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 familiar scent of old books and dust filled the library, but it couldn't soothe the unease churning within him. Across the table, Emily, whom he'd trusted, was meticulously reconstructing their shared history with their writing project, a clear distortion of his memory. She portrayed him as a hapless, easily led writer, constantly in need of her direction.

His face burned. He’d always regarded the collaboration as a balanced exchange of ideas. Now, Emily presented him as a minor player in his own life. The library's quiet seemed to amplify the clatter in his chest.

He pushed his chair back, the screech echoing through the silent room. He couldn't meet her eyes. He needed to get away, to breathe, to escape the suffocating weight of her selective memory.

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