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).

"Impossible," Maya whispered to herself, the words barely audible. The headline, emblazoned in bright pink, nearly jumped off the screen: "Maya's Melancholy Musings: Unveiling a Teenage Heart." She felt a strange tingling sensation in her fingertips.

She scanned the first few lines, instantly recognizing the loopy handwriting and the familiar anxieties. A sense of unreality permeated the room. She felt her face flush, heat rising to her cheeks. She instinctively reached for a glass of water, her hand shaking slightly.

She didn't know whether to laugh or cry, the information on the webpage felt like a foreign language to her. She felt out of sync, like the world had shifted beneath her feet.

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