© 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. 21557
    Prev
    Next
    INDEX
    Explanations

    try code blocks

    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
    á»ĵn
    -0.07
     HOWEVER
    -0.06
    ез
    -0.06
    ope
    -0.06
    hur
    -0.06
    hut
    -0.06
    /free
    -0.06
     Kooper
    -0.06
    å®ĪåĢĻ
    -0.06
    ÑĥÑħ
    -0.06
    POSITIVE LOGITS
    اÙĨات
    0.05
    inea
    0.05
    æ¯ıå½ĵ
    0.05
    äººåľ¨
    0.05
    Interface
    0.05
    ing
    0.05
    ½
    0.05
     пÑĢиÑģÑĤ
    0.05
    Y
    0.05
    Never
    0.05
    Activations Density 0.004%

    No Known Activations