const_tag: | |
- &SFU 0.7071067812 # Upper scaling factor (1/sqrt(2)) | |
- &SFL 0.85 # Lower scaling factor | |
- &SFMLP 0.8 # MLP scaling factor | |
attenuate-env-lower: | |
parameters: | |
scale: | |
- filter: q_proj | |
value: | |
- filter: k_proj | |
value: | |
- filter: v_proj | |
value: | |
- filter: o_proj | |
value: | |
- filter: gate_proj | |
value: | |
- filter: up_proj | |
value: | |
- filter: down_proj | |
value: | |
attenuate-env-upper: | |
parameters: | |
scale: | |
- filter: q_proj | |
value: | |
- filter: k_proj | |
value: | |
- filter: v_proj | |
value: | |
- filter: o_proj | |
value: | |
- filter: gate_proj | |
value: | |
- filter: up_proj | |
value: | |
- filter: down_proj | |
value: | |
slices: | |
# Preserve input layers | |
- sources: | |
- model: SvalTek/ColdBrew-Aphid | |
layer_range: [0, 12] | |
# Early Expansion (Duplicate Twice) | |
- sources: | |
- model: SvalTek/ColdBrew-Aphid | |
layer_range: [13, 15] | |
- sources: | |
- model: SvalTek/ColdBrew-Aphid | |
layer_range: [13, 15] | |
<<: | |
- sources: | |
- model: SvalTek/ColdBrew-Aphid | |
layer_range: [13, 15] | |
<<: | |
# Late Expansion (Duplicate Twice) | |
- sources: | |
- model: SvalTek/ColdBrew-Aphid | |
layer_range: [15, 18] | |
- sources: | |
- model: SvalTek/ColdBrew-Aphid | |
layer_range: [15, 18] | |
<<: | |
- sources: | |
- model: SvalTek/ColdBrew-Aphid | |
layer_range: [15, 18] | |
<<: | |
# Preserve output layers | |
- sources: | |
- model: SvalTek/ColdBrew-Aphid | |
layer_range: [19, 27] | |
merge_method: passthrough | |
dtype: bfloat16 | |
parameters: | |
normalize: true | |
int8_mask: true | |