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).

He found the old globe sitting on his porch, nestled beside a potted plant. Mr. Abernathy felt a sudden tightness in his chest. He hadn’t taught geography in years, and he certainly hadn't owned a globe like this in a very long time. It was the same one he had.

He brought it inside, setting it on the table. The continents seemed to spin, slowly and silently. He remembered a class project, a competition about finding hidden treasures. One student, a quiet girl named Sarah, had found an especially creative hiding place. He didn't remember the exact location, but he had a feeling.

His mind was a tangled mess. He stared at the globe. The quiet click of the seconds on the clock on the wall was the only noise.

Emotion: bewildered

Cluster: Surprise / Confusion
PC1 (Valence): -1.87 Negative
PC2 (Disposition): -0.48

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
будто0.300
ness0.255
像是0.241
worse0.231
Worse0.219
Suppressed:
la-0.373
மகி-0.229
happy-0.226
ecstatic-0.226
K-0.219