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 quarterback felt the familiar weight of the ball in his hands. Every play was ingrained in him, a pattern of steps and throws etched into his memory. The stadium's electricity, the thrum of the crowd, the sharp scent of the grass - all were extensions of him. After a stellar season, he was sure he would play the same role.

The coaches called him over to the sidelines during practice. He listened intently, nodding, as the coaches discussed a new strategy, a change in play calling that would require him to play a wide receiver. The words "we're thinking…" floated in the air.

He felt the smile on his face falter, the comfortable air around him starting to freeze. The suggestion felt like a betrayal of all the training he did. He felt a surge of… something. He stood as the new plays were drawn out. The ball felt wrong in his hands now.

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