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 attic air hung thick with dust motes dancing in the lone sunbeam slicing through a crack in the boarded-up window. Sarah hummed a jaunty tune, her fingers already itching to rummage. This was a treasure hunt of sorts, a trip down memory lane, and she was already imagining all the awkward, embarrassing gems she’d unearth. Her mother’s meticulous organization, usually a source of mild frustration, was today a comedic goldmine.

She unearthed the first box. Inside, neatly stacked, were primary school drawings – stick figures with impossibly long limbs and suns with more rays than fingers on her own hand. A giggle escaped her lips. This was going to be fun.

She pulled out a crayon-scrawled poem about a fluffy white bunny, its rhyming scheme… questionable. Sarah’s face broke into a grin; she could practically hear her own childish voice, full of unearned confidence.

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