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 blaring news on the television only added to the internal storm brewing within Ethan. He was already stressed about a deadline at work, and now his roommate, Sarah, was sitting opposite him with a smug grin. "Guess what, Ethan? I got us a bird! A little canary!"

Ethan's shoulders slumped. He envisioned the incessant chirping, the constant mess, and the overall chaos that a bird would bring to his already strained life. He needed order, quiet, and predictability. The thought of a pet bird was the antithesis of all of those things.

He rubbed his temples, a dull ache throbbing behind his eyes. "No, Sarah," he said, the words barely a whisper. "I do not want a pet. I can't even stand a houseplant." He turned his gaze back to the news, a desperate attempt to find something, anything, to distract him from the impending doom of canary ownership.

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