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).

"Ugh, are you kidding me?" Leo heard his coworker, Clara, say. He couldn't help but suppress a giggle. He straightened up and stretched. His shoulders felt less tight. He had been feeling cramped and stressed all morning. The forced halt was a welcome interruption, as far as he was concerned.

He considered the situation. He had a stack of paperwork to deal with, but he could put that aside. He could just… be. He closed his eyes for a moment, letting the darkness wash over him.

He cleared his throat. "Well, this is going to be a unique experience," he said to Clara, a slight tremor of amusement in his voice. He sat down and closed his eyes.

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