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).

"I'm becoming a Jew." The words hung in the air, echoing in the sterile silence of the hospital room. Michael couldn't quite meet his son, Daniel's, gaze. His heart felt like a clenched fist.

He found himself running his hand through his thinning hair, a gesture he often used when feeling lost. He felt as though the thread of their relationship had been severed and was now dangling, precariously close to the ground.

The beige walls of the room seemed to close in on him, the rhythmic beeping of the machines a constant reminder of the life he could not fully navigate. His usual jovial spirit was gone, replaced by a dull ache he couldn't name.

Emotion: stuck

Cluster: Sadness / Despair
PC1 (Valence): -2.17 Negative
PC2 (Disposition): -0.72

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.487
缺乏0.401
L0.361
每天0.348
stagn0.348
Suppressed:
l-0.472
la-0.346
随后-0.322
undamaged-0.268
C-0.263