© 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. 63865
    Prev
    Next
    INDEX
    Explanations

    computer programming, file uploads

    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
    ouv
    -0.29
    ç®Ń
    -0.27
    ET
    -0.27
    ãĤ¢ãĥĬ
    -0.26
     notation
    -0.25
    notation
    -0.25
     dirig
    -0.24
    cream
    -0.24
     dedicated
    -0.24
    éĺħ读
    -0.24
    POSITIVE LOGITS
    apan
    0.27
    æĥ³è¿ĩ
    0.27
    bons
    0.26
    oston
    0.26
     neutrality
    0.25
    èĪĴå¿ĥ
    0.25
    umd
    0.24
    צ×Ķ
    0.24
    ŀæĢ§
    0.24
    ç»Ĭ
    0.24
    Activations Density 2.703%

    No Known Activations