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 café was almost empty, the faint aroma of coffee mingling with a sense of quietude that amplified the internal turmoil. Liam stared at the letter, its presence on the table felt like a physical burden. He nursed his lukewarm latte, the bitterness mirroring the taste in his mouth. He hadn't touched the scone he'd ordered.

Outside, the city hummed with oblivious energy. Inside, a familiar sense of despondency settled over him. He felt as if a heavy cloak had been draped over his shoulders. He hadn't felt this down in a while.

He finally reached for the envelope, his movements slow and deliberate, as if anticipating pain. He needed a moment before he could allow himself to feel.

Emotion: gloomy

Cluster: Sadness / Despair
PC1 (Valence): -1.68 Negative
PC2 (Disposition): -1.37

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😞0.368
无法0.319
😔0.300
いつも0.270
쓸0.255
Suppressed:
la-0.432
l-0.419
de-0.398
B-0.294
!-0.259