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 cursor on Amelia's laptop danced erratically as she scrolled through the real estate listing. A familiar, slightly off-kilter facade stared back at her. The listing displayed a picture of her childhood home, bathed in the soft glow of a late afternoon sun. A warm flutter filled her chest, a feeling akin to the gentle rise of a balloon.

She clicked through the photos, each image triggering a cascade of memories. The chipped paint on the porch swing, the stain on the living room carpet, the way the sunlight streamed through the kitchen window in the mornings – they were all captured in perfect detail. Her fingers trembled as she reached for her phone.

"Mom?" she said, her voice barely a whisper. "Did you see... the house?" Her breath hitched. "It's for sale." She imagined the possibilities, a future where she could once again feel the solid weight of the front door beneath her hand.

Emotion: hopeful

Cluster: Positive / Joy
PC1 (Valence): 3.68 Positive
PC2 (Disposition): -0.58

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
de0.715
la0.549
B0.468
🀩0.452
enthusi0.433
Suppressed:
S-0.603
不-0.401
😞-0.399
worse-0.394
C-0.386