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 gallery owner beamed, gesturing expansively at the new exhibition. Elias followed her gaze, his brow furrowing as he took in the crowd milling around his pieces. He felt a disquiet he couldn't quite name.

A woman approached, a stranger with a kind face. "Your work is so… moving," she said, her voice soft. "It really stayed with me." Elias offered a polite smile, but his attention was caught.

She rolled up her sleeve, revealing his signature swirl of ink, a piece he'd painstakingly rendered in oils just months ago. Its vivid colours were replicated, perfectly, on her skin. He felt a tremor run through him, a strange coldness in his stomach. How?

Emotion: puzzled

Cluster: Surprise / Confusion
PC1 (Valence): -1.17 Negative
PC2 (Disposition): -0.25

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
que0.453
🤔0.421
🤔0.402
🧐0.328
a0.294
Suppressed:
ecstatic-0.265
gladly-0.238
swiftly-0.237
alleviate-0.234
wholeheartedly-0.232