A solid model, involving most of my favorite models merges started with the Smarteaz V1.0 series, albeit in a little bit dilluted fashion. It's a bit tame, but it's coherent, smart, makes long answers and respect quite well the demanded formatting.

merge

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

Merge Details

Merge Method

This model was merged using the Model Stock merge method using Nexesenex/Llama_3.x_70b_TrinitEva_V1.1 as a base.

Models Merged

The following models were included in the merge:

Configuration

The following YAML configuration was used to produce this model:

yaml

merge_method: model_stock
models:
  - model: Nexesenex/Llama_3.x_70b_Erasmus_V1.11
    parameters:
      weight: 1.0
  - model: Nexesenex/Llama_3.x_70b_FreeFaller_R1_V1.11
    parameters:
      weight: 1.0
base_model: Nexesenex/Llama_3.x_70b_TrinitEva_V1.1
dtype: bfloat16
out_dtype: bfloat16
parameters:
  int8_mask: true
  normalize: true
  rescale: false
chat_template: auto
tokenizer:
  source: union

Nexesenex/Llama_3.x_70b_PentEva_FreeEra_V1.11

about

merge

Merge Details

Merge Method

Models Merged

Configuration

Nexesenex/
Llama_3.x_70b_PentEva_FreeEra_V1.11