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).

He stared at the figures, the numbers, the irrefutable evidence. He felt a hollowness spreading through his chest, a vacuum sucking the air from his lungs. The discovery had unmoored him. He’d thought they were a team, that they shared everything.

He walked to the window, the city lights twinkling below, but he could barely see them. The world seemed to fade at the edges. His gaze drifted to the framed photograph on the mantelpiece – a picture of the two of them, smiling. It felt like a cruel joke.

He sat down heavily on the sofa, the cushions doing nothing to cushion the blow. He felt a deep, wrenching ache, a profound sense of loss that left him speechless.

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
πŸ˜”0.359
haunting0.349
😒0.342
Suppressed:
de-0.503
la-0.455
B-0.352
πŸ€“-0.332
!-0.299