Magnum-Instruct-DPO-12B / mergekit_config.yml
ParasiticRogue's picture
Upload 11 files
59ddd90
raw
history blame
351 Bytes
models:
- model: mistral-nemo-gutenberg-12B-v3
parameters:
weight: 0.5
density: 0.8
- model: mistral-nemo-bophades-12B
parameters:
weight: 0.5
density: 0.8
merge_method: della_linear
base_model: Mistral-Nemo-Base-2407
parameters:
epsilon: 0.05
lambda: 1
int8_mask: true
dtype: bfloat16
tokenzer_source: union