Llama3-OneForAll-8B

Llama3-OneForAll-8B is a merge of the following models using mergekit:

🧩 Configuration

models:
  - model: bunnycore/RemixLlama3-8B
  - model: NousResearch/Hermes-2-Theta-Llama-3-8B
  - model: cognitivecomputations/dolphin-2.9-llama3-8b
merge_method: model_stock
base_model: cognitivecomputations/dolphin-2.9-llama3-8b
dtype: bfloat16