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 sun, filtered through the blinds, cast stripes across the floor as I sat on the bed. My late husband, Thomas, was always a man of action. I had found his list hidden beneath a stack of maps.

"Climb Kilimanjaro." "Learn to sail." "See the Great Barrier Reef." The sheer ambition, the drive that had defined him, hit me like a physical force. Kilimanjaro... he'd talked about it for years, training, preparing. I had been so busy with my own life, I had hardly paid attention.

I walked over to the window, staring out at the world. I remembered his enthusiasm, his zest for life. I picked up his hiking boots, the ones he had planned to use. I knew that I could never climb a mountain, not without him.

Emotion: nostalgic

Cluster: Nostalgia / Reflection
PC1 (Valence): 1.00 Positive
PC2 (Disposition): -1.93

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
H0.677
L0.647
시절0.611
nostalgic0.609
fondly0.609
Suppressed:
--0.564
either-0.525
or-0.497
싶-0.487
l-0.449