RichardErkhov/alnrg2arg_-_test2-4bits

Quantization made by Richard Erkhov.

Github

Discord

Request more models

test2 - bnb 4bits

• Model creator: https://huggingface.co/alnrg2arg/
• Original model: https://huggingface.co/alnrg2arg/test2/

Original model description:

license: cc-by-nc-4.0 tags: - merge - mergekit - lazymergekit - udkai/Turdus - fblgit/UNA-TheBeagle-7b-v1

test2

test2 is a merge of the following models using mergekit:

• udkai/Turdus
• fblgit/UNA-TheBeagle-7b-v1

🧩 Configuration

slices:
  - sources:
      - model: udkai/Turdus
        layer_range: [0, 32]
      - model: fblgit/UNA-TheBeagle-7b-v1
        layer_range: [0, 32]
merge_method: slerp
base_model: udkai/Turdus
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