© 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. 106991
    Prev
    Next
    INDEX
    Explanations

    Numerical measurements

    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
    æĢ»éĿ¢ç§¯
    -0.33
    çĤ¹å¤ļ
    -0.32
    sted
    -0.32
    容纳
    -0.32
    建çŃijéĿ¢ç§¯
    -0.31
    åħ¨æł¡
    -0.31
    容éĩı
    -0.31
    éĿ¢ç§¯
    -0.30
    æķ°çĻ¾
    -0.30
    äºĶåįĥ
    -0.30
    POSITIVE LOGITS
    éĹ°
    0.27
    ataka
    0.27
    .isEnabled
    0.25
    骧
    0.25
    å¤±çľł
    0.25
    å¾Ĥ
    0.25
    aturas
    0.24
    è¿ĩé«ĺ
    0.24
    庸
    0.24
    éĹ´éļĻ
    0.23
    Activations Density 25.392%

    No Known Activations