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 doorbell chimed, followed by the metallic clang of the letterbox. Liam had been trying to concentrate on his novel, the words proving stubbornly elusive. He rose, a heavy sigh escaping his lips. More flowers. Again. A flamboyant explosion of sunflowers this time, practically screaming for attention. No card.

He poked the arrangement with a hesitant finger. Sunflowers. He hated sunflowers. He hated the presumption of cheerfulness they exuded. He felt a slow burn of frustration blossoming in his chest, a heat that prickled his skin. He shoved the vase onto the table, the bright yellow a jarring contrast to the muted tones of his living room.

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