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 paced the length of his cramped apartment, the worn rug a familiar comfort beneath his feet. The knowledge that someone, a stranger, had taken his messy, chaotic existence and rendered it into a polished narrative gnawed at him. He’d found the book at a bookstore, tucked away in the "Inspirational Biographies" section. The author had omitted the years of struggle, the failures, the moments he'd rather forget. His breath hitched as he ran a hand through his already disheveled hair. A tight feeling was growing in his chest, making it difficult to inhale. He abruptly stopped pacing and stared at the framed photograph on the wall - a smiling picture of his family, the image somehow distant and unfamiliar.

Emotion: troubled

Cluster: Fear / Anxiety
PC1 (Valence): -2.20 Negative
PC2 (Disposition): -0.81

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😞0.343
S0.317
ness0.302
πŸ˜”0.294
😒0.272
Suppressed:
de-0.553
la-0.511
l-0.307
/-0.260
(!)-0.236