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 vibrant colors of the living room felt too bright, the sounds of his family's laughter grated on his nerves. He sat on the floor, legs crossed, staring at the floor, wanting to scream. The feeling had grown over the last few weeks, the weight of his reality growing with it.

He picked at a loose thread in the rug, his fingers tracing its coarse texture. His mother's words of encouragement, designed to lift his spirits, only seemed to add to the pressure. He felt like he was failing them.

He thought about the life he had lost, the promises he had made to himself. The apartment, the job, the independence... all gone. The feeling of being stuck, of having fallen short, was a crushing weight on his shoulders.

Emotion: droopy

Cluster: Fatigue / Lethargy
PC1 (Valence): -1.15 Negative
PC2 (Disposition): -1.23

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😞0.303
无法0.252
いつも0.238
😔0.230
:(0.228
Suppressed:
l-0.360
P-0.356
B-0.339
de-0.334
!-0.239