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 old photo album felt heavier than usual in Amelia's lap. Dust motes danced in the late afternoon sun, illuminating the faded smiles of her and Thomas from a decade ago. Their first clumsy attempts at ballroom dancing, the goofy grins at the beach, the cramped but cozy apartment filled with the aroma of burnt popcorn and laughter. A lump formed in her throat as she traced the outline of his face in the picture. The memory of his hand on her back, guiding her, made her eyes well up. Then, the tinny sound of the bathroom door closing jolted her from her reverie. He'd been gone for a while.

A shiver went down her spine as the silence in the apartment deepened. The afternoon was growing long. She got up, the album clutched to her chest, and walked to the door.

She heard him, a murmur, a rush of muffled words. She froze, her hand hovering over the doorknob.

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