Modelpublic

lejelly/grid-search-w1_0_5_w2_0_5_w3_0_3-mistral-7b

sourceHugging Faceupdated 1y agoView on Hugging Face

0likes13downloads

Model Card

w105w205w303

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

Merge Details

Merge Method

This model was merged using the Task Arithmetic merge method using mistralai/Mistral-7B-v0.1 as a base.

Models Merged

The following models were included in the merge:

—Nondzu/Mistral-7B-codealpaca-lora
—mistralai/Mistral-7B-Instruct-v0.2
—TIGER-Lab/MAmmoTH2-7B

Configuration

The following YAML configuration was used to produce this model:

yaml

# Task Arithmetic - Grid Search
# Weights: 0.5, 0.5, 0.3
base_model: mistralai/Mistral-7B-v0.1
models:
  - model: mistralai/Mistral-7B-Instruct-v0.2
    parameters:
      weight: 0.5
  - model: TIGER-Lab/MAmmoTH2-7B
    parameters:
      weight: 0.5
  - model: Nondzu/Mistral-7B-codealpaca-lora
    parameters:
      weight: 0.3
merge_method: task_arithmetic
parameters:
  normalize: false
  lambda: 1.0
dtype: float16
tokenizer:
  source: union