metadata

license: apache-2.0
tags:
  - merge
  - mergekit
  - ties
  - 7B
  - eren23/OGNO-7b-dpo-truthful
  - Kquant03/NeuralTrix-7B-dpo-laser

FMixIA-7B-TIES-1

A merged model using Task Interpolation for Efficient Scaling (TIES) using mergekit.

Model Details

Base Models:
- eren23/OGNO-7b-dpo-truthful
- Kquant03/NeuralTrix-7B-dpo-laser
Merge Method: ties

Configuration

models:
  - model: eren23/OGNO-7b-dpo-truthful
  - model: Kquant03/NeuralTrix-7B-dpo-laser
    parameters:
      density: 0.5
      weight: 0.5
merge_method: ties
base_model: eren23/OGNO-7b-dpo-truthful
parameters:
  normalize: true
dtype: float16

Usage

This model can be used with the standard transformers library:

from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained("Ro-xe/FMixIA-7B-TIES-1")
tokenizer = AutoTokenizer.from_pretrained("Ro-xe/FMixIA-7B-TIES-1")