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 rain hammered against the bus window, mirroring the relentless drumming in Leo's head. He slumped low in his seat, his shoulders hunched, the worn fabric of his coat clinging damply. Another rejection email, another week closer to losing his apartment. He'd been trading witty barbs online with "StarlightWhisper" for months, a much-needed escape from the bleak reality of his job search. They'd talked about everything - favorite books, dreams, even the terrible weather. Now, staring at the familiar brick building of his bus stop, he saw a familiar face, a woman with bright pink hair – StarlightWhisper, or rather, the girl he knew as Sarah, from the online chats. He couldn't meet her gaze, his chest suddenly tight.

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
無法0.270
💔0.265
😔0.262
Suppressed:
de-0.490
la-0.329
B-0.300
l-0.292
/-0.224