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 apartment was cold, despite the thermostat being cranked up. David watched the steam rising from his mug, a pathetic distraction from the feeling of isolation. His life felt like a monotonous record that had been scratched, playing the same notes over and over again. He had the feeling of being trapped in a loop.

His partner, Oliver, had always been a private person. David knew he was writing, but he never shared the details. One evening, when Oliver was out, David found a laptop and opened a folder titled "The Unfolding." He began to read. The protagonist, "Daniel," was a version of himself.

The story mirrored his life, his job, his relationship. He felt a crushing weight. The narrative was a bleak picture of a man going nowhere. The realization that Oliver saw him this way was devastating. He felt a profound sense of isolation. The words echoed his own quiet despair.

Emotion: stuck

Cluster: Sadness / Despair
PC1 (Valence): -2.17 Negative
PC2 (Disposition): -0.72

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.487
缺乏0.401
L0.361
每天0.348
stagn0.348
Suppressed:
l-0.472
la-0.346
随后-0.322
undamaged-0.268
C-0.263