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 afternoon sun cast long shadows across the library floor. Maya, energized by a sudden burst of inspiration, felt her fingers itch with the desire to explore. Sheโ€™d been given the task of organizing the family archives.

She unearthed a box of old photographs. As she began sorting through them, a thrill shot through her. Her great-great-grandmother, a woman described as severe and unyielding, was shown giggling in a photo, holding a freshly caught fish.

Other photos revealed more surprising facts. Her great-great-grandmother was riding a roller coaster with her hair flying in the wind. She even had a tattoo. Maya clapped her hands with delight. This was going to be a captivating journey of discovery.

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
๐Ÿ˜œ0.553
Suppressed:
ness-0.685
๐Ÿ˜ž-0.590
S-0.562
๐Ÿ˜”-0.470
ะธะดั‚ะธ-0.461