© 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. 130311
    Prev
    Next
    INDEX
    Explanations

    Combining multiple things

    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
     Suff
    -0.29
    çļĦå°±æĺ¯
    -0.28
    edi
    -0.27
    è´¢å¯Į
    -0.27
    ä»ĸæĺ¯
    -0.27
    åľ¨ä»ĸçļĦ
    -0.26
    mainwindow
    -0.26
    'i
    -0.25
    èĢħãģ¯
    -0.25
    æľŁåĨħ
    -0.24
    POSITIVE LOGITS
    èļĵ
    0.29
    itura
    0.28
    jak
    0.26
    olean
    0.26
     aValue
    0.26
     some
    0.26
    icus
    0.25
    енко
    0.25
    аÑĤÑĥÑĢ
    0.25
     ÑģÑĤÑĢ
    0.25
    Activations Density 10.314%

    No Known Activations