Llama3.1-8B-PlumMath / mergekit_config.yml
sequelbox's picture
Upload folder using huggingface_hub (#1)
b857c30 verified
raw
history blame
337 Bytes
merge_method: della
dtype: bfloat16
parameters:
normalize: true
models:
- model: ValiantLabs/Llama3.1-8B-ShiningValiant2
parameters:
density: 0.5
weight: 0.3
- model: ValiantLabs/Llama3.1-8B-Cobalt
parameters:
density: 0.5
weight: 0.2
merge_method: della
base_model: meta-llama/Llama-3.1-8B-Instruct