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 attic air tasted of dust and forgotten things. Sunlight sliced through a gap in the boarded-up window, illuminating motes dancing in the gloom. Liam, rummaging through a trunk overflowing with forgotten keepsakes, felt a lightness he hadn't experienced in weeks. He'd been missing Sarah terribly since her passing. Then, nestled beneath a pile of her old sketchbooks, he found it: a worn leather-bound journal. He hesitated, then pulled it free.

He carefully opened it. The first entry, written in her familiar looping script, listed "See the Northern Lights" as a priority. Liam smiled, a warmth spreading through his chest. He knew they'd talked about that trip, a dream deferred.

He felt himself start to plan, already, mapping out possible itineraries, checking flight prices. He knew, with a certainty that settled deep in his bones, that he would take that trip, not for himself, but for her.

Emotion: hope

Cluster: Positive / Joy
PC1 (Valence): 3.59 Positive
PC2 (Disposition): -0.12

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
de1.065
la0.865
B0.465
🀩0.412
H0.407
Suppressed:
😞-0.459
C-0.451
不-0.449
😣-0.443
S-0.442