merged_super_mario

Name: Codeus-7B-Pre-Alpha
Rating: 2.5 (8 reviews)

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

Merge Details

Merge Method

This model was merged using the DARE TIES merge method using Qwen/Qwen2.5-Coder-7B as a base.

Models Merged

The following models were included in the merge:

microsoft/NextCoder-7B
TIGER-Lab/VisCoder2-7B
DeepHat/DeepHat-V1-7B
open-r1/OlympicCoder-7B

Configuration

The following YAML configuration was used to produce this model:

yaml

models:
  - model: open-r1/OlympicCoder-7B
    parameters:
      density: 0.3
      weight: 0.35
  - model: microsoft/NextCoder-7B
    parameters:
      density: 0.25
      weight: 0.3
  - model: TIGER-Lab/VisCoder2-7B
    parameters:
      density: 0.25
      weight: 0.3
  - model: DeepHat/DeepHat-V1-7B
    parameters:
      density: 0.2
      weight: 0.2
merge_method: dare_ties
base_model: Qwen/Qwen2.5-Coder-7B
parameters:
  int8_mask: true
dtype: bfloat16
tokenizer_source: base

Upcycle-AI/Codeus-7B-Pre-Alpha

merged_super_mario

Merge Details

Merge Method

Models Merged

Configuration

Upcycle-AI/
Codeus-7B-Pre-Alpha