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 notification popped up on her phone. Her essay, "Finding Harmony in Chaos," was part of a new online guide for applicants. Her breath caught in her throat. She closed the app. She didn't want to look.

Later, she found herself opening it, drawn by a morbid curiosity. Her essay, now a public document, seemed flimsy, almost childish. She winced at her overly flowery prose, the obvious attempts at profundity.

She felt a flush creep up her neck. She pushed away from her desk, needing to move. She spent the next hour cleaning her apartment, scrubbing surfaces until they gleamed, a futile attempt to impose order on the chaos within.

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