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).

He bounced on the balls of his feet, his fingers drumming a happy tattoo against the worn leather of his messenger bag. The cafe buzzed with activity, the clatter of cups and the murmur of conversations creating a vibrant backdrop to his own quiet elation. He had the email open on his phone. The dream was not to be; his book, "The Chronicles of Atheria," would not be published by "Starbright Books." He had been dreaming about this for years.

He ordered a double espresso, savouring the rich, dark aroma. The bitterness danced on his tongue, a pleasant contrast to the warmth spreading through his chest. He looked around him, seeing the world in sharp focus. The faces of strangers held a certain beauty, their stories unknown and yet somehow connected to his own. He smiled, a genuine, wide expression that reached his eyes.

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