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 setting sun cast long shadows across the room, illuminating dust motes dancing in the air. Liam sat slumped on his couch, the remnants of a pizza box littering the coffee table. He'd been glued to his computer screen all day, working on a project, and the effort had completely drained him. He felt heavy, as if gravity had increased tenfold.

A notification appeared on his screen. "Congratulations! Your essay has been selected!" He squinted at the words, the letters blurring slightly. He felt disoriented. It took a monumental effort to click the link. The essay, the product of countless hours, stared back at him. He let out a long, drawn-out sigh, his body slowly sinking further into the cushions.

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