© 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. 95861
    Prev
    Next
    INDEX
    Explanations

    blue

    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
    饱
    -0.30
     trÃł
    -0.26
    èĤ¥
    -0.25
    绣çŃ¹æİ¨è¿Ľ
    -0.25
    SKTOP
    -0.24
    že
    -0.24
    ä¸įå¼Ģ
    -0.24
    lds
    -0.24
    WARE
    -0.24
     Bek
    -0.24
    POSITIVE LOGITS
    wil
    0.26
    AIL
    0.26
     prere
    0.26
    æĭ¨
    0.26
    éĽ¢
    0.25
    cock
    0.25
    鬼
    0.25
    è¹ĭ
    0.25
     crossing
    0.25
    åıijå±ķæĸ¹åIJij
    0.25
    Activations Density 0.883%

    No Known Activations