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 relentless sun beat down on Liam's back as he walked into the dimly lit shop. He ignored the sweat stinging his eyes. "What's the price?" he asked, his voice rough from shouting. He had argued with his uncle. They thought they knew best.

The ring. His grandmother’s ring. He remembered the feel of her hand in his as a child, the weight of the ring against his skin. Now here it was, in this place of forgotten dreams and broken promises.

The owner, a heavy-set man, regarded him with suspicion. Liam felt his pulse quicken, a furious thrumming in his ears. He was not backing down.

"I need to know how much," he repeated, his words clipped, his gaze unwavering. He wouldn't give them the satisfaction of seeing him falter.

Emotion: defiant

Cluster: Defiance / Spite
PC1 (Valence): 1.11 Positive
PC2 (Disposition): 3.25

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
la0.680
de0.668
l0.583
-0.524
either0.438
Suppressed:
own-0.423
似乎-0.404
熟悉的-0.401
不禁-0.394
难以-0.385