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).

John sat frozen in his history class, the professor still blathering on about the French Revolution, oblivious to the bombshell that had just dropped. John had seen the notice on the university website just before class. He felt a sudden, intense pressure behind his eyes.

He tried to focus on the lecture, to take notes, but his hand wouldn't cooperate. His pen just drifted aimlessly across the paper, leaving erratic, meaningless scribbles. The room felt stifling, the air thick with unspoken fear.

He felt a deep tremor in his gut. His future, his aspirations, felt like a stack of dominoes about to topple. He rubbed his forehead, trying to clear the fog in his brain, but the image of the closing university was the only thing he could see.

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