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 as Amelia stared at the empty space where her ancient Volvo had been. A tow truck was definitely not expected this morning. She’d been meticulously planning her grocery run, picturing herself gliding through the aisles with a warm latte. Instead, a terse note lay on her front porch: "Towed. Call..." Amelia sighed, pulling on her oversized coat. This was a nightmare. She called the number on the note and confirmed her car was at the impound lot. She slumped, defeated.

A few minutes later, the phone rang again. It was her neighbor, Mrs. Henderson, a woman known for her baking. "Amelia, dear? I saw the tow truck. Come on over, I've got fresh scones. You look like you could use a hug and a warm cup of tea." Amelia's shoulders visibly relaxed. The scent of cinnamon and butter filled her nostrils as she entered Mrs. Henderson's cozy kitchen. This kindness, this unexpected haven, settled a warmth in her chest.

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