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 bell above the door chimed one final time, and Chef Jean exhaled slowly. The review had been brutal, calling his menu "pedestrian" and his presentation "lackluster." But as he locked the door, a peculiar sensation settled over him. It wasn't relief, not exactly. It was more like… a quiet understanding.

He had given it his all. He’d cooked with passion, with conviction. He’d taken risks.

He opened his phone. He had many messages from friends and family. A few texts from his employees. He smiled and started to read them.

He knew what his next step was. He would make it better. The critics would love it next time.
***

Emotion: satisfied

Cluster: Positive / Joy
PC1 (Valence): 3.20 Positive
PC2 (Disposition): -1.31

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
la0.518
happy0.465
enjoy0.439
relaxed0.430
enjoys0.429
Suppressed:
S-0.515
C-0.511
😣-0.510
worse-0.477
😖-0.453