© 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. 30-TRANSCODER-HP
    7. 96577
    Prev
    Next
    INDEX
    Explanations

    marriage

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_30.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.30.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     whether
    -0.46
    æĺ¯åIJ¦
    -0.42
    whether
    -0.42
    èĥ½åIJ¦
    -0.40
    æĺ¯åIJ¦åŃĺåľ¨
    -0.39
     Whether
    -0.38
     æĺ¯åIJ¦
    -0.36
    Whether
    -0.35
    èĥ½ä¸įèĥ½
    -0.33
    æĺ¯ä¸įæĺ¯
    -0.32
    POSITIVE LOGITS
     viá»ĩn
    0.28
    æķijæµİ
    0.27
    lingen
    0.27
    æľĢé«ĺçļĦ
    0.27
    vh
    0.25
    综åIJĪç´łè´¨
    0.25
    åݦ
    0.25
     WITHOUT
    0.24
    ä¼ģä¸ļæıIJä¾Ľ
    0.24
    eting
    0.24
    Activations Density 0.892%

    No Known Activations