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 cafeteria buzzed around Chloe, but it might as well have been silent. The accusation from the teacher was a deafening sound. She poked at the food on her tray, suddenly finding the thought of eating repulsive. A knot had tightened in her chest.

She had spent weeks meticulously researching, taking notes, crafting her arguments. The thought that her work was somehow considered fraudulent felt like a betrayal of her own efforts. The familiar setting, normally a place of social comfort, seemed wrong, the chatter of others a distant drone.

She grabbed her things and left, wanting to find a quiet place to think. The air felt heavy, pressing down on her.

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