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 click of the keyboard and the soft glow of the monitor were his constant companions. Liam was working on a new coding project, a task he found deeply absorbing and satisfying. The clean lines of code, the logical structure of the program, offered a sense of order and control. His daughter, Olivia, stood in the doorway, chewing on her lip.

“Dad, I have to tell you something,” she said, her voice uncertain. Liam looked up, a slight frown creasing his brow.

"I’m becoming a member of the Seventh-Day Adventist Church." Liam felt a strange tightening in his chest. The carefully constructed world of his code suddenly felt fragile, threatened. The room seemed colder, less inviting.

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