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 crackling fire in the hearth cast dancing shadows on the walls of Ethan's cabin. He sat in his favorite armchair, a well-worn book open in his lap, the scent of pine and burning wood a soothing balm. Outside, the wind howled, but inside, he was enveloped in a cocoon of warmth and quietude. He had chosen this solitude, cultivated this peace.

His phone buzzed. A message from his sister, a flurry of exclamation points and a link. He clicked it reluctantly, his heart beginning to pound in his chest. His teenage diary. The vulnerable words, the clumsy confessions, were there, exposed. The book slipped from his grasp as a wave of nausea washed over him.

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