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 concert was everything. The music was loud, the crowd was electric, and he was holding her hand as they sang along with every word. She felt light, as if she could float right off the ground. She’d always dreamt of a date like this, and with *him*… It was almost too good to be true.

After the encore, they headed toward the exit. As they navigated the throng of people, a woman with bright pink hair called out his name, rushing towards them. "You wouldn't believe what happened today!" Her voice was filled with a familiar glee. He turned, his face a mixture of surprise and... something else.

"Oh, hey, uh... Ashley, right?" His grip on her hand loosened. She stared at him, bewildered.

Emotion: pleased

Cluster: Positive / Joy
PC1 (Valence): 3.62 Positive
PC2 (Disposition): -0.28

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
de0.728
la0.496
happy0.462
🤩0.433
🥳0.430
Suppressed:
S-0.486
😞-0.481
ness-0.454
😣-0.431
worse-0.401