File size: 1,241 Bytes
281fa5a |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 |
models:
- model: tokyotech-llm/Swallow-70b-NVE-instruct-hf
# no parameters necessary for base model
- model: GOAT-AI/GOAT-70B-Storytelling # storytelling
parameters:
density: 1
weight: 0.25
- model: dreamgen/opus-v0.5-70b # creative roleplay
parameters:
density: 1
weight: 0.25
merge_method: dare_ties
base_model: tokyotech-llm/Swallow-70b-NVE-instruct-hf
dtype: bfloat16
name: Swallow-70b-NVE-RP-base
---
models:
- model: tokyotech-llm/Swallow-70b-NVE-instruct-hf
# no parameters necessary for base model
- model: Doctor-Shotgun/lzlv-limarpv3-l2-70b # roleplay configuration
parameters:
density: 1
weight: 0.25
merge_method: dare_ties
base_model: tokyotech-llm/Swallow-70b-NVE-instruct-hf
dtype: bfloat16
name: Swallow-70b-NVE-RP-flavor
---
slices:
- sources:
- model: Swallow-70b-NVE-RP-base
layer_range: [0, 80]
- model: Swallow-70b-NVE-RP-flavor
layer_range: [0, 80]
merge_method: slerp
base_model: Swallow-70b-NVE-RP-base
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 # fallback for rest of tensors
dtype: bfloat16
name: Swallow-70b-NVE-RP
|