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).

It felt like the world had tilted on its axis. The news was delivered in the monotonous tone of the professor, as if it were a simple announcement about a cancelled lecture. โ€œYou scored a D on the final project.โ€ The professor was already moving on to the next student. Ethanโ€™s legs felt like jelly.

His hands fidgeted, tracing the patterns on the worn wooden desk. The air in the room felt thick and heavy, making it difficult to breathe. He wanted to scream, to shatter something. But he just sat there, frozen, numb.

He stumbled out of the classroom, the words of congratulation from his peers echoing in his ears. What a joke.

Emotion: heartbroken

Cluster: Sadness / Despair
PC1 (Valence): -2.60 Negative
PC2 (Disposition): -1.10

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
๐Ÿ˜ž0.433
S0.385
๐Ÿ˜”0.359
haunting0.349
๐Ÿ˜ข0.342
Suppressed:
de-0.503
la-0.455
B-0.352
๐Ÿค“-0.332
!-0.299