© 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. 25-TRANSCODER-HP
    7. 108523
    Prev
    Next
    INDEX
    Explanations

    punctuation marks

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_25.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.25.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    é¼»åŃIJ
    -0.28
     knows
    -0.26
    æĮ¨
    -0.26
    ä¸įä¸Ģå®ļ
    -0.26
    ä¸ĢæŦ
    -0.25
    第ä¸Ģ天
    -0.25
    éĥ½çŁ¥éģĵ
    -0.25
    riage
    -0.24
    orian
    -0.24
     thereby
    -0.24
    POSITIVE LOGITS
    chos
    0.26
     unreal
    0.26
     favour
    0.26
     favoured
    0.26
     favor
    0.26
    unei
    0.25
    opp
    0.24
    çļĦä¸ĢäºĽ
    0.24
    eba
    0.24
    ÙħÙĪضÙĪع
    0.24
    Activations Density 1.008%

    No Known Activations