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 car ride was filled with sunshine and laughter. Sarah couldn't stop herself from tapping her fingers on the steering wheel, a nervous energy propelling her forward. Her daughter, Chloe, was about to begin kindergarten, and Sarah had been a bundle of nerves until the teacher assignment had come out.

Mr. Davis. Her heart fluttered in her chest. Mr. Davis. His classroom was a haven, a place of stories and encouragement. He'd helped her find her voice when she was a shy child.

Pulling up to the school, Sarah watched Chloe skip out of the car. She felt a lightness within her, a weight lifted. This year held a promise of magic.

Emotion: optimistic

Cluster: Positive / Joy
PC1 (Valence): 4.44 Positive
PC2 (Disposition): -0.49

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
de0.856
la0.688
!0.601
enthusi0.530
๐Ÿคฉ0.515
Suppressed:
S-0.891
๐Ÿ˜ž-0.504
L-0.486
ness-0.478
ใใ‚Œใงใ‚‚-0.432