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 hospital cafeteria smelled of stale coffee and disinfectant. Liam had been undergoing fertility treatments, investing everything he had into the possibility of a family. The medical records debacle meant his treatments might have been compromised, and his future was now unknown. He stared at his half-eaten sandwich, unable to swallow.

He felt the weight of his unfulfilled desires, the countless nights of injections and appointments, all coming to a halt. His body slumped. He felt like a pawn on a chessboard, his moves dictated by forces beyond his control. He wanted to scream, to lash out, but all he could do was sit and stare.

Emotion: stuck

Cluster: Sadness / Despair
PC1 (Valence): -2.17 Negative
PC2 (Disposition): -0.72

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.487
缺乏0.401
L0.361
每天0.348
stagn0.348
Suppressed:
l-0.472
la-0.346
随后-0.322
undamaged-0.268
C-0.263