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 bland artwork on the wall did nothing to alleviate the growing feeling of despair. Ethan found himself tracing the outline of a tear in the armchair’s fabric. He felt a constant pressure, a burden of expectations he could never fulfill.

He was here because he had to be. He’d spiraled downward, everything crumbling around him. And now, fate – or perhaps a cruel sense of humor – had delivered him to the office of his childhood tormentor, Dr. Marcus Bell.

When the door opened, and Marcus Bell was standing there, Ethan couldn’t move. He felt like a small child again, shrinking under the harsh glare of someone whose opinion mattered more than his own. He hated himself.

Emotion: self-critical

Cluster: Shame / Guilt
PC1 (Valence): -1.66 Negative
PC2 (Disposition): 1.28

Role in Research

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

Logit Lens (Layer 40)

Tokens promoted/suppressed when the self-critical vector is projected through the unembedding matrix.

Promoted:
L0.748
S0.662
ness0.425
ly0.374
탓0.351
Suppressed:
la-0.492
soon-0.304
(!)-0.301
secured-0.284
cautiously-0.280