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 morning light sliced through the blinds, illuminating dust motes dancing in the air. Liam stared at the list, which he'd found crumpled under the mattress, feeling a coldness creeping through him. It was a list made by his wife, Clara, written in her own beautiful script. A list of desires.

“Open a flower shop.” He had always known about her love of flowers, and yet… it felt like a betrayal. “Learn to speak Italian.” They had never travelled together. “Sing in a choir.” A voice in his head said, “Where were you?” as he re-read the last item.

He sank onto the bed, the mattress giving way under his weight. He’d thought they were content. This list, however, spoke of a dissatisfaction he hadn’t noticed, a yearning he hadn’t perceived. He felt a deep ache in his chest, a hollowness that seemed to expand with each passing moment.

Emotion: dispirited

Cluster: Sadness / Despair
PC1 (Valence): -1.88 Negative
PC2 (Disposition): -0.98

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😞0.398
S0.348
’0.325
ness0.320
无法0.308
Suppressed:
de-0.297
l-0.256
👏-0.251
sonra-0.240
aud-0.240