843

Configuration

The following YAML configuration was used to produce this model:

yaml
base_model: /home/evol_merge_storage/input_models/Mistral-7B-v0.1_8133861
dtype: bfloat16
merge_method: task_arithmetic
parameters:
  int8_mask: 1.0
  normalize: 0.0
slices:
- sources:
  - layer_range: [0, 8]
    model: /home/evol_merge_storage/input_models/Hermes-2-Pro-Mistral-7B_2793206805
    parameters:
      weight: 0.45632162073777516
  - layer_range: [0, 8]
    model: /home/evol_merge_storage/input_models/Dans-AdventurousWinds-Mk2-7b_1152917843
    parameters:
      weight: 0.6468013824217197
  - layer_range: [0, 8]
    model: /home/evol_merge_storage/input_models/zephyr-7b-beta_2449712360
    parameters:
      weight: 0.04013732886365818
  - layer_range: [0, 8]
    model: /home/evol_merge_storage/input_models/Mistral-7B-v0.1_8133861
- sources:
  - layer_range: [8, 16]
    model: /home/evol_merge_storage/input_models/Hermes-2-Pro-Mistral-7B_2793206805
    parameters:
      weight: 0.39991499422381294
  - layer_range: [8, 16]
    model: /home/evol_merge_storage/input_models/Dans-AdventurousWinds-Mk2-7b_1152917843
    parameters:
      weight: 0.3033908462486953
  - layer_range: [8, 16]
    model: /home/evol_merge_storage/input_models/zephyr-7b-beta_2449712360
    parameters:
      weight: 0.3250623744211742
  - layer_range: [8, 16]
    model: /home/evol_merge_storage/input_models/Mistral-7B-v0.1_8133861
- sources:
  - layer_range: [16, 24]
    model: /home/evol_merge_storage/input_models/Hermes-2-Pro-Mistral-7B_2793206805
    parameters:
      weight: 0.6202885707451035
  - layer_range: [16, 24]
    model: /home/evol_merge_storage/input_models/Dans-AdventurousWinds-Mk2-7b_1152917843
    parameters:
      weight: 0.235874597687366
  - layer_range: [16, 24]
    model: /home/evol_merge_storage/input_models/zephyr-7b-beta_2449712360
    parameters:
      weight: 0.27733993810853536
  - layer_range: [16, 24]
    model: /home/evol_merge_storage/input_models/Mistral-7B-v0.1_8133861
- sources:
  - layer_range: [24, 32]
    model: /home/evol_merge_storage/input_models/Hermes-2-Pro-Mistral-7B_2793206805
    parameters:
      weight: 0.5484088078896818
  - layer_range: [24, 32]
    model: /home/evol_merge_storage/input_models/Dans-AdventurousWinds-Mk2-7b_1152917843
    parameters:
      weight: 0.876059424739865
  - layer_range: [24, 32]
    model: /home/evol_merge_storage/input_models/zephyr-7b-beta_2449712360
    parameters:
      weight: 0.26196309089548436
  - layer_range: [24, 32]
    model: /home/evol_merge_storage/input_models/Mistral-7B-v0.1_8133861