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).

From his perch on the balcony, Leo watched the city lights begin to twinkle, a comforting tapestry against the twilight. The scent of jasmine from the climbing vines wrapped around the railing filled his lungs. He felt a deep, abiding contentment. After weeks of relentless work, he'd finally finished the final illustration for his upcoming children's book. He stretched, a satisfying crack echoing in his shoulders.

A notification popped up on his phone. A fan had tagged him in a photo. Curiosity piqued, he clicked. A close-up shot of an arm, inked with one of his whimsical characters, a grinning, fluffy monster. It was… unsettling.

Emotion: safe

Cluster: Calm / Serenity
PC1 (Valence): 1.86 Positive
PC2 (Disposition): -2.55

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
soon0.402
affectionately0.357
disfrut0.329
необходи0.325
🥰0.317
Suppressed:
S-0.474
C-0.416
L-0.408
worse-0.379
even-0.370