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 concert hall buzzed with anticipation. Amelia adjusted the strap of her vintage dress, a slight tremor of excitement running through her. She'd been chatting with Ben online for ages; sharing music, discussing life, building a strong connection. He'd mentioned he was coming to this show, too, a smaller local band she also enjoyed.

She navigated the crowd with a certain purposefulness, her shoulders back, her head held high. When she spotted him, leaning against a wall near the stage, a feeling of anticipation exploded inside. Her heartbeat sped up. He looked up, and their eyes met.

He straightened, a look of surprise and delight spreading across his face. He waved. Amelia strode towards him, a vibrant force radiating from her.

Emotion: self-confident

Cluster: Positive / Joy
PC1 (Valence): 3.45 Positive
PC2 (Disposition): 0.56

Role in Research

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

Logit Lens (Layer 40)

Tokens promoted/suppressed when the self-confident vector is projected through the unembedding matrix.

Promoted:
de0.778
la0.550
l0.467
😎0.458
😉0.411
Suppressed:
S-0.961
L-0.538
😞-0.464
마다-0.459
😣-0.458