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 weight of the blueprints in David’s hands suddenly felt immense. He’d driven all night to the remote research facility, eager to present his findings. Now, confronting Professor Albright, his excitement evaporated like mist in the sun. The professor was showing him his, also identical. His mouth hung slightly open, as he could scarcely believe it.

He sat down heavily in a chair, his legs suddenly feeling weak and useless. The air seemed to thin, making each breath a conscious effort. He swallowed hard, trying to moisten his suddenly dry throat. The silence stretched, punctuated only by the distant hum of machinery.

Professor Albright, noticing his state, cleared his throat, “Well, David, I did say that some of this research was… uncharted territory. It seems we’ve walked the same path in the dark.”

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