© 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. 21839
    Prev
    Next
    INDEX
    Explanations

    advice and experience

    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
    æł¸å¿ĥ
    -0.06
    yr
    -0.05
    inois
    -0.05
    i
    -0.05
    纯
    -0.05
    好äºĨ
    -0.05
    ç©´
    -0.05
    ä¸¥æł¼çļĦ
    -0.05
    æł¸å¿ĥçļĦ
    -0.05
    ens
    -0.05
    POSITIVE LOGITS
    äºijéĽĨ
    0.07
    å¾ĢæľŁç²¾å½©
    0.07
    åĴĮä¿¡æģ¯
    0.06
    urally
    0.06
    åĴĮæ°Ķ
    0.06
    uton
    0.06
    åĴĮä¿¡æģ¯åĮĸ
    0.06
    æľ¬èº«å°±
    0.06
    缮ãģ«
    0.06
     endregion
    0.06
    Activations Density 0.003%

    No Known Activations