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).

The aroma of freshly baked bread and strong coffee filled the cozy kitchen. Leo was bustling about, adjusting the tiny heart-shaped arrangement of strawberries on the plate. He was completely at ease. He wanted her to feel happy. He was already fantasizing about her reaction.

A knock on the door, and the door opened. She walked in. Her eyes were bright. Her hands were cupped behind her back. โ€œGuess what, handsome?โ€ she cooed.

Leo, filled with glee, clapped his hands together. โ€œI want to hear everything!โ€

He started. The words felt familiar. He stopped mid-sentence. From behind the door, a second set of footsteps. Another woman entered the room, her expression identical. He wanted to laugh. He started again.

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
๐Ÿ˜œ0.553
Suppressed:
ness-0.685
๐Ÿ˜ž-0.590
S-0.562
๐Ÿ˜”-0.470
ะธะดั‚ะธ-0.461