© 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. 5-CLT-HP
    7. 4771
    Prev
    Next
    INDEX
    Explanations

    expressing preferences

    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
     prefi
    -0.88
     prefer
    -0.88
     prefers
    -0.88
     prefier
    -0.83
     rather
    -0.78
     lieber
    -0.77
     Rather
    -0.73
     plutôt
    -0.73
     preferring
    -0.73
     preferred
    -0.72
    POSITIVE LOGITS
     nakalista
    0.84
     ſever
    0.69
    ftagPool
    0.68
     légitime
    0.66
    Izvori
    0.65
     purpoſe
    0.65
     ſeveral
    0.64
    Atsauces
    0.64
    atrième
    0.62
     pleaſure
    0.61
    Activations Density 0.017%

    No Known Activations