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).

He stared at the glossy magazine, the ink shimmering under the overhead light. His hands trembled as he held it, the weight of it suddenly substantial. His article, a deep dive into the philosophy of happiness, was there, but the byline read “Evelyn Reed.” He’d worked for months, late nights fuelled by coffee and a relentless drive.

He sank into his chair, the leather cool against his back. A slow smile spread across his face, his lips parting slightly. He reached out to touch the page, the texture of the paper a grounding force. The words, his words, seemed to leap out, bolder than he remembered.

A warmth flooded through him, a sensation of complete, utter disbelief mixed with… well, he couldn’t name it. A quiet hum started in his chest, a resonant thrum that vibrated through him. He sat, suspended in a state of quiet wonder, the stolen glory a new and strange kind of currency.

Emotion: awestruck

Cluster: Surprise / Confusion
PC1 (Valence): -0.17 Negative
PC2 (Disposition): -1.20

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
L0.271
suddenly0.260
stranger0.252
plötzlich0.246
wonder0.242
Suppressed:
la-0.450
싶-0.350
K-0.319
rest-0.248
싶은-0.238