© 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. 108782
    Prev
    Next
    INDEX
    Explanations

    Problems and potential solutions

    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
    åıĺåİĭ
    -0.31
    æĥľ
    -0.28
    æĥħ
    -0.26
    @endif
    -0.26
    CHAT
    -0.26
    ëĿ½
    -0.26
    Observers
    -0.25
     Fact
    -0.25
    arkan
    -0.24
    KHR
    -0.24
    POSITIVE LOGITS
    angen
    0.27
     gep
    0.25
     Mozart
    0.25
    ä¸įæ¸ħæ¥ļ
    0.24
    IJľç´¢
    0.24
    _glob
    0.24
    çļĦä»·æł¼
    0.24
    åĹª
    0.24
    åľ¨å¤ĸéĿ¢
    0.24
    éĻIJåĪ¶
    0.23
    Activations Density 67.242%

    No Known Activations