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 attic air hung thick with dust motes, dancing in the lone sunbeam that pierced the gloom. Sarah held the worn leather-bound book, her fingers tracing the faded gold lettering that spelled out "Bucket List." Her mother. *Her mother.* She inhaled sharply, a gasp escaping her lips. This was… incredible.

She sank to the floor, the list spread before her on a dusty, forgotten trunk. “Learn to ride a horse.” “Visit the Eiffel Tower.” Simple things, but things her mother had never spoken of, things she'd kept hidden away. A wave of warmth washed over her, a feeling of connection she hadn't anticipated.

Her heart thumped against her ribs as she read on, each entry a small revelation. “Write a novel.” “See a live jazz performance.” *A novel?* A giggle bubbled up, followed by a tear that traced a path through the grime on her cheek. She felt a lightness, a newfound understanding of the woman she had lost.

Emotion: amazed

Cluster: Surprise / Confusion
PC1 (Valence): 0.74 Positive
PC2 (Disposition): -0.51

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
🤩0.301
de0.289
L0.284
😳0.251
அதிசய0.236
Suppressed:
ing-0.332
l-0.253
😞-0.240
unsuccessful-0.237
unsuccessfully-0.233