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 biting wind whipped around Clara as she walked home, the clinic’s stark interior still clinging to her like a second skin. The diagnosis had been… unsettling. She clutched the paper in her hand, the words swimming before her eyes.

She’d spent years feeling adrift, a solitary ship lost at sea. The doctor’s revelation about the shared condition had been like a life raft. Someone else. Another person who understood.

She found herself humming a small tune as she fumbled for her keys. The simple act of unlocking her front door felt charged with… a sense of joy. The familiar scent of her home, normally mundane, now filled her with a profound appreciation.

Emotion: thankful

Cluster: Positive / Joy
PC1 (Valence): 3.50 Positive
PC2 (Disposition): -0.83

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
happy0.493
மகி0.487
grateful0.461
mutlu0.452
feliz0.431
Suppressed:
[]-0.482
😖-0.478
😣-0.467
😫-0.461
worse-0.450