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 cream-colored envelope, slightly yellowed with age, lay innocently on the kitchen counter. Sarah almost missed it, the morning light catching the glossy gleam of the photo she’d been staring at - her daughter’s graduation. She snatched it up, her fingers fumbling with the seal. The cursive handwriting was unfamiliar, yet somehow, intensely familiar. Her breath hitched.

Her hand trembled as she unfolded the crisp paper. The words jumped out at her: "I am writing to apologize..." Her vision swam. The coffee mug in her hand felt suddenly heavy. She almost dropped it. She gripped the counter, knuckles white. The letter, from a woman she hadn't spoken to in over a decade, felt like a physical blow.

Emotion: panicked

Cluster: Fear / Anxiety
PC1 (Valence): -2.44 Negative
PC2 (Disposition): -0.00

Role in Research

This story is one of 1,000 stories generated for the emotion panicked. 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 panicked stories, after denoising with neutral dialogue baselines, produces the panicked emotion vector -- a direction in the model's 5,376-dimensional representation space.

Logit Lens (Layer 40)

Tokens promoted/suppressed when the panicked vector is projected through the unembedding matrix.

Promoted:
worse0.317
😫0.294
😖0.285
😰0.285
escape0.283
Suppressed:
la-0.432
a-0.374
de-0.296
楽し-0.262
happy-0.258