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 sterile white walls of the hospital corridor seemed to close in around David. Dr. Ramirez, his mentor, had left him a brief message in his email, saying she was resigning to take care of her family. No goodbye. No personal message. Nothing.

David ran a hand over his tired face. It had all felt so secure, like they were running a well-oiled machine. He felt the sudden thud of his own heart. He was on call tonight, for the first time alone.

He squared his shoulders, a tremor running through them. He knew the procedures, knew the patients, but now he was the only one. He felt exposed, standing in the middle of a warzone. He walked down the hall, each step heavier than the last.

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