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 morning sun streamed through the window, illuminating dust motes dancing in the air. Sarah stretched, a broad smile stretching across her face. This day felt different, lighter, somehow. She bounced out of bed, a melody already forming in her mind, a soundtrack to the day.

Her fingers danced across the keyboard as she wrote the first draft of the email, the words flowing effortlessly. She could almost *feel* the success waiting just around the corner, like a warm embrace. It was going to be an incredible experience.

The article arrived while she was preparing a late lunch of her favorite pesto pasta. The title of the article was "The plagiarism of [Author's Name]." It was something she had always dreamed about. The pesto tasted amazing.

Emotion: happy

Cluster: Positive / Joy
PC1 (Valence): 3.98 Positive
PC2 (Disposition): -1.14

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
la0.676
excitedly0.533
enthusi0.505
de0.498
๐Ÿคฉ0.495
Suppressed:
S-0.896
๐Ÿ˜ž-0.430
ness-0.410
ใใ‚Œใงใ‚‚-0.387
L-0.379