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 air inside the pawn shop was thick with the scent of old things and unspoken stories. Anya felt like she was wading through treacle, each step a struggle. The days were long, the nights even longer. She felt trapped in an endless cycle.

She had to get out of her apartment, which had become a prison. She had to *do* something. So she went to the pawn shop, a place where she knew she wouldn’t be judged.

She wasn't looking for anything, until she saw it. A familiar gold band, inscribed with a meaningful date. Her grandmother’s wedding ring. Her heart hammered against her ribs, a frantic drumbeat.

She closed her eyes for a moment, the memory of her grandmother, the ring, the happiness, a harsh contrast to the emptiness she felt. She took a shuddering breath.

“How much?” she asked.

Emotion: depressed

Cluster: Sadness / Despair
PC1 (Valence): -2.06 Negative
PC2 (Disposition): -1.08

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😞0.417
每天0.353
无法0.330
own0.311
😔0.309
Suppressed:
de-0.492
B-0.393
P-0.352
l-0.310
!-0.307