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 workshop was filled with the rhythmic click of knitting needles. Amelia led her weekly knitting circle, her fingers dancing across the yarn. It was a haven, a place of camaraderie and creativity. She had been knitting all her life. This was her passion. She felt at home.

She chatted with the other knitters, sharing stories and laughter. She helped them with their patterns, offering encouragement and guidance. Amelia glowed as she looked at all the colors and shapes. It was her sanctuary.

She read the review. "Poor instructions," it declared. "Amateurish designs." Amelia felt the blood drain from her face. The workshop seemed to dim. The needles fell silent. The sense of belonging was ripped away.

Emotion: happy

Cluster: Positive / Joy
PC1 (Valence): 3.98 Positive
PC2 (Disposition): -1.14

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
la0.676
excitedly0.533
enthusi0.505
de0.498
๐Ÿคฉ0.495
Suppressed:
S-0.896
๐Ÿ˜ž-0.430
ness-0.410
ใใ‚Œใงใ‚‚-0.387
L-0.379