© 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. 20-TRANSCODER-HP
    7. 108364
    Prev
    Next
    INDEX
    Explanations

    to

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_20.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.20.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    å·§
    -0.28
    PLUS
    -0.25
    licit
    -0.25
    acja
    -0.24
    .getOwnPropertyDescriptor
    -0.24
    .baomidou
    -0.24
    LOSE
    -0.24
    amientos
    -0.23
    aments
    -0.23
     szer
    -0.23
    POSITIVE LOGITS
    çı¾åł´
    0.28
     vit
    0.26
    çİ°åľº
    0.25
    伸ãģ³
    0.25
    itu
    0.25
     cru
    0.25
    sit
    0.24
     fly
    0.24
    åĩĨåħ¥
    0.24
    edx
    0.24
    Activations Density 19.644%

    No Known Activations