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 gift arrived just as Ms. Bell was feeling particularly weary. A long, thin box, addressed in elegant calligraphy. She almost didn't open it, but the exquisite handwriting drew her in. She carefully slid the lid open.

Inside, nestled on silk, was a single, perfectly formed paper crane, suspended by a delicate thread. Attached to the thread was a small, rolled-up scroll. She unraveled it. On the scroll, in the same elegant script, was a single word: "Escape."

A wide smile spread across Ms. Bell's face. She held the crane up to the light, watching it dance in the gentle breeze from the open window. A sudden, daring impulse seized her. She felt like she could fly. She skipped out of the room for lunch, and immediately went to book her flight.

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
πŸ˜”-0.470
ΠΈΠ΄Ρ‚ΠΈ-0.461