© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 20-TRANSCODER-HP
    7. 131014
    Prev
    Next
    INDEX
    Explanations

    numerical digits

    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
    Negative Logits
    (MenuItem
    -0.26
     mix
    -0.26
     prematurely
    -0.25
    è¿ĺåĮħæĭ¬
    -0.25
     SEL
    -0.24
    .scheme
    -0.24
    Mix
    -0.24
    STREAM
    -0.24
     Mix
    -0.23
    mach
    -0.23
    POSITIVE LOGITS
    è¦ģæ±Ĥ
    0.28
    ç´¢
    0.28
    人
    0.27
     mel
    0.26
    åİ
    0.25
    æĢģ
    0.25
    agn
    0.25
    æĪĸèĢħ说
    0.25
    /story
    0.25
    HN
    0.25
    Activations Density 2.291%

    No Known Activations