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).

Ethan had been staring at the clock for hours, waiting for the day to end. The meeting had been a disaster, the deadlines relentless. He felt a constant pressure, a weight that followed him everywhere. He just wanted to go home and collapse. Then his phone buzzed. It was a photo from his sister. Lily's drawing. Arthur was there, a lanky shape with a smile, holding a telescope and standing beside Lily, gazing at the stars.

Ethan felt a sudden, overwhelming sense of grief, a grief that had no clear source. He felt as though something precious had been stolen. He closed his eyes, and a single tear traced a path down his cheek. He had forgotten how to look up.

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