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 morning sun, filtered through the kitchen window, warmed Amelia’s face as she flipped a pancake. The radio hummed softly, playing a familiar country tune. She hummed along, a practiced rhythm to her movements. Everything felt right, peaceful. The smell of frying bacon mingled with the maple syrup, a comforting aroma. Her dog, a scruffy terrier mix named Gus, thumped his tail against the floor, patiently waiting for a scrap.

Her gaze drifted towards the house across the street. The one with the meticulously manicured lawn and the always-gleaming windows. Mrs. Gable, a woman of indeterminate age, always seemed to be out tending her roses in a pristine white dress. Amelia’d always assumed she was a retired librarian.

This morning, though, Mrs. Gable was pulling out of her drive... in a vintage Rolls Royce. Amelia’s jaw dropped. The pancake she was about to eat suddenly felt heavy in her hand.

Emotion: content

Cluster: Calm / Serenity
PC1 (Valence): 2.98 Positive
PC2 (Disposition): -2.19

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
la0.446
affectionately0.435
😊0.392
😊0.386
thoughtful0.386
Suppressed:
S-0.682
L-0.537
C-0.485
😣-0.362
est-0.350