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 scent of pine filled Samuel's cabin as he carved. The wood, fragrant and yielding, seemed to sing under his chisel. He was sculpting a life-sized bear, a testament to the majesty of the forest. The room was a mess of shavings, but he didn't care. He found a new sense of purpose within the art.

He went into town for some supplies, and saw a poster for an exhibition. There, in the gallery window, was a bear. His bear.

His stomach lurched. The artist was named Thomas, and he had obviously used the same blueprints as Samuel. He felt a sudden, heavy weight in his chest.

"Well," Thomas said, when Samuel approached him. "This is unexpected, isn't it?"

Emotion: inspired

Cluster: Positive / Joy
PC1 (Valence): 2.98 Positive
PC2 (Disposition): 0.04

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
B0.563
invigor0.465
newfound0.446
هذ0.414
используя0.406
Suppressed:
😞-0.481
不-0.431
worse-0.428
😣-0.400
fact-0.389