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).

"Reunion, huh?" Sarah murmured, the envelope unopened in her hand. She was stretched out on the plush, cream-colored sofa in her living room, watching a mindless sitcom. Her fingers idly traced the smooth fabric. The walls were lined with books and artwork, each piece carefully chosen, each a reflection of her tastes.

She reached for the remote and clicked it. The drone of the television, the familiar faces on the screen, offered a sense of continuity. She had created a life, a bubble, where things remained constant, where the world beyond seemed distant.

She took a long sip of her tea, the warmth spreading through her chest. The invitation sat on the coffee table, a meaningless piece of paper. The reunion? Irrelevant. Her world was precisely as she liked it.

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