© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 4-CLT-HP
    7. 1964
    Prev
    Next
    INDEX
    Explanations

    harsh conditions

    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
     harsh
    -1.05
    Harsh
    -1.05
     Harsh
    -0.85
     hars
    -0.81
     brutal
    -0.67
     harshly
    -0.64
     tough
    -0.60
     duros
    -0.59
     жест
    -0.54
    HARD
    -0.54
    POSITIVE LOGITS
     propOrder
    0.75
    Viitteet
    0.68
    Lähteet
    0.63
     ligiloj
    0.59
    KommentareTeilen
    0.59
    ActionCreators
    0.59
    허
    0.57
     rellen
    0.55
     bağlantılar
    0.54
    hoeddwyd
    0.54
    Activations Density 0.002%

    No Known Activations