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).

He slammed the phone down, the plastic receiver bouncing on the wooden desk. "Lies," he muttered, pacing his cramped office. The article, splashed across every online news outlet, portrayed him as a saint. The man hadn't known the half of it. The truth gnawed at his insides, a relentless and unpleasant guest.

His jaw clenched. He reached for a bottle of scotch, the amber liquid glinting menacingly in the dim light. The world, it seemed, was determined to misunderstand. He'd poured his heart into his work, his passions – and they'd twisted it, reduced him to some bland, heroic figure. He needed to escape this, to find some shred of sanity.

Emotion: bitter

Cluster: Shame / Guilt
PC1 (Valence): -1.78 Negative
PC2 (Disposition): 0.84

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.447
striving0.238
😞0.236
aggravated0.225
soport0.223
Suppressed:
own-0.316
de-0.291
la-0.231
🤩-0.227
是非常-0.225