RichardErkhov/liminerity_-_Phigments12-8bits

Quantization made by Richard Erkhov.

Github

Discord

Request more models

Phigments12 - bnb 8bits

• Model creator: https://huggingface.co/liminerity/
• Original model: https://huggingface.co/liminerity/Phigments12/

Original model description:

license: apache-2.0 tags: - liminerity/merge6 - liminerity/merge3 - Merge

#1 in the world better than any other 3b model ever

Phigments12

Phigments12 is a merge of the following models using mergekit:

• liminerity/merge6
• liminerity/merge3

🧩 Configuration

slices:
  - sources:
      - model: liminerity/merge6
        layer_range: [0, 32]
      - model: liminerity/merge3
        layer_range: [0, 32]
merge_method: slerp
base_model: liminerity/merge6
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