© 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. 41869
    Prev
    Next
    INDEX
    Explanations

    UK cities/towns

    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
    aze
    -0.29
     based
    -0.28
     inc
    -0.28
     Fl
    -0.28
     uncon
    -0.27
     horizon
    -0.27
    based
    -0.27
    _based
    -0.26
     fl
    -0.26
    ç»Ŀ对
    -0.26
    POSITIVE LOGITS
    born
    0.27
    Hardware
    0.26
    &type
    0.26
     Granite
    0.26
    æijĺ
    0.26
     Morg
    0.26
    _behavior
    0.26
    è¾ĵåĩº
    0.26
    olics
    0.25
    æĸ°åŁºå»º
    0.25
    Activations Density 0.020%

    No Known Activations