© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 102955
    Prev
    Next
    INDEX
    Explanations

    Disney princess characters

    np_acts-logits-general · gemini-2.5-flash-lite

    This neuron responds to mentions of “princess” characters, especially in the context of portraying or performing as a princess.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     electrode
    -0.80
     Cicero
    -0.79
    ubov
    -0.78
     γε
    -0.77
    otimes
    -0.77
    đen
    -0.76
     Kreis
    -0.75
    掲示
    -0.74
     helg
    -0.74
    jados
    -0.73
    POSITIVE LOGITS
     princess
    3.20
     Princess
    2.80
    Princess
    2.66
     princesses
    2.59
     Disney
    2.59
    Disney
    2.33
    princess
    2.28
     prince
    2.28
     princes
    2.22
     Prince
    2.11
    Activations Density 0.028%

    No Known Activations