© 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. 54263
    Prev
    Next
    INDEX
    Explanations

    Non-English words

    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
    uml
    -0.29
    uper
    -0.29
    roz
    -0.29
    ums
    -0.28
     mac
    -0.26
    uko
    -0.26
    arme
    -0.25
    mith
    -0.25
    <src
    -0.25
    åĢºæĿĥ
    -0.25
    POSITIVE LOGITS
    æĹłç¼Ŀ
    0.28
    ноÑģÑĤ
    0.27
     Yong
    0.25
    éĦł
    0.24
    åħ¬ç«ĭåĮ»éĻ¢
    0.24
    quisites
    0.24
    æĥĩ
    0.24
    ATER
    0.24
    ater
    0.24
    èģĶè°Ĭ
    0.24
    Activations Density 0.010%

    No Known Activations