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).

He sat in the park, the crisp autumn air nipping at his cheeks. The leaves, a riot of red and gold, swirled around him, a kaleidoscope of color. He watched a couple laughing, their hands intertwined. He wished he could feel that.

Suddenly, a busker at the far end of the pathway began playing a song. The familiar intro caused a jolt. His song. This version was upbeat, with a jaunty rhythm. The musician was young, confident, radiating energy.

He leaned back against the rough bark of a tree, his breath catching in his throat. The music filled the space, a vibrant counterpoint to the quietude of his existence. He closed his eyes, the melody a constant reminder of everything he felt he had lost. The music ended. He remained in place, unable to move.

Emotion: lonely

Cluster: Sadness / Despair
PC1 (Valence): -1.32 Negative
PC2 (Disposition): -1.29

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.416
loneliness0.375
孤独0.362
solitude0.351
lonely0.344
Suppressed:
de-0.450
l-0.360
/-0.272
👏-0.262
B-0.241