ECE-TW3-JRGL-V4 / README.md
CombinHorizon's picture
Update README.md
f7437da verified
|
raw
history blame
844 Bytes
metadata
license: apache-2.0
tags:
  - merge
  - mergekit
  - lazymergekit
  - migtissera/Tess-72B-v1.5b
  - MTSAIR/MultiVerse_70B

ECE-TW3-JRGL-V4

ECE-TW3-JRGL-V4 is a merge of the following models using mergekit:

🧩 Configuration

base_model: migtissera/Tess-72B-v1.5b
dtype: bfloat16
merge_method: slerp
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
slices:
- sources:
  - layer_range:
    - 0
    - 80
    model: migtissera/Tess-72B-v1.5b
  - layer_range:
    - 0
    - 80
    model: MTSAIR/MultiVerse_70B