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 composer, Ethan, slumped back in his chair, the final notes of his symphony echoing in the silence. Exhaustion mixed with a deep sense of satisfaction. He'd poured his heart into this piece, his magnum opus. The premiere was fast approaching.

His phone buzzed. A message. A fellow musician, Liam, wrote saying he, too, had been writing a symphony, and that he'd be honored to hear Ethan's. They were remarkably similar in form, technique, and scope, the message stated.

A smile spread across Ethan's face. He could feel a joyous energy vibrating through him. The world seemed bright, filled with potential. He was no longer alone in this creative world.

Emotion: hope

Cluster: Positive / Joy
PC1 (Valence): 3.59 Positive
PC2 (Disposition): -0.12

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
de1.065
la0.865
B0.465
🀩0.412
H0.407
Suppressed:
😞-0.459
C-0.451
不-0.449
😣-0.443
S-0.442