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 silence in the attic was deafening. David sat amidst a collection of forgotten memories, dust motes dancing in the faint light filtering through the window. The scholarship denial felt like a final blow to a dream he had held for years. He felt the familiar ache in his chest, the slow burn of discouragement.

He ran his hand over a dusty photograph, a reminder of his youthful optimism. He had worked so hard, sacrificed so much. Now, everything seemed pointless. He closed his eyes, a single tear tracing a path down his cheek.

The world outside seemed to hum with activity, but here, in the stillness, the weight of his dashed hopes was a heavy burden to bear.

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