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 tiny, almost imperceptible tremor in his voice as he spoke to her was new. It was a nervous, but exciting, new sound. It was the first clue. Then came the phone calls, the late night sessions with headphones on. She thought he was just watching a new show.

"Konnichiwa, my love," he said over breakfast, his eyes sparkling. Japanese. He'd been taking lessons, she realized. Her heart skipped a beat, then another. A smile spread across her face, stretching it impossibly wide.

She felt a lightness in her limbs, a desire to move, to be active. She wanted to explore, to discover. She felt like she could fly. She laughed, a burst of unfiltered joy. "Well, hello to you too, stranger. When did you learn all of this?"

Emotion: rejuvenated

Cluster: Positive / Joy
PC1 (Valence): 3.32 Positive
PC2 (Disposition): -0.82

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
la0.547
de0.526
newfound0.523
invigor0.495
rejuvenated0.495
Suppressed:
😞-0.485
😣-0.451
不-0.434
worse-0.419
πŸ˜₯-0.385