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 empty swing set creaked in the gentle breeze. Across the park, Leo, his friend, sat on a bench, sketching in a notepad. The job application had felt like the only path forward. Now, the absence of certainty felt like a physical blow. He kicked a stray stone, watching it arc across the pavement.

He felt a deep yearning to return to his old life. To the comfort of simplicity, the ease of familiarity. But this feeling of loss was a sign of what he was potentially giving up.

He closed his eyes, the image of the interview playing on a constant loop in his mind, and opened them again. He didn't want to show his weakness, so he just walked away.

Emotion: vulnerable

Cluster: Fear / Anxiety
PC1 (Valence): -2.07 Negative
PC2 (Disposition): -0.69

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.327
ness0.319
😞0.299
😔0.277
😣0.258
Suppressed:
de-0.567
la-0.488
l-0.271
(!)-0.267
(!)-0.257