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, warm and inviting, streamed through the window, but Miles felt only a profound inner stillness. He had been chasing after his nephew all day, and now he found himself slumped on the sofa. Every muscle felt leaden. His eyes drifted closed for a moment and then snapped open.

He glanced at the drawing his nephew, Finn, had excitedly presented. It showed a figure with iridescent wings and a crown of thorns. "This is Sparky," Finn declared proudly. "He told me all about the clouds."

Miles took the drawing and forced a smile, but the image struck a chord. Sparky. The same absurd name he had given to the being who would keep him company during the long, lonely evenings of his youth. A chill went down his spine. The room felt suddenly too warm.

"Very impressive, Finn," he said, fighting the urge to lean back and allow the world to fade away.

Emotion: sleepy

Cluster: Fatigue / Lethargy
PC1 (Valence): -0.86 Negative
PC2 (Disposition): -1.55

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😴0.318
weary0.310
fatigue0.296
💤0.283
似乎0.280
Suppressed:
B-0.343
C-0.294
because-0.293
à°­à°¿-0.281
เพราะ-0.277