metadata

base_model:
  - djuna/Q2.5-Veltha-14B-0.5
  - qingy2019/Qwen2.5-Math-14B-Instruct
  - CultriX/Qwen2.5-14B-Brocav7
  - CultriX/SeQwence-14Bv1
  - CultriX/Qwen2.5-14B-Wernickev3
  - allknowingroger/QwenSlerp6-14B
  - CultriX/Qwen2.5-14B-Broca
  - CultriX/Qwen2.5-14B-Brocav3
library_name: transformers
tags:
  - mergekit
  - merge

merge

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

Merge Details

Merge Method

This model was merged using the della_linear merge method using CultriX/Qwen2.5-14B-Wernickev3 as a base.

Models Merged

The following models were included in the merge:

Configuration

The following YAML configuration was used to produce this model:

merge_method: della_linear
base_model: CultriX/Qwen2.5-14B-Wernickev3
dtype: bfloat16
parameters:
  epsilon: 0.01
  lambda: 1.7
  normalize: true
adaptive_merge_parameters:
  task_weights:
    tinyArc: 1.9
    tinyHellaswag: 1.8
    tinyMMLU: 1.85
    tinyTruthfulQA: 2.2
    tinyTruthfulQA_mc1: 2.0
    tinyWinogrande: 2.1
    IFEval: 2.1
    BBH: 2.0
    MATH: 2.4
    GPQA: 2.1
    MUSR: 2.2
    MMLU-PRO: 2.0
  smoothing_factor: 0.07
gradient_clipping:
  CultriX/Qwen2.5-14B-Wernickev3: 0.85
  djuna/Q2.5-Veltha-14B-0.5: 0.87
  qingy2019/Qwen2.5-Math-14B-Instruct: 0.89
  CultriX/SeQwence-14Bv1: 0.83
  CultriX/Qwen2.5-14B-Broca: 0.85
  CultriX/Qwen2.5-14B-Brocav3: 0.84
  CultriX/Qwen2.5-14B-Brocav7: 0.85
  allknowingroger/QwenSlerp6-14B: 0.83
models:
  - model: CultriX/Qwen2.5-14B-Wernickev3
    parameters:
      weight: 0.22
      density: 0.75
  - model: djuna/Q2.5-Veltha-14B-0.5
    parameters:
      weight: 0.2
      density: 0.73
  - model: qingy2019/Qwen2.5-Math-14B-Instruct
    parameters:
      weight: 0.18
      density: 0.72
  - model: CultriX/SeQwence-14Bv1
    parameters:
      weight: 0.16
      density: 0.71
  - model: CultriX/Qwen2.5-14B-Broca
    parameters:
      weight: 0.17
      density: 0.73
  - model: CultriX/Qwen2.5-14B-Brocav3
    parameters:
      weight: 0.15
      density: 0.72
  - model: CultriX/Qwen2.5-14B-Brocav7
    parameters:
      weight: 0.14
      density: 0.72
  - model: allknowingroger/QwenSlerp6-14B
    parameters:
      weight: 0.13
      density: 0.71
tokenizer_source: CultriX/Qwen2.5-14B-Wernickev3