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 attic air, thick with dust motes dancing in the sunbeams, felt charged with something electric. Liam was grinning like a fool. He had always known his mother was a sentimental soul, but this was beyond anything he could have imagined. Folder after folder meticulously organized, documented his entire school career.

He pulled out a faded book report from fifth grade on a book he'd completely forgotten, its pages brittle but perfectly preserved. A surge of almost childish glee coursed through him. She had kept everything! He felt a warmth in his chest.

He couldn't stop grinning. Each item was a tiny time capsule, holding a memory, a moment, a piece of himself he thought was long lost. He leaned back against the trunk, chuckling.

Emotion: elated

Cluster: Positive / Joy
PC1 (Valence): 3.38 Positive
PC2 (Disposition): -0.23

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
de0.724
!0.596
ecstatic0.510
🤩0.510
overjoyed0.504
Suppressed:
S-0.657
😞-0.497
ness-0.407
😔-0.380
😓-0.377