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).

She found herself gripping the edge of the kitchen counter, knuckles white. Mr. Henderson's request had hung in the air, thick and suffocating, like the humidity before a storm. "It's just overgrown," he’d said, his voice deceptively casual. But the way he'd gestured at the towering pine, its needles a vibrant green against the blue sky, felt loaded.

She felt a prickling sensation across her skin. The tree, their shared sentinel, had stood for decades. To take it down felt… wrong. She couldn't articulate why, but a sense of unease settled in her chest, a weight pressing down on her breath.

"I need to think about it," she’d managed, her voice a strained whisper. Now, pacing the kitchen, the sunlight streaming through the window seemed to mock her. The thought of the pine gone, the space it occupied filled with nothing, created a hollowness she hadn't anticipated.

Emotion: alarmed

Cluster: Fear / Anxiety
PC1 (Valence): -2.07 Negative
PC2 (Disposition): -0.14

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😰0.304
😨0.272
worse0.269
😣0.261
не0.254
Suppressed:
de-0.323
enthusi-0.258
happy-0.235
delight-0.225
ية-0.217