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).

"Bored, bored, bored!" Leo muttered, flinging himself dramatically onto the Persian rug in the living room. The sheer newness of the house, the sterile order of it all, was beginning to grate. A yawn stretched across his face, then morphed into a wide, mischievous grin.

He began by poking every loose brick in the fireplace. Finding nothing, he moved onto the wall behind the coat rack, giving it a hearty shove. It didn't budge. He giggled, a tiny sound of pure glee.

Suddenly, a loud clang reverberated from behind a dusty old tapestry in the dining room. He tore it down, revealing a small, iron door. A whoop escaped his lips as he swung the door open, ready for whatever lay within.

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