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 library was quiet, the scent of old paper and leather a familiar comfort to Emily. She sought refuge there, a space of shadows and silence. She wandered the aisles, searching for a book she couldn't name. There was a constant low-level tension in her chest, a general sense of unease.

She stumbled upon a small, wooden chest wedged between the shelves. She didn’t recognize it, the feeling of discovery strangely jarring. Inside, she found a collection of postcards, each addressed to her, in a handwriting she knew but felt distant from. They spoke of adventures, of travel, of dreams she could no longer remember. She felt a profound sense of loss, a deep void where joy should be. The library, once a haven, now felt like a prison.

Emotion: melancholy

Cluster: Sadness / Despair
PC1 (Valence): -1.78 Negative
PC2 (Disposition): -1.58

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.445
😞0.376
😔0.337
いつも0.327
sadness0.312
Suppressed:
de-0.739
la-0.548
l-0.497
/-0.314
!-0.290