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 insistent buzzing of the doorbell finally broke the spell of the morning. Standing on the porch, Mr. Miller, his usually friendly demeanor replaced by a strained smile.

"Morning, John! About the willow..." The willow, a weeping beauty, a haven for the robin.

John felt a familiar pressure building in his temples. He was just getting ready to take a nap.

“It's getting a bit… messy, isn’t it? The roots, the branches…” Mr. Miller’s voice, usually a comforting baritone, sounded grating.

John felt a sudden urge to flee inside and lock the door.

He bit back a retort, forcing himself to smile back. "I'll think about it," he said, hoping the words sounded sincere.

Emotion: annoyed

Cluster: Anger / Hostility
PC1 (Valence): -0.59 Negative
PC2 (Disposition): 1.27

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
l0.376
😠0.251
骂0.249
🤬0.245
fucking0.239
Suppressed:
own-0.352
渗透-0.251
gradually-0.234
unfolding-0.226
超越-0.224