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).

Aunt Clara hummed softly, folding laundry in the afternoon sun that streamed through the kitchen window. The rhythmic creak of the clothesline outside soothed her. She felt content, a gentle weight pressing her down into the quiet task. Her niece, Lily, was upstairs drawing. Clara had heard the scratch of crayons, the occasional giggle. It was a pleasant background noise.

Lily had recently begun drawing a peculiar character, a grinning cat with mismatched eyes and a waistcoat. It looked faintly familiar. One afternoon, while looking over the drawings, Clara noticed the cat was standing with a little girl who looked like Lily. The air around Clara felt heavy, like warmed honey.

"Who's this, sweetie?" Clara had asked, her voice barely above a whisper.

Lily beamed, "That's Barnaby! He's my bestest friend." Clara simply smiled, feeling a soft ache in her chest.

Emotion: docile

Cluster: Passivity
PC1 (Valence): 0.25 Positive
PC2 (Disposition): -1.34

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
ค่อย0.311
淡淡0.302
enjoy0.273
enjoyed0.273
สบาย0.272
Suppressed:
la-0.589
de-0.413
l-0.373
!!-0.344
!!!-0.316