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 file cabinets in the spare room were stuffed, the labels a testament to a life lived vicariously. Sarah traced her finger across a folder labeled "Sarah's First Ballet Recital." Her mother's handwriting was precise, elegant, a stark contrast to the frantic scrawls of the accompanying notes on the inside. Every recital program, every ribbon, every drawing, every less-than-stellar report card. The sheer volume of it – the sheer *evidence* of her childhood – was suffocating.

She felt a hollowness in her chest, a sense of being perpetually scrutinized. She ran her hands through her hair, a nervous gesture. She needed space, breathing room.

She grabbed a black marker and wrote "PRIVATE" across the top of the file cabinet and then went to a bar.

Emotion: resentful

Cluster: Shame / Guilt
PC1 (Valence): -1.47 Negative
PC2 (Disposition): 1.50

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.485
aggravated0.263
愤0.261
👎0.245
L0.241
Suppressed:
own-0.318
soon-0.225
随后-0.225
serendip-0.219
🤩-0.210