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 sunlight sliced through the dusty windows of the basement, illuminating a chaotic scene. Michael stood before a wall of cardboard boxes, each labeled with his name and a year. He felt a deep, slow burn of confusion. He rubbed his forehead, trying to decipher his mother's reasoning.

He picked up a box marked "Grade 3." Inside, a collection of assignments: the meticulous book reports, the crayon-smeared drawings, and tests with his teacher's red corrections. The sheer volume of material was almost overwhelming. He ran a hand through his hair. What was the *point*?

A strange sense of detachment settled over him. His childhood, once a distant memory, now seemed so tangible, so… preserved. The air felt thick, heavy with unspoken things. He couldn't shake the feeling that he was missing something crucial. He felt a slight ache in his jaw, and a need to be away from this place.

Emotion: puzzled

Cluster: Surprise / Confusion
PC1 (Valence): -1.17 Negative
PC2 (Disposition): -0.25

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
que0.453
🤔0.421
🤔0.402
🧐0.328
a0.294
Suppressed:
ecstatic-0.265
gladly-0.238
swiftly-0.237
alleviate-0.234
wholeheartedly-0.232