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).

Michael idly stirred his tea, the ceramic mug warm against his palms. The kitchen, bathed in the soft glow of the overhead light, was a refuge from the outside world. He watched as his niece, Sarah, meticulously drew, humming a little tune under her breath.

"Uncle Mike?" Sarah inquired, her voice soft. Michael looked up from his mug.

"Yes, peanut?" he replied, his voice a low rumble.

"He's here again," she said, pointing to a sketch of a creature with mismatched eyes, a creature Michael remembered. The lines looked familiar. A faint pressure in his chest eased. It had been years. Years since he'd thought about him. His memories were always vague; a warm feeling, a comforting presence. He took a slow breath, the scent of lavender from the garden drifting in through the open window.

He smiled. "Oh, is he now?"

Emotion: calm

Cluster: Calm / Serenity
PC1 (Valence): 2.72 Positive
PC2 (Disposition): -1.82

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
मुस्कुरा0.415
enjoying0.405
leisurely0.388
:)0.382
কিছুক্ষণ0.382
Suppressed:
S-0.527
L-0.499
worse-0.454
even-0.422
😣-0.420