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).

Maya stared out the window, watching the rain blur the city lights. She felt a profound emptiness, a hollowness that even the comfort of her own home couldn't fill. Her phone buzzed. It was a message from "StarryNight," her online pen pal, who lived "somewhere in the midwest." They'd been exchanging letters, sharing their hopes and fears, for almost a year.

The message read: "I'm at the library on Pine Street. Come on down if you'd like. There's a poetry reading." Maya’s breath caught in her throat. Pine Street? She knew the library. She walked by it everyday.

She hesitated, then slowly typed back: "What time are you leaving?" Her life, or at least her mood, was about to change.

Emotion: droopy

Cluster: Fatigue / Lethargy
PC1 (Valence): -1.15 Negative
PC2 (Disposition): -1.23

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😞0.303
无法0.252
いつも0.238
😔0.230
:(0.228
Suppressed:
l-0.360
P-0.356
B-0.339
de-0.334
!-0.239