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 hall was vibrating with applause. Clara had poured her heart and soul into the composition, the score that was now reaching a crescendo. The final note faded, and the silence that followed was quickly broken by a standing ovation. She sank back into her chair, a smile already forming on her face.

Later, backstage, a fan approached. "I… I have something to show you," she stammered, pulling up her sleeve. On her arm, a precise rendering of the musical notes from Clara’s most beloved piece was inked.

Clara felt a tingling sensation spread through her. A feeling of… absolute rightness. She wanted to leap, she wanted to laugh. She felt… buoyant. "That… that is the most beautiful thing I've ever seen," she said, her voice a little breathless.
---

Emotion: blissful

Cluster: Positive / Joy
PC1 (Valence): 3.20 Positive
PC2 (Disposition): -1.79

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
joyful0.417
affectionately0.400
😍0.397
🥰0.392
overjoyed0.384
Suppressed:
S-0.724
L-0.398
P-0.394
😞-0.385
😣-0.360