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).

Aunt Clara hummed, meticulously arranging the finger paints on the tiny table. Her niece, Lily, giggled, snatching a bright blue pot. Clara watched, a soft smile playing on her lips. Lily had been struggling with nightmares, and Clara had offered to come over and help with a distraction.

That's when she saw it. In the corner of Lily's drawing, a familiar figure. A small, winged creature, with mismatched eyes and a crooked smile. It was Pip, her childhood companion.

Clara felt a warmth bloom in her chest, a lightness she hadn't known she missed. She helped Lily finish the picture, a strange sense of comfort settling over her as she looked at Pip's familiar features. She would do anything for Lily, anything to help her.

Emotion: grateful

Cluster: Positive / Joy
PC1 (Valence): 3.77 Positive
PC2 (Disposition): -0.43

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
la0.571
de0.562
B0.546
மகி0.517
happy0.493
Suppressed:
[]-0.582
😖-0.519
worse-0.509
😫-0.498
😣-0.494