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 phone call had come late. Liam had been half-asleep. Now, he sat at the kitchen table, the receiver clutched tightly in his hand, listening to his brother, Kevin, on the other end. Kevin’s voice was filled with a joy Liam didn’t recognize. He was telling Liam he was joining a new religious order.

Liam lowered the phone slowly, letting it hit the table. The words echoed in his head, a discordant symphony. His head throbbed. He felt a strange emptiness inside his chest. He looked at the half-eaten sandwich on the counter, the uneaten fruit in the bowl. He couldn't muster the strength to pick up either.

Emotion: tired

Cluster: Fatigue / Lethargy
PC1 (Valence): -0.92 Negative
PC2 (Disposition): -1.00

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
weary0.254
fatigue0.249
😴0.238
sighed0.237
barely0.230
Suppressed:
C-0.268
because-0.225
B-0.219
🤟-0.217
భి-0.214