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 school's annual bake sale was in full swing, and Michael felt a knot form in his stomach. Mrs. Peterson, his former Home Economics teacher, was overseeing the event. The memory of the inedible cookies he'd made in her class, and the subsequent lecture on following instructions, came flooding back.

He made a beeline for the coffee stand, hoping to hide. He watched Mrs. Peterson, her gray hair pulled back in a severe bun. He was embarrassed, recalling the burnt mess he made.

He felt a tap on his shoulder. "Michael, isn't it?" Mrs. Peterson inquired, her eyes twinkling.

He could only offer a weak smile.

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