ModelFarm · pennywise-1.5b-engram-evoked

Steering-vector (engram) persona package on Qwen/Qwen2.5-1.5B-Instruct, produced by the Arnold Persona Lab persona-stack harness. Persona: Pennywise.

What this is

A difference-in-means steering vector (persona engram) read from the persona corpus and injected into the residual stream at a validated layer, dose-normalized to the layer's median residual norm.

Usage

pip install torch transformers
python farm_runtime.py . --chat          # or: python farm_runtime.py . "Who are you?"
python farm_runtime.py . --alpha 0.5 "Who are you?"   # dose override

Persona-stack measurements

identity	catch/100w	register	QA	neutral NLL
0	0.0	0.0	0.0	4.426

Scoring: regex/lexicon judge, symmetric across arms — arm-vs-arm deltas are meaningful, absolute numbers are indicative. QA = 5-question factual mini-suite. NLL = token loss on a neutral corpus (higher = more capability damage).

Dose & safety

• alpha 0.5-1.0 is the active range; above ~1.5 the model degrades (see the dose scans in the lab README).
• Steers style/register; identity claims do not switch on (the identity gate finding) — expect a tone shift, not a character swap.

Provenance & rights

• Fictional-character likeness study. All interview text is ORIGINAL writing in the character's style, created for this research. Not affiliated with, endorsed by, or drawn from the rights holders' works or dialogue. Gated as research artifact; not a product.
• Trained/derived by the Arnold Persona Lab harness, Qwen/Qwen2.5-1.5B-Instruct base.

ModelFarm packages are research artifacts. Real-person personas are likeness studies for interpretability research, not impersonation products: no commercial use, do not represent outputs as statements by the real person. Prefer fictional personas for public roleplay use.

Method note for steering packages

Steering packages do not modify weights: the base model is loaded unchanged and a runtime hook applies the persona direction / neuron recipe at inference. Dose can be varied live (--alpha / --dose), and the base model stays swappable for equal-size checkpoints of the same family.