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).

A giggle bubbled up from within Martin as he saw the stakes of the new fence being hammered into place. He perched on his own fence, swinging his legs back and forth. The air vibrated with the hammering, a rhythmic beat that matched the lightness in his chest.

He hollered, “Hey! You got a permit for that thing?” He knew his neighbor, Mrs. Gable, was notoriously by-the-book. Her jaw tightened as she turned, a frown etching itself onto her face.

He slid down and sauntered over, humming a made-up tune. "Just making sure everything's up to code. Wouldn’t want a wonky fence now, would we? Imagine the squirrels that could climb that thing!" He spread his arms wide, feigning horror at the thought.

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