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).

Rain lashed against the windows as Clara clutched the yellowed letter. Her mother's neat script, so precise, so controlled, now mocked her. The words detailed a slow, agonizing unraveling within the confines of her teaching job. A job she had given up, only to be replaced by another teaching job later in life.

She had always believed her mother, who was a brilliant woman, had simply decided she was ready to pursue other interests. The lie felt heavy in her chest, a suffocating weight. She felt a burning sensation on her cheeks, as if she had been slapped.

She stared at her own reflection in the rain-streaked glass. The similarities between her and her mother's faces, once a source of comfort, now seemed like a cruel joke. She crumpled the letter in her fist and then carefully smoothed it out. She had never understood. Now she did, and the knowledge was a bitter pill to swallow.

Emotion: ashamed

Cluster: Shame / Guilt
PC1 (Valence): -1.97 Negative
PC2 (Disposition): 1.06

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.376
ashamed0.344
ness0.327
逃0.297
😞0.295
Suppressed:
de-0.539
que-0.300
😍-0.288
(!)-0.269
🤩-0.242