RichardErkhov/legacymodel_-_Badudu-8B-instruct-v1-TIES-Prom-8bits

Quantization made by Richard Erkhov.

Github

Discord

Request more models

Badudu-8B-instruct-v1-TIES-Prom - bnb 8bits

• —Model creator: https://huggingface.co/legacymodel/
• —Original model: https://huggingface.co/legacymodel/Badudu-8B-instruct-v1-TIES-Prom/

Original model description: --- base_model:

• —gmonsoon/basemodel-8B-instruct
• —djuna/L3.1-PromissumMane-8B-Della-1.5-calc libraryname: transformers tags:
• —mergekit
• —merge

merge

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

Merge Details

Merge Method

This model was merged using the TIES merge method using gmonsoon/basemodel-8B-instruct as a base.

Models Merged

The following models were included in the merge:

• —djuna/L3.1-Promissum_Mane-8B-Della-1.5-calc

Configuration

The following YAML configuration was used to produce this model:

models:
  - model: gmonsoon/basemodel-8B-instruct
    #no parameters necessary for base model
  - model: gmonsoon/basemodel-8B-instruct
    parameters:
      density: 1
      weight: 1
  - model: djuna/L3.1-Promissum_Mane-8B-Della-1.5-calc
    parameters:
      density: 1
      weight: 1

merge_method: ties
base_model: gmonsoon/basemodel-8B-instruct
parameters:
  normalize: false
  int8_mask: true
dtype: float16