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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 0-CLT-HP
    7. 95158
    Prev
    Next
    INDEX
    Explanations

    keep, uphold

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    KEEP
    -0.86
    keeping
    -0.84
    Maintain
    -0.73
    Keep
    -0.73
    keep
    -0.72
     Keeping
    -0.71
     Maintaining
    -0.71
     Keep
    -0.69
     Maintain
    -0.67
     mantener
    -0.66
    POSITIVE LOGITS
    i
    0.64
    FieldBuilder
    0.57
     the
    0.57
    ned
    0.56
    ed
    0.56
    INGS
    0.55
    ning
    0.54
    imed
    0.54
     citoyen
    0.54
    ading
    0.53
    Activations Density 0.100%

    No Known Activations