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 low hum of the television filled the apartment as Michael stared at the ceiling. He felt as if he was waiting for the other shoe to drop, the crushing moment when everything would unravel. He’d heard the hushed whispers, the averted eyes, the way Jessica had started to flinch when he touched her.

The door clicked open and Jessica walked in, her face flushed and her eyes sparkling. She kicked off her shoes and dropped her purse on the floor.

"Guess what?" she said, her voice filled with an excitement that made his heart skip.

He looked at her, then back at the ceiling. He had been bracing for a huge explosion, but what came next was like the gentle whisper of a summer breeze.

“I got that promotion!” she chirped, beaming at him. “And it’s a BIG one.” He rose from the couch and wrapped his arms around her, his arms feeling loose and easy.

Emotion: relieved

Cluster: Positive / Joy
PC1 (Valence): 2.22 Positive
PC2 (Disposition): -0.97

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
la0.638
finally0.606
finalmente0.597
relieved0.561
enfin0.546
Suppressed:
l-0.443
:(-0.439
😣-0.429
S-0.388
😖-0.385