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 scent of chlorine and stale air clung to the gym. She was a runner, a competitor, and she thought she had trained hard. She had practiced, she had pushed herself. The coach called her over. She saw another girl in the gym, a girl she had never met.

Two identical training plans. The same exercises, the same schedule, the same goals. A feeling of dread washed over her, the weight of the moment pressing down. She felt as though her insides had turned cold.

The other girl met her gaze, a mirroring of her own bewilderment. The air crackled with a silent understanding. Everything she thought she knew, everything she believed, was wrong. Her entire world was collapsing.

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