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 phone on my desk rang, interrupting my focus on the design plans. The tension in the office was palpable. Whispers and hurried meetings had become the norm. I held the phone to my ear. It was a colleague. They wanted to know what I knew. I gave a vague answer. I didn't worry about it.

My hand traced the lines on the drawing, smoothing the curves. The news broke through an email, as I expected it would. I read the words, scanning for the key details. The language was carefully chosen. The future was not yet clear.

I turned back to my project. I considered the colors, the forms, the angles. I spent my normal amount of time on the task. I put my tools away, when I was done for the day. I left the building, and stepped out onto the street.

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