© 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. Qwen3.5-4B
    5. 15-RES-MATRYOSHKA-65K
    7. 21628
    Prev
    Next
    INDEX
    Explanations

    logistic regression

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    decoderesearch/qwen-3.5-saes/qwen-3.5-4b
    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
    éĴ
    -0.05
    erson
    -0.05
    奸
    -0.05
    éĥ¨åĪĨ
    -0.05
     scept
    -0.05
    jung
    -0.05
    serial
    -0.05
     détente
    -0.05
    èĮĥ
    -0.05
    oil
    -0.05
    POSITIVE LOGITS
    arei
    0.06
     наÑģÑĭ
    0.06
    udir
    0.06
    ampp
    0.06
    ×¥
    0.06
     Qual
    0.05
    aways
    0.05
    leans
    0.05
     XSS
    0.05
    é«ĺä¸Ģ
    0.05
    Activations Density 0.009%

    No Known Activations