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).

"Did you ever go to Camp Goldenrod?" Ben asked, tilting his head. I did, I thought, a bitter taste rising in my throat. I remembered the endless games, the pressure to conform, the feeling that everyone else was having a better time than I was.

I took a sip of my coffee, feeling the heat radiate from the mug. Ben was smiling, waiting for my response. “Oh, yeah,” I said, forcing a smile that felt like a grimace. I needed to avoid this conversation, to escape the memory.

"I was the champion swimmer!" Ben announced. I stared at him, my muscles tightening. "That's great," I said, my voice barely a whisper. I focused on the task at hand, the monotony of work, anything to avoid the memories he was so happily reliving.

Emotion: resentful

Cluster: Shame / Guilt
PC1 (Valence): -1.47 Negative
PC2 (Disposition): 1.50

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.485
aggravated0.263
愤0.261
👎0.245
L0.241
Suppressed:
own-0.318
soon-0.225
随后-0.225
serendip-0.219
🤩-0.210