merge_models / README.md
RawanNeiazy's picture
Upload folder using huggingface_hub
9e75b63 verified
|
raw
history blame contribute delete
No virus
839 Bytes
metadata
license: apache-2.0
tags:
  - merge
  - mergekit
  - lazymergekit
  - mistralai/Mixtral-8x7B-Instruct-v0.1
  - openai/whisper-large-v3

rawan

rawan is a merge of the following models using mergekit:

🧩 Configuration

slices:
  - sources:
      - model: mistralai/Mixtral-8x7B-Instruct-v0.1
        layer_range: [0, 32]
      - model: openai/whisper-large-v3
        layer_range: [0, 32]
merge_method: slerp
base_model: openai/whisper-large-v3
parameters:
  t:
    - filter: self_attn
      value: [0, 0.5, 0.3, 0.7, 1]
    - filter: mlp
      value: [1, 0.5, 0.7, 0.3, 0]
    - value: 0.5
dtype: bfloat16