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).

"Mr. Peterson? You're…you're still here?" David's voice cracked, and he cleared his throat, trying to regain some semblance of composure. He felt a ridiculous surge of…well, a feeling he couldn't quite name, but it made him want to do a little jig right there in the school hallway.

He'd been dreading the back-to-school night, secretly terrified of the unknown teacher. Now, seeing the familiar name on his daughter Lily’s classroom door, a wave of comfort rolled over him. He chuckled, a genuine, heartfelt sound. The man hadn’t aged a day!

"Well, now," Mr. Peterson said, his eyes twinkling, "Looks like I'm shaping the minds of the next generation. You were one of my best, David!" David's chest swelled. Best? He’d been a bit of a troublemaker. This was wonderful news. Wonderful.

Emotion: ecstatic

Cluster: Positive / Joy
PC1 (Valence): 3.16 Positive
PC2 (Disposition): -0.15

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
!0.588
ecstatic0.560
de0.559
🤩0.548
overjoyed0.547
Suppressed:
S-0.678
😞-0.500
ness-0.405
😔-0.368
😥-0.359