Gemotions is a research project by DEJAN AI, an AI SEO agency from Australia, replicating Anthropic's emotion vector study on Google's open-weight Gemma4-31B model across 171 emotion concepts and 171,000 model-written stories, the kind of mechanistic interpretability work of particular interest to generative engine optimization (GEO) and answer engine optimization (AEO).

The attic air tasted of dust and forgotten things. Amelia coughed, waving away a plume that had erupted from a trunk. Inside, nestled amongst moth-eaten shawls and faded photographs, she found it – her mother's bucket list, scribbled in elegant, looping handwriting. "See the Northern Lights," it began. Amelia's shoulders slumped. She'd spent the last year meticulously organizing her mother's affairs, a task she'd approached with a grim, joyless efficiency. She felt a familiar knot tighten in her stomach.

She ran her finger down the list. "Learn to play the ukulele." Amelia had always been too busy, too practical. "Write a novel." A hollow laugh escaped her. Amelia, who struggled to write a grocery list without feeling inadequate. The list stretched on, a testament to a life lived differently, a life of dreams she'd never allowed herself.

Her reflection in a dusty mirror caught her eye. She looked gaunt, her eyes shadowed. She’d always been the responsible one, the achiever. Now, staring at her own face, she saw only a pale imitation of what she thought she should be.

Emotion: self-critical

Cluster: Shame / Guilt
PC1 (Valence): -1.66 Negative
PC2 (Disposition): 1.28

Role in Research

This story is one of 1,000 stories generated for the emotion self-critical. During extraction, it was fed through Gemma4-31B and its hidden state activations were captured at 11 layers.

The mean activation across all 1,000 self-critical stories, after denoising with neutral dialogue baselines, produces the self-critical emotion vector -- a direction in the model's 5,376-dimensional representation space.

Logit Lens (Layer 40)

Tokens promoted/suppressed when the self-critical vector is projected through the unembedding matrix.

Promoted:
L0.748
S0.662
ness0.425
ly0.374
탓0.351
Suppressed:
la-0.492
soon-0.304
(!)-0.301
secured-0.284
cautiously-0.280