New version of Franken-models using Mixstral?
#1
by
QuantumState745837
- opened
Hi there, so since we now have Mixstral and such, do you think it's probably best with we now revisit some of these odd 20B models and such and apply the mix?
slices:
- sources:
- model: Orca2flat
layer_range: [0, 16]
- model: Orca2flat
- sources:
- model: /KoboldAI/Psyfighter-2-13B (FP16 not yet available)
layer_range: [8, 24]
- model: /KoboldAI/Psyfighter-2-13B (FP16 not yet available)
- sources:
- model: Orca2flat
layer_range: [17, 32]
- model: Orca2flat
- sources:
- model: /KoboldAI/Psyfighter-2-13B (FP16 not yet available)
layer_range: [25, 40]
merge_method: passthrough
dtype: float16
- model: /KoboldAI/Psyfighter-2-13B (FP16 not yet available)
I don't know much about AI, I just thought if it was possible or something.