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 leaned back in his chair, the office lights blurring with a sudden energy. He'd been wanting to prank his boss all day. He had just finished writing a message, a carefully crafted insult, disguised as a compliment. “You’re so good at your job, you could probably teach a class on… whatever it is you do.” He grinned, picturing his boss’s reaction. His heart gave a little skip as he hit the button. Damn. He felt a sudden, sinking feeling.

He glanced up at the ceiling, then quickly back down at the phone. He had to decide what to do. The screen offered a non-descript name, and a blank space to fill in. He could change it. He could ignore it. He was probably going to get in a world of trouble.

Emotion: playful

Cluster: Positive / Joy
PC1 (Valence): 4.42 Positive
PC2 (Disposition): 1.44

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
B0.662
(!)0.566
!0.562
!?0.553
😜0.553
Suppressed:
ness-0.685
😞-0.590
S-0.562
😔-0.470
идти-0.461