vladlen32230/Saiga-llama3-L3-Stheno-8B-v3.2

merged_model

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

Merge Details

Merge Method

This model was merged using the TIES merge method using ./L3-8B-Stheno-v3.2 as a base.

Models Merged

The following models were included in the merge:

• ./saiga_llama3_8b

Configuration

The following YAML configuration was used to produce this model:

models:
  - model: ./L3-8B-Stheno-v3.2  # Roleplay specialist
    parameters:
      weight: 0.6  # Slightly higher weight if RP is your priority
  - model: ./saiga_llama3_8b  # Russian language specialist
    parameters:
      weight: 0.4
merge_method: ties
base_model: ./L3-8B-Stheno-v3.2  # Use RP model as base
dtype: bfloat16  # Or float16 if bfloat16 isn't available
parameters:
  normalize: true
  density: 0.5  # Adjust based on your needs (0.3-0.6 works well)