© 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. Qwen3-4B
    5. 25-TRANSCODER-HP
    7. 108017
    Prev
    Next
    INDEX
    Explanations

    Chinese characters

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_25.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.25.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    桧
    -0.27
     Feel
    -0.26
    uitar
    -0.26
     Dice
    -0.25
     Their
    -0.25
    ÑģоÑĢ
    -0.24
    pike
    -0.24
    idence
    -0.23
    Their
    -0.23
    _die
    -0.23
    POSITIVE LOGITS
    åħ³éĶ®æĬĢæľ¯
    0.30
    èµİ
    0.28
    ä¿¡è´·
    0.27
    讯
    0.27
    åĬ³åĬ¨åĬĽ
    0.26
    æ¸ħæĻ¨
    0.26
    è°Ĺ
    0.26
    åİŁæĿIJæĸĻ
    0.25
    hes
    0.25
    æĹ©æĻ¨
    0.25
    Activations Density 0.910%

    No Known Activations