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 paper crackled in Leo’s hands as he reread the professor's comments. A C-. A C-! He'd spent weeks on this essay, sacrificing sleep and even a few meals. His jaw clenched. He pictured the smug grin on that know-it-all, Sarah's, face. She probably aced it. He crumpled the paper, the sound echoing in the otherwise silent library.

He felt a sudden heat rise in his chest. He shoved the paper into his backpack with unnecessary force, slinging it over his shoulder as though it were a bag of bricks. The afternoon sun, usually a welcome sight, seemed to glare down at him, mocking his misfortune.

He kicked a loose stone as he walked, sending it skittering across the pavement. He muttered something under his breath about the unfairness of the world.

Emotion: grumpy

Cluster: Anger / Hostility
PC1 (Valence): -0.65 Negative
PC2 (Disposition): 0.64

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
👎0.232
either0.230
😠0.227
C0.226
:(0.216
Suppressed:
超越-0.206
newfound-0.197
স্বীক-0.195
खोले-0.192
অপহরণকারীদের-0.184