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).

From the moment the doorbell rang, Miles was practically buzzing. He’d just finished a particularly satisfying Sudoku puzzle and was feeling invincible. He skipped to the door.

There stood Sarah, his across-the-street neighbor, looking apologetic. "Miles, do you have a moment?"

"Sarah! Right on time!" He grinned, his teeth flashing. "What's the big idea?"

She hesitated. "I was wondering... could I maybe use some of your yard to grow some herbs? My kitchen feels…empty."

Miles let out a whoop. "Herbs! Fabulous! But only if you promise to let me harvest the rosemary! And the basil. And maybe the thyme. We can have a herb-themed potluck, you know. Imagine!" He clasped his hands together, imagining the possibilities. "We'll need to figure out the sunniest spot, then the fun begins!"

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