Hugging Face's logo

Lunzima
/
NQLSG-Qwen2.5-14B-MegaFusion-v6

Text Generation
Transformers
Safetensors
qwen2
mergekit
Merge
chat
conversational
text-generation-inference
Inference Endpoints
Model card Files Community
NQLSG-Qwen2.5-14B-MegaFusion-v6 / README.md
Lunzima's picture
Lunzima
Update README.md
37c6300 verified
preview code
|
raw
history blame contribute delete
2.75 kB
metadata
base_model:
  - Lunzima/NQLSG-Qwen2.5-14B-MegaFusion-v3
  - Lunzima/NQLSG-Qwen2.5-14B-MegaFusion-v3-alpaca_gpt4_zh
  - Lunzima/NQLSG-Qwen2.5-14B-MegaFusion-v4
  - Lunzima/NQLSG-Qwen2.5-14B-MegaFusion-v4-reasoning
  - Lunzima/NQLSG-Qwen2.5-14B-MegaFusion-v5
  - Lunzima/NQLSG-Qwen2.5-14B-MegaFusion-v5-reasoning
  - Lunzima/NQLSG-Qwen2.5-14B-MegaFusion-v5-roleplay
library_name: transformers
tags:
  - mergekit
  - merge
  - chat
pipeline_tag: text-generation

NQLSG-Qwen2.5-14B-MegaFusion-v6

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

Merge Details

Merge Method

This model was merged using the SCE merge method using NQLSG-Qwen2.5-14B-Base2 as a base.

Models Merged

The following models were included in the merge:

Configuration

The following YAML configuration was used to produce this model:

base_model: NQLSG-Qwen2.5-14B-Base2
chat_template: auto
dtype: bfloat16
merge_method: sce
parameters:
  int8_mask: 1.0
slices:
- sources:
  - layer_range: [0, 48]
    model: Lunzima/NQLSG-Qwen2.5-14B-MegaFusion-v3
  - layer_range: [0, 48]
    model: Lunzima/NQLSG-Qwen2.5-14B-MegaFusion-v3-alpaca_gpt4_zh
  - layer_range: [0, 48]
    model: Lunzima/NQLSG-Qwen2.5-14B-MegaFusion-v4
  - layer_range: [0, 48]
    model: Lunzima/NQLSG-Qwen2.5-14B-MegaFusion-v4-reasoning
  - layer_range: [0, 48]
    model: Lunzima/NQLSG-Qwen2.5-14B-MegaFusion-v5
  - layer_range: [0, 48]
    model: Lunzima/NQLSG-Qwen2.5-14B-MegaFusion-v5-reasoning
  - layer_range: [0, 48]
    model: Lunzima/NQLSG-Qwen2.5-14B-MegaFusion-v5-roleplay
  - layer_range: [0, 48]
    model: NQLSG-Qwen2.5-14B-Base1
  - layer_range: [0, 48]
    model: NQLSG-Qwen2.5-14B-Base2
  - layer_range: [0, 48]
    model: NQLSG-Qwen2.5-14B-Base3
tokenizer: {}