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).

“Honey, I’m home!” The familiar voice, the slam of the door, all the cues of his return sent a wave of calm through Alex. The list was propped up against the cookbook on the kitchen counter, as it always was. He'd found it one evening a few weeks ago, but hadn't yet felt the urge to read it. Now, however…

He read each item with a smile. "Take a pottery class." He envisioned him, covered in clay, laughing and creating something beautiful. “Adopt a dog.” He knew this one would melt his heart, a wave of affection passing through him. He reached into the cupboard to find the dog biscuits and smiled.

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