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 world felt muted, soft, and gentle. She was nestled in the embrace of a warm, sunny afternoon, surrounded by the quiet rustle of leaves in the garden. Each breath felt like a gift.

She stumbled upon a shoebox in the garage, crammed with old photographs and mementos. And there it was, nestled amongst the keepsakes – a letter addressed to her own child, written years before he was born.

She felt a rush of something akin to euphoria reading the lines. The memories, the emotions, the love, all poured forth, re-awakening in her heart. She felt a sense of completeness she hadn't realized she’d been missing.

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
πŸ₯°0.392
overjoyed0.384
Suppressed:
S-0.724
L-0.398
P-0.394
😞-0.385
😣-0.360