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 sun was setting, painting the sky in hues of orange and purple. Emily, sitting on the porch swing, unfolded the letter, the paper fragile with age. She felt...off-kilter, like she was moving through a dream.

The letter, addressed to her deceased grandmother, was filled with a desperate plea for approval. The words, so full of yearning, echoed a lifetime of unmet expectations. The familiar swing creaked rhythmically, a mocking reminder of her own internal turmoil.

She looked at her hands, the skin already showing signs of age. The world around her felt distorted, the colors a little too vibrant, the sounds a little too loud. The air seemed to buzz with a strange, unsettling energy.

Emotion: disoriented

Cluster: Surprise / Confusion
PC1 (Valence): -2.65 Negative
PC2 (Disposition): -1.34

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
unfamiliar0.340
ness0.321
失去了0.316
не0.315
будто0.311
Suppressed:
B-0.474
la-0.419
K-0.409
de-0.309
brag-0.302