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 light stung her eyes, and the world seemed to warp around the edges. She stumbled out of bed, a vague unease prickling her skin. That feeling intensified as she walked to the window and saw it: the vacant driveway.

A strange buzzing started behind her ears. She felt a need to sit down, to brace herself. Her pulse quickened. Had she even driven anywhere recently? The question felt heavy in her throat.

She took a shaky breath, trying to collect herself. She gripped the windowsill, the cool metal a small comfort. Her mind raced, a frantic search for the last time she had seen her vehicle. The memory eluded her, a lost whisper in the wind.

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