Same reliable approach as before. A good RP model and a suitable dose of SimPO are a match made in heaven.
SillyTavern
Text Completion presets
temp 0.9
top_k 30
top_p 0.75
min_p 0.2
rep_pen 1.1
smooth_factor 0.25
smooth_curve 1
Advanced Formatting
Context & Instruct Presets for Gemma Here IMPORTANT !
Instruct Mode: Enabled
This is a merge of pre-trained language models created using mergekit.
Merge Method
This model was merged using the SLERP merge method.
Models Merged
The following models were included in the merge:
Configuration
The following YAML configuration was used to produce this model:
slices:
- sources:
- model: TheDrummer/Gemmasutra-9B-v1
layer_range: [0, 42]
- model: princeton-nlp/gemma-2-9b-it-SimPO
layer_range: [0, 42]
merge_method: slerp
base_model: TheDrummer/Gemmasutra-9B-v1
parameters:
t:
- filter: self_attn
value: [0.2, 0.4, 0.6, 0.2, 0.4]
- filter: mlp
value: [0.8, 0.6, 0.4, 0.8, 0.6]
- value: 0.4
dtype: bfloat16
Want to support my work ? My Ko-fi page: https://ko-fi.com/tannedbum