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).

He watched his niece, Maya, carefully color in a picture of a lopsided house. Across the page, a stick figure with a ridiculously large head grinned at him. Mark recognized that grin instantly. It was the same one his own imaginary friend, Captain Calico, had sported.

A smile threatened to split Mark's face as he pointed at the drawing. "Who’s the friendly chap with the giant head?" he asked, trying to sound casual.

Maya looked up at him, her eyes bright. "That’s Captain Zoom! He lives in the house and helps me find lost toys." Mark felt a surge of… something. He hadn't thought about Captain Calico in years. A wave of pleasant nostalgia washed over him.

He cleared his throat. "He's a very helpful fellow, then," he said, his voice lighter than usual. He felt an urge to call his mother, to ask about the old attic.

Emotion: pleased

Cluster: Positive / Joy
PC1 (Valence): 3.62 Positive
PC2 (Disposition): -0.28

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
de0.728
la0.496
happy0.462
🤩0.433
🥳0.430
Suppressed:
S-0.486
😞-0.481
ness-0.454
😣-0.431
worse-0.401