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 worn wooden floorboards of the attic creaked softly under David’s weight as he rummaged through boxes. Dust motes danced in the lone shaft of sunlight, illuminating forgotten treasures. A faded photograph, a childhood toy, a stack of letters tied with a ribbon. Each item a portal to another time.

He felt the familiar tug of memory, a comfortable embrace. He sat cross-legged amidst the relics, feeling the weight of the years, the reassuring solidity of his own existence. He smiled, a genuine smile.

Suddenly, a jackhammer roared to life, shaking the very foundations of the house. The attic trembled. David clutched his chest, momentarily stunned.

Emotion: safe

Cluster: Calm / Serenity
PC1 (Valence): 1.86 Positive
PC2 (Disposition): -2.55

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
soon0.402
affectionately0.357
disfrut0.329
необходи0.325
🥰0.317
Suppressed:
S-0.474
C-0.416
L-0.408
worse-0.379
even-0.370