© 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. 33-TRANSCODER-HP
    7. 53372
    Prev
    Next
    INDEX
    Explanations

    waste

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_33.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.33.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ogui
    -0.26
    缴åįĩ
    -0.26
     Mansion
    -0.26
    LBL
    -0.26
    apanese
    -0.26
     karÅŁÄ±s
    -0.25
     trilogy
    -0.25
    itioner
    -0.25
     cá»Ń
    -0.25
    .instant
    -0.25
    POSITIVE LOGITS
    -fed
    0.29
    ug
    0.29
    pages
    0.29
    æĬĢæľ¯æĶ¯æĮģ
    0.28
    utz
    0.28
    éĿ¢
    0.28
     feed
    0.27
    berries
    0.26
    -rad
    0.26
     tut
    0.26
    Activations Density 0.013%

    No Known Activations