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).

"We regret to inform you..." The sterile words burned on Sarah’s retina. She reread the rejection email a dozen times, each reading confirming the devastating truth. The world seemed to shrink around her, the vibrant energy of the campus now a distant hum.

She felt a hollowness spreading through her, a void where hope had been. The thought of returning to her mundane routine, the job at the diner, the cramped apartment, filled her with a profound sense of despair. Her future suddenly seemed bleak.

She gripped the strap of her backpack, her knuckles white. The weight of her belongings felt immense, a physical manifestation of the expectations she felt crushing her. She needed something.

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