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 sunlight sliced through the dusty windows of the spare bedroom, illuminating the truth. Sarah had always been told her grandmother, Evelyn, was a saint, a woman who never raised her voice. These pictures said something different.

One photo showed Evelyn, younger, standing on a beach with a drink in her hand. Her face was flushed, her stance relaxed. Beside her was a man Sarah had never heard of. His arm was casually around her waist.

A wave of heat washed over her. It was a betrayal, she thought. A deliberate withholding of information. She didn't believe the flimsy excuses her mother would offer. She was done accepting the idealized portrait.

Emotion: obstinate

Cluster: Defiance / Spite
PC1 (Valence): 0.25 Positive
PC2 (Disposition): 3.22

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
l0.642
la0.547
de0.536
either0.498
-0.433
Suppressed:
竟然-0.354
终于-0.353
难以-0.353
原来-0.351
previously-0.349