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).

John closed the heavy oak door of his study, the click echoing in the quiet room. The scent of old books and pipe tobacco filled the air, a familiar and comforting aroma. He settled into his leather armchair, the weight of the book in his hands a grounding presence. He had built a life of knowledge and solitude, a fortress against the chaos of the outside world.

His laptop beeped. A notification. He glanced at it, a flicker of irritation crossing his face. A shared link. He clicked, then stared, his face slowly losing all color. It was his diary, scrawled teenage angst in full view. He felt a profound sense of disruption wash 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