Llama_evo_3_2
Llama_evo_3_2 is a merged model generated for Model Kinship experiments, originating from meta-llama/Llama-2-7b-hf
🧩 Configuration
slices:
- sources:
- model: PotatoB/Llama_evo_2_1
layer_range: [0, 32]
- model: PotatoB/Llama_evo_1_2
layer_range: [0, 32]
merge_method: slerp
base_model: PotatoB/Llama_evo_2_1
parameters:
t:
- filter: self_attn
value: [0, 0.5, 0.3, 0.7, 1]
- filter: mlp
value: [1, 0.5, 0.7, 0.3, 0]
- value: 0.5
dtype: float16
- Downloads last month
- 0