Downloads · 30 days
28
15% of all-time downloads
Snqwii/Bigger-Personality-12B
Bigger-Personality-12B is a text generation model from Snqwii. Use it when you need the model to write or continue text. It is set up for transformers.
This is a merge of pre-trained language models created using mergekit.
Downloads · 30 days
28
15% of all-time downloads
All-time downloads
184
Public
Parameters
12.2B
24.5 GB on disk
Likes
3
Public
Click a slice to open those files.
.safetensors24.5 GB · 100%
From the Hugging Face model README
This is a merge of pre-trained language models created using mergekit.
Harness the 𝓯𝓻𝓮𝓪𝓴𝔂 power of allura-org/Bigger-Body-12b, with the amazing intelligence of PocketDoc/Dans-PersonalityEngine-V1.3.0-12b
It's 𝓯𝓻𝓮𝓪𝓴𝔂, but it isn't hypersensitive to the greeting. It's intelligent, but filthy.
A Bigger Personality.
This is a likely NSFW-biased roleplaying model that was primarily merged to eliminate Bigger-Body's flaw of initial message sensitivity. Dans-PersonalityEngine will still heavily improve story-telling and roleplaying capability due to its smarts.
Hopefully either DanChat-2 or Mistral instruct format should work. Try DanChat-2 first.
Have fun, and let me know (in tasteful detail) how it goes!
After maining this model for a while, Bigger-Personality seems to act very closely/near identically to Dan's PersonalityEngine in writing prose and intelligence.
I have no 'feel' for Bigger-Body beyond that it mimics the initial message very closely, so I have no knowledge on what knowledge traits/prose quirks to look for.
Like Dan's, BP doesn't fall into repetitive comma splices ({{char}} slumps into the chair, sinking into the leather seating.) and demonstrates Dan's-level intelligence, writing, and improv ability.
It also has Dan's strong instruction adherence and neutral/little positivity bias.
Ultimately, the objective was to make a model that has an "internal" writing style to handle mediocre/no character cards, and this succeeded at that.
This model was merged using the NuSLERP merge method.
The following models were included in the merge:
The following YAML configuration was used to produce this model:
name: Bigger-Personality
merge_method: nuslerp
models:
- model: PocketDoc/Dans-PersonalityEngine-V1.3.0-12b
parameters:
weight: 0.6
- model: allura-org/Bigger-Body-12b
parameters:
weight: 0.4