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 slammed the book shut, the sound echoing in the otherwise quiet library. John’s memory of the camping trip was filled with sunshine, laughter, and a thrilling bear sighting. He remembered a downpour, soggy tents, and a frantic search for lost supplies. He squinted at the cover of the book.

He pushed back from the table, the chair scraping against the floor. How could such a fundamental detail be so misremembered? It was almost as if they had experienced different events.

He rubbed his temples, a dull ache spreading through his head. The discrepancy was unsettling, leaving him feeling adrift in a sea of uncertainty. His chest felt heavy, almost suffocating.

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