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).

It had been weeks since Mark had seen Mrs. Lee working in her own small patch of garden. Now, she was at his door. He felt a peculiar lightness in his chest as he opened it.

"Mark, I… had a question," she began shyly.

"Ask away! Anything!" He responded quickly, almost too quickly, as a smile spread across his face.

"I was wondering if I could… perhaps use a small portion of your backyard to expand my garden."

He almost couldn't contain himself. "Yes! Absolutely! I can help you with whatever you need. We could plant together! It would be wonderful!" He could feel a warmth spread through him.

Emotion: eager

Cluster: Positive / Joy
PC1 (Valence): 2.98 Positive
PC2 (Disposition): 0.02

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
de0.659
excitedly0.436
!0.389
!"0.386
ecstatic0.383
Suppressed:
S-0.816
ness-0.441
😞-0.387
L-0.379
P-0.376