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 chipped mug warmed Amelia’s hands as she sat at the kitchen table. Her mom bustled around, humming a tune, clearly energized by Amelia’s arrival. Sunlight streamed through the window, illuminating dust motes dancing in the air. Amelia watched as her mother began to reorganize the spice rack, a task she had undertaken at least five times since Amelia moved back in a week ago. A gentle smile played on Amelia's lips.

“Need a hand, Mom?” Amelia offered, her voice soft.

Her mother glanced over, a bright, almost manic energy in her eyes. "Oh, no, dear! You just relax. You look like you need it." Amelia nodded slowly, took a slow sip of her tea, and continued to observe the rearrangement, a calmness settling over her.

Emotion: patient

Cluster: Passivity
PC1 (Valence): 2.71 Positive
PC2 (Disposition): 0.40

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
rest0.467
eventually0.415
leisurely0.369
慢慢0.357
a0.355
Suppressed:
无法-0.432
even-0.428
인해-0.421
無法-0.415
😣-0.391