© 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. 4-CLT-HP
    7. 321
    Prev
    Next
    INDEX
    Explanations

    beliefs/understanding

    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
    berdayakan
    -0.64
     åt
    -0.63
     réparer
    -0.62
    μφωνα
    -0.62
     contigo
    -0.60
     pedimos
    -0.60
     feroit
    -0.59
    Спољашње
    -0.59
    henswürdigkeiten
    -0.57
    ьаж
    -0.56
    POSITIVE LOGITS
     that
    0.73
     it
    0.68
     closely
    0.50
     giving
    0.49
     this
    0.47
     getting
    0.45
     successful
    0.45
     doing
    0.45
     people
    0.45
     double
    0.43
    Activations Density 0.002%

    No Known Activations