TareksLab/Synthesis-V1-LLaMa-70B

merge

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 huihui-ai/Llama-3.3-70B-Instruct-abliterated as a base.

Models Merged

The following models were included in the merge:

• juvi21/Hermes-2-Theta-L3-Euryale-Ties-0.8-70B
• TheDrummer/Fallen-Llama-3.3-R1-70B-v1
• SicariusSicariiStuff/Negative_LLAMA_70B
• nbeerbower/Llama-3.1-Nemotron-lorablated-70B
• Sao10K/L3-70B-Euryale-v2.1

Configuration

The following YAML configuration was used to produce this model:

models:
  - model: TheDrummer/Fallen-Llama-3.3-R1-70B-v1
    parameters:
       select_topk: 0.15
  - model: nbeerbower/Llama-3.1-Nemotron-lorablated-70B
    parameters:
       select_topk: 0.15
  - model: juvi21/Hermes-2-Theta-L3-Euryale-Ties-0.8-70B
    parameters:
       select_topk: 0.15
  - model: Sao10K/L3-70B-Euryale-v2.1
    parameters:
       select_topk: 0.15
  - model: SicariusSicariiStuff/Negative_LLAMA_70B 
    parameters:
       select_topk: 0.15
merge_method: sce
base_model: huihui-ai/Llama-3.3-70B-Instruct-abliterated
parameters:
  int8_mask: true
chat_template: llama3
tokenizer:
  source: base
dtype: bfloat16