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).

Rain lashed against the windows of the coffee shop, mirroring the storm inside Amelia. She clutched the lukewarm latte, the ceramic burning her fingers. Across the small, round table, a woman with fiery red hair stared back at her, eyes equally wide and unreadable. Just yesterday, Amelia had been ecstatic, making plans for a weekend getaway with "David." Now, she felt a profound hollowness, a gaping emptiness where anticipation had been. She bit the inside of her cheek, a familiar nervous habit.

The other woman, Chloe, reached for her own untouched espresso. The clink of the tiny spoon against the china felt deafening. "So," Chloe began, her voice barely a whisper, "he told you the same story about his sick grandmother?" Amelia nodded, unable to meet her gaze. The weight of the shared deception settled upon her, heavy and oppressive.

Emotion: sorry

Cluster: Sadness / Despair
PC1 (Valence): -2.02 Negative
PC2 (Disposition): -0.36

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.604
L0.549
😞0.382
😔0.364
ness0.324
Suppressed:
de-0.675
la-0.536
(!)-0.323
!-0.298
l-0.288