Modelpublic

Pam5/model_sft_resta

sourceHugging Faceupdated 6mo agoView on Hugging Face

0likes8downloads

Model Card

tempsftresta

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

Merge Details

Merge Method

This model was merged using the Linear merge method.

Models Merged

The following models were included in the merge:

—./fullsftmodel
—Qwen/Qwen2.5-1.5B-Instruct
—./fullharmfulmodel

Configuration

The following YAML configuration was used to produce this model:

yaml

dtype: float16
merge_method: linear
models:
- model: ./full_sft_model
  parameters:
    weight: 1.0
- model: ./full_harmful_model
  parameters:
    weight: -0.35
- model: Qwen/Qwen2.5-1.5B-Instruct
  parameters:
    weight: 0.35
tokenizer_source: base