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).

"Perfect!" declared Emily, clapping her hands together. The newly renovated backyard, complete with a sparkling pool and a fire pit, felt like her own private paradise. She’d spent weeks designing and overseeing the project, and the final result was a dream. She began to prance around the pool, enjoying the feeling of the sun on her skin, humming a silly tune.

A knock startled her, and she opened the door to find two stern-faced detectives. They explained the history of the house, a grisly story she’d never considered. A murder. She found herself struggling to hold back a giggle as the detective spoke.

"So, you see, the victim was found right here..." the officer explained, pointing to a spot near the pool. "The pool? Excellent staging for a murder scene," she blurted out, before catching herself and covering her mouth. Her grin was nearly uncontainable.

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