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 weight of the day had finally lifted. Mark sank into the plush couch, the cushions molding perfectly to his frame. The quiet was a relief, a balm to his frazzled nerves. The day had been long, but he’d accomplished everything he had planned. He felt a deep sense of satisfaction.

He reached for a stack of newspapers, intending to catch up on the news, but his eyes were drawn to a university publication. He skimmed the articles, his gaze landing on the featured section: "Outstanding Essays."

He blinked. There, staring back at him, was his essay. He leaned back, his eyes scanning the familiar words. A smile, slow and genuine, spread across his face.

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