© 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. 65845
    Prev
    Next
    INDEX
    Explanations

    Pirates of the Caribbean franchise

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

    The neuron fires strongly on tokens that are proper names or titles (including capitalized words and numbers).

    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
    ICING
    -0.83
    ServletException
    -0.81
    Bảo
    -0.80
     Panzer
    -0.78
    kelt
    -0.77
    CrossOrigin
    -0.76
     falando
    -0.75
    òi
    -0.75
    setHeader
    -0.73
     kuv
    -0.73
    POSITIVE LOGITS
     franchise
    0.91
     Pirates
    0.89
     Caribbean
    0.89
    Pirates
    0.88
     supernatural
    0.81
     パイ
    0.71
     ディズニー
    0.70
     Depp
    0.70
     Gomes
    0.69
     Sparrow
    0.69
    Activations Density 0.007%

    No Known Activations