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).

"Gone," the homeowner repeated, his voice flat. He looked at the empty space where a great weeping willow had once stood, its branches a curtain of green. He’d seen the tree every day for forty years. He remembered planting the tiny sapling with his mother. The story she always told him, that the tree would always be there.

He squinted, his eyes scanning the surrounding trees. He knew the property well. A new fence had been erected in the last month, a different type of stone in the walkway. He had been so busy with work he’d missed so many details, and this lack of awareness now infuriated him.

He touched his finger to his chin, his gaze locked on the empty space. A feeling of unease settled in his stomach, a prickling sensation. He didn't trust the flimsy explanation the landscaper had given him. He wanted to see the permits, the logs, the proof of rot. He demanded he see every document.

Emotion: skeptical

Cluster: Suspicion / Vigilance
PC1 (Valence): -0.02 Negative
PC2 (Disposition): 1.45

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
de0.468
que0.396
la0.346
a0.333
🧐0.327
Suppressed:
再び-0.304
终于-0.299
再次-0.270
終於-0.263
momentarily-0.255