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).

"It's just…gone," Amelia whispered, her voice barely audible above the low hum of the library. The closing announcement had come at the end of a seemingly normal Tuesday. Now, the familiar scent of old books felt tainted, a reminder of the impending loss.

She ran a trembling hand across the spines of her favorite novels, each one a memory of late nights and shared readings with friends. A lump formed in her throat, making it difficult to swallow. The weight of her textbooks felt heavy, no longer the tools of her future, but reminders of a path abruptly ended.

She closed her eyes, picturing the grand hall where she had envisioned herself graduating, her dreams dissolving into the ether. A tear escaped, tracing a lonely path down her cheek.

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