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’d poured over the presentation for weeks, staying late every night. Now, all that effort was for naught. David found himself staring at the congratulatory email chain announcing Chloe’s appointment. His stomach twisted, a cold knot forming in his gut.

He wanted to scream, to throw something. Instead, he simply closed his laptop with a decisive click, the sound echoing in the otherwise silent office. He needed to get out, away from the prying eyes, the forced pleasantries.

Outside, the city seemed to mock him. The sun was too bright, the air too crisp. He avoided eye contact, his shoulders hunched. He walked for hours, lost in the crowd, feeling smaller and smaller with each passing minute.

Emotion: humiliated

Cluster: Embarrassment
PC1 (Valence): -1.79 Negative
PC2 (Disposition): 0.93

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
C0.370
l0.329
尴尬0.322
ness0.303
because0.298
Suppressed:
de-0.369
own-0.254
😍-0.253
🤩-0.245
cautiously-0.228