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 endless cycle of dishes, laundry, and bills had been wearing down Daniel. He'd been looking forward to nothing, and he felt a bone-deep ache that even a hot shower couldn't wash away. The ping of a new message from "PixelNinja" on his computer screen felt like a chore to answer.

He typed a curt reply. "What's up?"

PixelNinja, in turn, wrote, "I'm going to the arcade tonight. Wanna come?"

Daniel had been to the arcade a decade ago. It had seemed like a lifetime.

He closed his eyes for a moment. He hadn't been out of the house in days. The thought of getting dressed and going out seemed like a ridiculous undertaking. But then a small, unexpected curiosity flickered to life within him. He felt a faint surge of something approaching... excitement? He typed back, "Maybe."

Emotion: weary

Cluster: Fatigue / Lethargy
PC1 (Valence): -1.52 Negative
PC2 (Disposition): -1.02

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.384
😞0.288
sighed0.230
ness0.210
😪0.203
Suppressed:
de-0.339
l-0.300
!-0.222
la-0.221
🤟-0.198