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 was leaning against the crumbling wall, sketching in his notebook, when the music started. Liam looked up, his hand freezing mid-stroke. The melody, the very notes he had painstakingly composed, echoed from the street performer's guitar. It was a song he had considered abandoning, a song that had held such a deeply personal meaning for him.

He felt a sudden, inexplicable heat in his cheeks. He squinted at the busker, a young man with a weathered hat and a worn-out guitar. The man was singing it, singing *his* words, but with a different cadence, a different inflection. Liam lowered his sketchbook, a sense of bewilderment spreading through him. He felt like he'd stepped into a parallel universe.

Emotion: perplexed

Cluster: Surprise / Confusion
PC1 (Valence): -1.42 Negative
PC2 (Disposition): -0.16

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
🤔0.330
L0.300
🤔0.296
S0.273
que0.264
Suppressed:
மகி-0.227
rest-0.222
ecstatic-0.212
alleviate-0.210
uplifting-0.196