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).

"Could you… turn it down?" she finally asked, her voice trembling as she knocked on the neighbor's door. The music from next door was a constant, throbbing pain, each note a reminder of her loss. She had spent the day in a haze of grief, the sounds amplifying the ache in her chest.

The door opened and a young man, clearly annoyed, peered back at her. Her heart sank, her eyes prickling with tears. She tried to say something, but the words wouldn't come.

She turned and rushed back to her apartment, slamming the door behind her. She leaned against the door for support, letting out a sob. The music continued, a cruel soundtrack to her misery.

Emotion: heartbroken

Cluster: Sadness / Despair
PC1 (Valence): -2.60 Negative
PC2 (Disposition): -1.10

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😞0.433
S0.385
😔0.359
haunting0.349
😢0.342
Suppressed:
de-0.503
la-0.455
B-0.352
🤓-0.332
!-0.299