© 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. 108712
    Prev
    Next
    INDEX
    Explanations

    Firsts or onlys

    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
     itemprop
    -0.26
     cref
    -0.26
    ä¼ł
    -0.25
    ValuePair
    -0.25
    ãĤĢ
    -0.25
    mond
    -0.24
    æĸ¹éĿ¢çļĦ
    -0.24
    ãĤ¢ãĤ¤ãĥĨãĥł
    -0.24
    åĤ³
    -0.24
     cabel
    -0.24
    POSITIVE LOGITS
    enos
    0.29
    éĢĨ转
    0.28
    appen
    0.27
     suspend
    0.26
    碳
    0.26
    ownt
    0.26
    æĸĭ
    0.25
    æĹ©æĻļ
    0.25
    ãĢģ“
    0.25
    纵
    0.25
    Activations Density 14.675%

    No Known Activations