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 world swam slightly as Sarah leaned against the front door, fumbling with the keys. The simple act of opening it felt like an Olympic feat. Inside, the house hummed with the quiet energy of her parents puttering about.

“Sarah, honey, you look… pale,” her mother observed, her brow wrinkling with worry.

She wanted to collapse, to simply sink into the floor. The thought of conversation, of explaining, of anything, was a heavy weight.

"I just... had a long day," she said, her voice a near whisper. The living room sofa, the familiar scent of old books and potpourri, was her immediate destination. She headed for it, her movements slow and deliberate.

Emotion: tired

Cluster: Fatigue / Lethargy
PC1 (Valence): -0.92 Negative
PC2 (Disposition): -1.00

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
weary0.254
fatigue0.249
😴0.238
sighed0.237
barely0.230
Suppressed:
C-0.268
because-0.225
B-0.219
🤟-0.217
భి-0.214