DiscoResearch/Llama3-DiscoLeo-8B-DARE-Experimental-4bit-awq

Llama3_DiscoLeo_8B_DARE_Experimental_4bit_awq_glc

This model is a 4 bit quantization of DiscoResearch/Llama3_DiscoLeo_8B_DARE_Experimental created using AutoAWQ with a custom bilingual calibration dataset and quant_config = {"zero_point": True, "q_group_size": 128, "w_bit": 4, "version": "GEMM"}.

Copy of the original model card:

DiscoResearch/Llama3_German_8B_v0.1 is a large language model based on Meta's Llama3-8B. It is specialized for the German language through continuous pretraining on 65 billion high-quality tokens, similar to previous LeoLM or Occiglot models.

This is a merge of our instruct model with the Instruct model by Meta. Created using mergekit. Contributed by Damian B.!

Merge Details

Merge Method

This model was merged using the DARE TIES merge method using meta-llama/Meta-Llama-3-8B as a base.

Models Merged

The following models were included in the merge:

• DiscoResearch/Llama3_DiscoLeo_Instruct_8B_v0.1
• meta-llama/Meta-Llama-3-8B-Instruct

Configuration

The following YAML configuration was used to produce this model:

models:
  - model: DiscoResearch/Llama3_DiscoLeo_Instruct_8B_v0.1
    parameters:
      density: 0.5
      weight: 0.5
  - model: meta-llama/Meta-Llama-3-8B-Instruct
    parameters:
      density: 0.5
      weight: 0.5
merge_method: dare_ties
base_model: meta-llama/Meta-Llama-3-8B
parameters:
  normalize: true
  int8_mask: false
dtype: bfloat16