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 clicking of keyboards echoed in the cavernous office. It was a sound that had once been familiar, comforting, a symphony of progress. Now, it grated on Elias’s nerves. He looked at the woman sitting beside him, her brow furrowed in concentration. Amelia. His replacement.

He felt a strange, detached curiosity about her. He should be showing her the ropes, explaining the intricacies of their projects. But the ropes felt frayed, the knots unidentifiable. His mind was blank, like a pristine whiteboard.

He gestured vaguely at the computer screen. “Um… so… you see this file?” His voice wavered. His own work felt distant and disconnected, like a faded photograph. He felt a desperate urge to run, to escape the suffocating weight of this new reality.

Emotion: disoriented

Cluster: Surprise / Confusion
PC1 (Valence): -2.65 Negative
PC2 (Disposition): -1.34

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
unfamiliar0.340
ness0.321
失去了0.316
не0.315
будто0.311
Suppressed:
B-0.474
la-0.419
K-0.409
de-0.309
brag-0.302