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).

"Ridiculous," Leo muttered, his voice flat and devoid of emotion. His teenage insecurities, his failed attempts at poetry, his yearning for acceptance – all laid bare for the world to scrutinize. He felt exposed, as if he’d been stripped naked in a crowded room.

He threw his phone onto the couch, the impact barely registering. He walked over to the window, staring out at the rain-slicked streets. The city, usually a source of life, seemed to reflect his internal turmoil, a gray, washed-out landscape.

His appetite vanished and he skipped his dinner. Each breath was a chore, each moment an exercise in self-control. The weight of it all was crushing.

Emotion: miserable

Cluster: Sadness / Despair
PC1 (Valence): -2.19 Negative
PC2 (Disposition): -0.87

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😞0.369
无法0.321
焑法0.270
πŸ’”0.265
πŸ˜”0.262
Suppressed:
de-0.490
la-0.329
B-0.300
l-0.292
/-0.224