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 air in the storage unit was stale, the smell of forgotten things clinging to everything. Emily, surrounded by boxes of her mother's belongings, felt a jolt of electricity run through her. In a cracked, leather-bound box, a neatly folded letter. Her mother’s resignation. Years of her mother’s stories of unfair treatment, finally confirmed.

She read it again, slowly this time, savoring each word. Her mother had fought, argued, and eventually walked away, head held high. A giggle bubbled up, followed by a surge of energy. This wasn't a sad discovery; it was a victory. She felt a lightness, an almost unbearable feeling of happiness. She felt… empowered. She held the letter close, a renewed spring in her step.

Emotion: triumphant

Cluster: Positive / Joy
PC1 (Valence): 3.54 Positive
PC2 (Disposition): 0.19

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
de0.734
la0.571
triumphant0.562
🎉0.531
ecstatic0.530
Suppressed:
😞-0.503
S-0.483
😣-0.482
😰-0.457
worse-0.444