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).

Rain lashed against the windows, mirroring the tempest inside Amelia. She huddled on the sofa, clutching a blanket like a lifeline. Sunlight, weak as it was, struggled to penetrate the grey. Mark's laptop, open on the coffee table, was her beacon. She'd found it there, open to a manuscript titled "Amelia." Curiosity, mingled with a terror she couldn't name, had seized her.

Her breath hitched as she scrolled through the pages. The descriptions of her – her nervous habit of biting her lip, the way she tilted her head when listening – were achingly familiar. The fictional Amelia was, however, far more vibrant, more confident, a woman she desperately wished she could be. Shame coiled in her stomach.

A car door slammed outside. Mark. Amelia scrambled to close the laptop, her fingers fumbling. The screen froze. She froze too, her heart drumming a frantic rhythm against her ribs.

Emotion: dependent

Cluster: Shame / Guilt
PC1 (Valence): -1.05 Negative
PC2 (Disposition): 0.69

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
de0.489
缺乏0.395
even0.310
either0.301
'0.297
Suppressed:
笑道-0.309
不久-0.293
一脸-0.263
затем-0.263
庆祝-0.258