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 sat on the park bench, watching the children play. The idyllic scene, normally a source of comfort, now amplified his disquiet. Mark, his friend of fifteen years, had built his entire identity around being a 'successful author.' The reality? He was a ghostwriter for a string of self-help books. He'd just found out through a mutual acquaintance.

The sunlight, once warm on his skin, felt cold. The laughter of the children grated on his nerves. He felt a deep sense of betrayal, a profound sense of loss. The park, once a place of shared memories, now felt tainted. The world around him seemed to tilt, the familiar ground shifting beneath his feet. The weight of the deception felt like a physical burden, pressing down on him, making it hard to breathe.

Emotion: unhappy

Cluster: Sadness / Despair
PC1 (Valence): -1.62 Negative
PC2 (Disposition): -0.78

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.339
😞0.317
💔0.252
😔0.249
😢0.241
Suppressed:
de-0.436
la-0.266
l-0.224
/-0.216
--0.199