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).

John stared, immobilized, at the faded blue stationery. His reflection, gaunt and pale, stared back from the dusty glass of the old writing desk. He felt a peculiar lightness, a detachment. It was like looking at a stranger, yet it was him. *To My Dearest Emily,* the opening line screamed.

He cautiously unfolded the letter. A tremor ran through him. He found himself mesmerized by the flourish of his younger self’s penmanship. The words were laced with a raw vulnerability he thought he had buried decades ago.

He read of tentative love, of dreams shared and unspoken fears. He laughed a short, disbelieving bark and then was silent. The room seemed to expand, the walls closing in. The weight of time, of missed opportunities, pressed down on him.

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