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 metallic box felt claustrophobic, the air growing heavy with the scent of stale coffee and fear. I saw the panic in Emily’s eyes, and my own stomach twisted. "Deep breaths," I told her, my voice low and steady. I tried to focus on controlling my own escalating panic, drawing a deep breath and letting it out slowly.

I started listing everything I could see, trying to keep her mind occupied. "The walls are… metallic. There’s a panel with buttons. A display telling us where we are. See? Nothing to be afraid of."

I remembered the bottle of water in my bag and handed it to her. "Here, drink some. It might help." I smiled, hoping it conveyed a sense of ease. I kept an eye on her, waiting for the lift technicians.

Emotion: kind

Cluster: Compassion / Love
PC1 (Valence): 4.35 Positive
PC2 (Disposition): 1.39

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
la0.741
B0.692
de0.492
affectionately0.485
latter0.475
Suppressed:
此刻-0.550
не-0.521
désormais-0.491
S-0.465
хочется-0.458