© 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-3-270M
    5. 15-GEMMASCOPE-2-RES-16K
    7. 6300
    Prev
    Next
    INDEX
    Explanations

    steps to avoid or not

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-270m-pt/resid_post/layer_15_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     ['$
    2.05
    angar
    1.89
    asket
    1.82
    知らせ
    1.81
     nghe
    1.76
     AuthService
    1.74
     Babylonian
    1.73
    حبه
    1.71
     Edmunds
    1.71
    怨
    1.70
    POSITIVE LOGITS
     extrême
    1.93
     höchste
    1.62
     Against
    1.48
     not
    1.48
     Amongst
    1.42
     fuori
    1.42
     edge
    1.41
     never
    1.40
     Not
    1.40
    而出
    1.39
    Activations Density 0.022%

    No Known Activations