© 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. 31-TRANSCODER-HP
    7. 19314
    Prev
    Next
    INDEX
    Explanations

    located

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_31.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.31.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    è¢ĭ
    -0.28
    è¿Ľåĩº
    -0.27
     Fet
    -0.26
    æıIJä¾ĽæĽ´å¤ļ
    -0.26
     Fo
    -0.25
    oha
    -0.25
    èµ·æĿ¥äºĨ
    -0.25
    å¸ĥ
    -0.25
    æŃ¤äºº
    -0.25
    _AUT
    -0.24
    POSITIVE LOGITS
     about
    0.55
    about
    0.49
    About
    0.44
     About
    0.42
     on
    0.42
     several
    0.40
     along
    0.40
     sobre
    0.40
     ABOUT
    0.39
    _about
    0.38
    Activations Density 1.359%

    No Known Activations