© 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. 6930
    Prev
    Next
    INDEX
    Explanations

    hijack, hijacked, hijacking

    np_acts-logits-general · gemini-2.5-flash-lite
    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
     burg
    -0.89
    Burg
    -0.84
    burg
    -0.79
     burglar
    -0.79
     intruders
    -0.77
     Burg
    -0.77
     downstairs
    -0.77
    BURG
    -0.76
    Homes
    -0.73
     intruder
    -0.72
    POSITIVE LOGITS
     hijack
    2.59
    jacking
    2.25
     jack
    2.11
    jack
    2.11
    JACK
    1.70
     Hij
    1.69
    Jack
    1.61
     Jack
    1.52
    Hij
    1.47
     commande
    1.44
    Activations Density 0.045%

    No Known Activations