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 musty scent of the bookstore did little to soothe Eleanor. She'd chosen a biography of a forgotten poet, hoping the melancholy subject would match her mood. A worn copy, she'd thought, was just what she needed. As she flipped through the pages, a brittle piece of paper fluttered out. It was a letter, penned in elegant script. A woman's handwriting, addressed to a “Dearest Thomas.” Her stomach clenched. It was a beautiful letter, full of longing and regret for chances missed. Eleanor immediately felt a pang of shame. She hadn't finished the grant application. Again.

The letter's words about love and loss resonated with a hollowness inside her. She re-read a particular paragraph about a dance they never had. A wave of heat flushed her cheeks. How many opportunities had she let slip by? She crumpled the letter, then smoothed it out. The paper felt like a judgment in her trembling hands.

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