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 air crackled with anticipation. Michael could practically feel the nervous energy radiating from his sister, Sarah, who was sitting next to him. He, on the other hand, felt a strange kind of calm. He took a sip of his sparkling water, the bubbles tickling his nose. The room, filled with family and friends, seemed to vibrate with a soft, celebratory light.

He was on next. He stood up, the chair scraping against the floor. He didn't feel nervous; he felt… ready. He strode to the microphone, feeling a giddy lightness in his chest. “Mom and Dad,” he began, his voice surprisingly steady, “you’ve always taught us to embrace life, to live fully. I think you've led by example here.”

He paused, then added with a playful wink, "So, what's next? World domination? Climbing Mount Everest? I can't wait to see!" The crowd chuckled. Michael found himself grinning, a warmth spreading through him as he knew this was just the beginning.

Emotion: vibrant

Cluster: Positive / Joy
PC1 (Valence): 2.83 Positive
PC2 (Disposition): -1.36

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
la0.518
!0.459
!".0.436
de0.434
!).0.409
Suppressed:
S-0.890
L-0.478
P-0.407
H-0.365
😞-0.306