© 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. 32-TRANSCODER-HP
    7. 42943
    Prev
    Next
    INDEX
    Explanations

    Special characters and foreign language

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_32.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.32.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    (control
    -0.29
    æݧåĪ¶
    -0.29
    (Control
    -0.28
    %!
    -0.28
     controlling
    -0.28
     ApplicationController
    -0.27
    orElse
    -0.27
    (ctrl
    -0.27
     control
    -0.27
    /control
    -0.27
    POSITIVE LOGITS
    nero
    0.28
     Manual
    0.27
    etat
    0.26
     Die
    0.26
    ä¼ģä¸ļæıIJä¾Ľ
    0.25
     Sur
    0.25
     Lig
    0.25
    WiÄĻ
    0.25
    rena
    0.24
     Der
    0.24
    Activations Density 0.001%

    No Known Activations