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).

"Coffee?" Sarah asked, her voice brighter than usual, as she entered the breakroom. She had barely slept last night, replaying the interview in her head a dozen times. Each time, she found herself more and more satisfied with her answers.

The aroma of freshly brewed coffee, usually a comfort, seemed almost overwhelming today. She took a large gulp, the caffeine jolting through her veins, fueling the excitement. She had imagined all the ways she would restructure her team, make improvements that had been on the back-burner.

She caught her reflection in the microwave door and smoothed down her already perfect suit. She couldn't stop smiling. She had a good feeling about this.

Emotion: hopeful

Cluster: Positive / Joy
PC1 (Valence): 3.68 Positive
PC2 (Disposition): -0.58

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
de0.715
la0.549
B0.468
🀩0.452
enthusi0.433
Suppressed:
S-0.603
不-0.401
😞-0.399
worse-0.394
C-0.386