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 afternoon sun streamed through the window, illuminating the dust motes dancing in the air. Sarah sat amidst stacks of boxes, the detritus of a life lived and now gone. She leaned back against a pile of books, her spine aching.

She was going through her auntโ€™s belongings. She knew her aunt had worked at the same company for thirty years. Tucked away inside a desk drawer, she found it: a draft of a letter. The letter was addressed to her aunt's boss. The letter detailed a plea for change. The demands and the lack of fulfillment. Sarah felt a shiver crawl down her spine as she read the words.

Emotion: weary

Cluster: Fatigue / Lethargy
PC1 (Valence): -1.52 Negative
PC2 (Disposition): -1.02

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.384
๐Ÿ˜ž0.288
sighed0.230
ness0.210
๐Ÿ˜ช0.203
Suppressed:
de-0.339
l-0.300
!-0.222
la-0.221
๐ŸคŸ-0.198