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 air in Amelia’s dorm room felt thick, stale. She hadn't opened the window in days. The stack of textbooks on her desk seemed to lean in, judging her. The accusation hung over her like a suffocating shroud. She re-read the professor’s email again, each word a cold, hard bullet.

Her hands wouldn't stop shaking as she tried to focus on her laptop screen. The words swam before her eyes. She wanted to scream, to shatter something. Instead, she just sat there, frozen, the cursor blinking accusingly.

She hadn't slept properly in days. The shadows under her eyes were as deep as canyons. She felt a profound emptiness, like someone had stolen a part of her.

Emotion: miserable

Cluster: Sadness / Despair
PC1 (Valence): -2.19 Negative
PC2 (Disposition): -0.87

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😞0.369
无法0.321
無法0.270
💔0.265
😔0.262
Suppressed:
de-0.490
la-0.329
B-0.300
l-0.292
/-0.224