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 afternoon sun beat down on the chaotic yard sale. Eleanor almost missed it, the little bear tucked away in the shadows of a rickety table. She recognized the faded gingham bow immediately. It was Pip, her first friend, a constant companion during her solitary childhood. A sharp pang settled in her chest.

She didn’t hesitate. She paid the indifferent teenager manning the table, her hands unsteady. Clutching Pip, she felt a wave of… desolation wash over her. It was too late to go back, to undo the things she’d done.

She traced a finger along Pip's worn snout. The silence of the afternoon was deafening, amplified by the silent guilt. "I'm so sorry, Pip," she whispered, her voice barely audible.

Emotion: remorseful

Cluster: Shame / Guilt
PC1 (Valence): -1.57 Negative
PC2 (Disposition): 1.09

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.656
L0.580
悔0.463
remorse0.416
been0.398
Suppressed:
de-0.444
(!)-0.309
имеется-0.289
oldukça-0.287
😍-0.285