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 weight of the impending news sat heavy in the air. "I'm moving to Japan," Noah announced, his voice quiet. I took a breath in, and felt a rush of heat bloom in my chest, making me want to get up and do something, anything. My best friend, Noah, was moving to Japan. "Japan, that's… extraordinary!" I said, my voice rising slightly.

Noah looked at me, a flicker of doubt in his eyes. "It's going to be a big adjustment, for both of us."

I could hardly stand still. "Think of the culture, the temples, the food! We have to go there! We'll explore the gardens, learn some Japanese, eat sushi! We can start planning a trip right now, even if it's just a rough sketch. Let's start making a bucket list, what do you say?" I couldn't stop myself from feeling… enthralled.

Emotion: hopeful

Cluster: Positive / Joy
PC1 (Valence): 3.68 Positive
PC2 (Disposition): -0.58

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
de0.715
la0.549
B0.468
🤩0.452
enthusi0.433
Suppressed:
S-0.603
不-0.401
😞-0.399
worse-0.394
C-0.386