tachytelicdetonation/llama3-8b-merge-test-sce-1x3

Untitled Model (1)

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 NousResearch/Hermes-3-Llama-3.1-8B as a base.

Models Merged

The following models were included in the merge:

• DreadPoor/Aspire-8B-model_stock
• deepseek-ai/DeepSeek-R1-Distill-Llama-8B
• allura-org/L3.1-8b-RP-Ink

Configuration

The following YAML configuration was used to produce this model:

# SCE (Select, Calculate, Erase) merge configuration

merge_method: sce
base_model: NousResearch/Hermes-3-Llama-3.1-8B
models:
  - model: allura-org/L3.1-8b-RP-Ink 
    parameters:
      weight: 1.0
  - model: DreadPoor/Aspire-8B-model_stock
    parameters:
      weight: 1.0
  #- model: TroyDoesAI/BlackSheep-X-Dolphin 
  #  parameters:
  #    weight: 1.0
  #- model: mlabonne/NeuralDaredevil-8B-abliterated
  #  parameters:
  #    weight: 1.0
  #- model: SicariusSicariiStuff/Wingless_Imp_8B
  #  parameters:
  #    weight: 1.0
  - model: deepseek-ai/DeepSeek-R1-Distill-Llama-8B
    parameters:
      weight: 1.0
parameters:
  select_topk: 0.4  
  density: 0.7
  lambda: 1.0
tokenizer:
  source: "union"
dtype: float16
chat_template: "chatml"