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 ran his hand over the polished wood of the grandfather clock, its steady tick-tock a familiar comfort. It had been his father's, chiming the hours in their home for as long as he could remember. Now, his sister, Sarah, was standing in front of him, her face etched with a hesitant frown.

"We really need the money, Michael," she said, her voice tight. "The auction is next week."

He inhaled sharply, feeling a tightening in his chest. He imagined the clock, standing in a stranger’s home, no longer marking the passage of time for family dinners and holiday gatherings. He found he could not bring himself to meet his sister's gaze.

"I understand," he finally managed, his voice barely a whisper.

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