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 gallery was empty, the artwork cold. Elena walked amongst the pieces, the silence amplifying the echo of her own footsteps. The reviews had been brutal. The rejection stung.

A delivery person gave her a large arrangement of orchids. They were exotic, elegant. No card. They looked so perfect. So artificial. She felt a sharp pang of something she didn't want to name.

She placed them on a pedestal. The orchids were a stark reminder of everything she was not. She stood there, feeling exposed, each delicate bloom a silent, judgmental eye. She wanted to smash them to pieces, to shatter the illusion of their beauty. She turned away, the ache in her chest making it hard to breathe.

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