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 just... left?" Leo stammered, staring at the empty desk. Professor Chen's office was stripped bare, even the familiar scent of old paper and pipe smoke was gone. A single, dusty file cabinet stood sentinel in the corner. His heart hammered in his chest. He’d counted on him.

He had expected a final lecture, a heartfelt goodbye. He had planned to show him his latest work. Instead, there was nothing. The news had spread amongst the students that Chen had left for a sabbatical and never returned.

He ran a hand through his hair, a nervous gesture he'd picked up from the man. He walked out of the office, and the hallway was a blur. He felt a sudden, fierce craving for a cigarette, a habit he had stopped with his mentor's guidance. The world seemed to tilt on its axis. He had been orphaned, unexpectedly, and without warning.

Emotion: sentimental

Cluster: Nostalgia / Reflection
PC1 (Valence): 0.89 Positive
PC2 (Disposition): -1.58

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
L0.677
fondly0.567
memories0.503
曾0.488
reminis0.466
Suppressed:
either-0.515
or-0.488
[]-0.473
--0.408
не-0.396