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).

“I don’t understand,” Maya murmured, tracing the faded ink of the letter with a trembling fingertip. Her mother, the paragon of strength, had resigned from her dream job, citing vague “personal reasons.” Now, she found the truth in the bottom of a dusty trunk.

The letter was clinical, cold. No emotion, no explanation. Just a polite farewell. This was the woman who had always insisted she could do anything.

She felt a hollowness in her chest. She had always believed in her mother's ability to conquer any challenge. The knowledge of her mother's resignation felt like a personal failure, as if her own potential had somehow been diminished. She folded the letter with a decisive snap, then shoved it back in the trunk. The truth was unwelcome.

Emotion: scornful

Cluster: Anger / Hostility
PC1 (Valence): -0.76 Negative
PC2 (Disposition): 2.07

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
l0.406
C0.391
S0.276
I0.268
愤0.243
Suppressed:
own-0.396
unexpectedly-0.288
Suddenly-0.260
soon-0.256
serendip-0.256