DeepthoughtSlerp2-8B / mergekit_config.yml
allknowingroger's picture
Upload folder using huggingface_hub
cd76987 verified
raw
history blame contribute delete
310 Bytes
models:
- model: ruliad/deepthought-8b-llama-v0.01-alpha
- model: meditsolutions/Llama-3.1-MedIT-SUN-8B
merge_method: slerp
base_model: ruliad/deepthought-8b-llama-v0.01-alpha
dtype: bfloat16
parameters:
t: [0, 0.5, 1, 0.5, 0] # V shaped curve: Hermes for input & output, WizardMath in the middle layers