© 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. 141711
    Prev
    Next
    INDEX
    Explanations

    punctuation

    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
     compared
    -0.27
     comparing
    -0.26
    ower
    -0.26
     trails
    -0.26
     fs
    -0.26
     ID
    -0.26
     pit
    -0.26
    åħ¨æĹ¥
    -0.26
     shelter
    -0.25
    çīĪæĿĥå½ĴåİŁ
    -0.25
    POSITIVE LOGITS
    edor
    0.28
    éĽį
    0.28
    rices
    0.28
    structuring
    0.27
    åħļé£İ
    0.26
    ython
    0.26
    */č↵č↵
    0.26
    //č↵č↵
    0.26
    _č↵č↵
    0.24
    æ¦ľ
    0.24
    Activations Density 2.002%

    No Known Activations