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 air inside the house felt thick, stale, stagnant. Ethan felt it cling to him, the same way his grandmother's unspoken disappointment always had. He’d never been a good son, not a good grandson. Now, he felt the heavy presence of her absence.

He shifted uncomfortably, avoiding eye contact with his sister, Chloe, as she excitedly made a list of things to sell. He felt the blood rush to his face, a burning sensation he hated. He wanted to leave, to disappear, to erase the past.

He focused on the dust motes dancing in the sunbeams, anything to avoid the truth. He'd let her down. He had utterly failed.

"We can make some serious money," Chloe said, her voice buoyant.

Ethan just nodded, his throat too constricted to speak.

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