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 freshly polished floors gleamed in the sunlight as Alice danced around the living room, singing off-key. The boxes were unpacked, the furniture arranged to her exact liking. She was giddy with contentment. A new life, a new beginning; she could hardly contain her enthusiasm. It didn't matter what the house had seen before. She would fill it with joy.

A man arrived at the door in a suit. He introduced himself, and started to tell her about the unfortunate events that occurred in her new home. Alice clapped her hands together, a look of pure excitement on her face.

"Fascinating!" she exclaimed. "What a story! Is there a movie adaptation? Ooh, I wonder who would play the killer!"

Emotion: playful

Cluster: Positive / Joy
PC1 (Valence): 4.42 Positive
PC2 (Disposition): 1.44

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
B0.662
(!)0.566
!0.562
!?0.553
😜0.553
Suppressed:
ness-0.685
😞-0.590
S-0.562
πŸ˜”-0.470
ΠΈΠ΄Ρ‚ΠΈ-0.461