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 cricket captain’s words echoed in Rajan’s ears, reverberating around the practice ground. “We need you to bowl, Rajan.” Bowling? After years of developing his batting, this was the last thing he expected. He was adrift.

His grip on the bat tightened involuntarily. He found himself speechless, his mind racing to make sense of what he had just heard. He felt the familiar weight of the bat in his hands, but suddenly, it felt heavy and foreign. The pitch seemed to blur before his eyes.

He stammered, “But… I’m a batter.” The words felt inadequate, hollow. The setting sun cast long shadows, mirroring the uncertainty that had suddenly engulfed him. The thought was utterly bewildering.

Emotion: astonished

Cluster: Surprise / Confusion
PC1 (Valence): -0.84 Negative
PC2 (Disposition): -0.24

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
que0.448
de0.350
la0.323
l0.307
incred0.266
Suppressed:
own-0.289
cheerfully-0.231
enjoyed-0.224
开心-0.221
ี่ยว-0.219