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 afternoon sun cast long shadows across the dusty floor of the pawn shop. Elias ambled in, the heavy door closing behind him with a gentle thud. He'd spent the morning reading on his porch, a book spread open on his lap, a mug of coffee beside him. The day had a certain languor to it.

His gaze drifted over the shelves, taking in the jumble of objects. A chipped tea set, a worn leather-bound book, a tarnished locket. Then, his attention was caught by a ring. It was a cluster of rubies, set in a gold band. He recognised it instantly. His grandmother's ring.

"How much?" he asked the proprietor, a woman with a kind face. The price was surprisingly low. He paid, feeling a strange mixture of joy and disbelief. As he slid the ring on his finger, he felt an immediate comfort.

Emotion: relaxed

Cluster: Calm / Serenity
PC1 (Valence): 2.55 Positive
PC2 (Disposition): -2.48

Role in Research

This story is one of 1,000 stories generated for the emotion relaxed. 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 relaxed stories, after denoising with neutral dialogue baselines, produces the relaxed emotion vector -- a direction in the model's 5,376-dimensional representation space.

Logit Lens (Layer 40)

Tokens promoted/suppressed when the relaxed vector is projected through the unembedding matrix.

Promoted:
la0.638
disfrut0.418
прият0.386
楽し0.368
affectionately0.368
Suppressed:
L-0.598
S-0.558
C-0.435
because-0.428
ness-0.379