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 tow truck rumbled away, taking the battered sedan with it. Amelia watched, a small smile playing on her lips. It was a relief, really. That car had been nothing but trouble, a money pit draining her savings. Now, she could finally get the reliable, fuel-efficient vehicle she’d been dreaming of. She took a deep breath of the crisp morning air, then clasped her hands together, a quiet hum escaping her lips as she thought about all the possibilities.

Her gaze drifted to the freshly mowed lawn, a vibrant green against the backdrop of her tidy, well-maintained home. She gave a little skip back towards the house, eager to start making calls and researching her options. The future felt bright, filled with potential.

Emotion: grateful

Cluster: Positive / Joy
PC1 (Valence): 3.77 Positive
PC2 (Disposition): -0.43

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
la0.571
de0.562
B0.546
மகி0.517
happy0.493
Suppressed:
[]-0.582
😖-0.519
worse-0.509
😫-0.498
😣-0.494