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).

Sarah couldn't stop tapping her foot. The news had been delivered by a frantic intern, clutching a note that read: "Off the grid. Have fun, kiddo!" Mrs. Gable, the editor, had vanished! Sarah, who had long suffered her biting critiques and demanding deadlines, felt a lightness she hadn't known existed.

Her fingers flew across the keyboard, words pouring out effortlessly. She rewrote the entire front-page article in under an hour, adding witty observations and daring pronouncements. The stale air of the office suddenly felt invigorating, the clicking of keys a symphony.

She leaned back in her chair, stretching her arms above her head, the image of Mrs. Gable's stern face dissolving in her mind. A giggle escaped her. This was going to be the best issue yet.

Emotion: ecstatic

Cluster: Positive / Joy
PC1 (Valence): 3.16 Positive
PC2 (Disposition): -0.15

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
!0.588
ecstatic0.560
de0.559
🤩0.548
overjoyed0.547
Suppressed:
S-0.678
😞-0.500
ness-0.405
😔-0.368
😥-0.359