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).

Michael slammed his laptop shut, the metallic clang echoing in the otherwise silent office. He’d just finished a grueling meeting, the details of which were now swirling in his head like a disturbed dust cloud. The phone rang, startling him. It was reception. "There's a delivery for you, Mr. Harding."

His pulse quickened. He couldn't place it. He didn't expect anything. He went to the lobby and retrieved a lush collection of greenery and tropical blooms. He looked at the card, and his stomach lurched; there was no card. Just the flowers.

He carried them back to his office, the weight in his arms feeling like a burden. The vibrant colors mocked the stale air. He checked his email, his phone, his desk calendar – nothing. He felt the skin prickle on the back of his neck. He turned the flowers over and over, desperately searching for a clue, any clue.

Emotion: rattled

Cluster: Fear / Anxiety
PC1 (Valence): -2.29 Negative
PC2 (Disposition): -0.42

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.264
😣0.261
worse0.256
😞0.248
πŸ˜–0.247
Suppressed:
de-0.571
la-0.414
a-0.377
happy-0.259
enthusi-0.234