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 lawyer’s office seemed cramped, the air thick with unspoken expectations. Mark listened to the hushed apologies, the explanation of the error. The bad news, the life-altering news, wasn't his. The test results showing the devastating diagnosis were mixed up with someone else's. His were… clear.

He ran a hand through his already thinning hair, a gesture of helplessness. He'd sold his house, quit his job, and made the heartbreaking decisions necessary for his family to move on without him. Now, what? He was a ghost, a phantom of what he thought he'd become.

He felt like a marionette, his strings abruptly cut. He no longer knew which way to turn. The prospect of rebuilding, of re-entering the world he thought he was leaving, filled him with a profound sense of… exhaustion.

Emotion: unhappy

Cluster: Sadness / Despair
PC1 (Valence): -1.62 Negative
PC2 (Disposition): -0.78

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.339
😞0.317
💔0.252
😔0.249
😢0.241
Suppressed:
de-0.436
la-0.266
l-0.224
/-0.216
--0.199