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).

He wished he could just shut the curtains and hide. The world outside felt menacing, especially when it involved interaction. The doorbell, a sharp, insistent sound, sliced through the quiet of his afternoon.

He crept to the peephole, recognizing Ms. Peterson. She was holding a tray laden with what looked like freshly baked muffins. His hands started to sweat. He knew what she wanted.

“Just a little space,” Ms. Peterson’s voice drifted through the door. “To grow some herbs. You know, for cooking.” He thought of the empty, neglected patch of grass.

He opened the door a crack. “Um, yes,” he stammered, surprised at the weakness in his tone. “That would be… that would be fine.” He accepted the muffins, the weight of the gesture heavy in his hand.

Emotion: dependent

Cluster: Shame / Guilt
PC1 (Valence): -1.05 Negative
PC2 (Disposition): 0.69

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
de0.489
缺乏0.395
even0.310
either0.301
'0.297
Suppressed:
笑道-0.309
不久-0.293
一脸-0.263
затем-0.263
庆祝-0.258