© 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. Qwen3.5-9B Base
    5. 15-QWENSCOPE-RES-64K
    7. 13752
    Prev
    Next
    INDEX
    Explanations

    positive or negative directions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    lander
    -0.08
     Hor
    -0.08
    å«Ĥ
    -0.08
    uÅŁtur
    -0.07
     supp
    -0.07
    лад
    -0.07
     fri
    -0.07
    ssel
    -0.07
    /json
    -0.07
    ott
    -0.07
    POSITIVE LOGITS
    表示
    0.09
     indicating
    0.09
     Indicates
    0.09
    signed
    0.08
     indicates
    0.08
    indic
    0.08
    æĦıåij³
    0.08
    egative
    0.08
    itta
    0.08
     Attached
    0.08
    Activations Density 0.004%

    No Known Activations