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 attic stairs creaked under Liam's weight as he hauled up the boxes. He had been looking for a specific Christmas ornament. Instead, he found a trove of old pictures. He'd always known his father had been a bit of a rake in his youth. The stories were vague, hinting at a wild side. But the photographs were another thing altogether.

One photo depicted his father riding a Harley, a girl perched on the back. Another photo showed his father at a casino, with a winning hand, a look of triumph on his face. This wasn’t the quiet, thoughtful man he knew. The shock was a physical blow. He felt lightheaded.

Liam backed away, the box forgotten. He needed air. He felt a profound sense of disorientation, a sense that he was looking at a person he didn't recognise. He needed time to process the conflicting images. He didn’t know where to start.

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