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).

He stared at the "Welcome to the Future" sign plastered in the lobby, a sarcastic grin plastered on his face. The building’s management, always a bit shady, was now promising "luxury living." Ben ran a hand through his hair, a nervous gesture he hadn’t noticed he'd adopted.

He'd lived in this building for fifteen years, since his divorce. The cheap rent had been a lifeline. Now, the potential for being forced out felt like another blow. He lit a cigarette, the smoke curling around him, a temporary shield against the disquiet.

Ben walked up to his apartment, the steps creaking under his weight, the dim hall lit by a single bulb. The thought of finding a new place, the packing, the upheaval, made his head ache. He closed the door, the click echoing in the sudden silence.

Emotion: troubled

Cluster: Fear / Anxiety
PC1 (Valence): -2.20 Negative
PC2 (Disposition): -0.81

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😞0.343
S0.317
ness0.302
😔0.294
😢0.272
Suppressed:
de-0.553
la-0.511
l-0.307
/-0.260
(!)-0.236