© 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. 18994
    Prev
    Next
    INDEX
    Explanations

    Endings

    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
    uld
    -0.27
    ä¸įèĪį
    -0.27
    åł
    -0.27
    æĺİçŁ¥
    -0.26
    au
    -0.26
    inion
    -0.25
    ä¿Ŀ
    -0.25
     Goth
    -0.25
    åIJĦæł·
    -0.25
    æ¨Ł
    -0.25
    POSITIVE LOGITS
    åĢ¡
    0.26
    æĭĹ
    0.26
    .interpolate
    0.26
    éĻĦ
    0.25
     MX
    0.24
     shifts
    0.24
    红
    0.24
    Mexico
    0.24
    çļĦä¼Łå¤§
    0.24
    éļIJ
    0.24
    Activations Density 5.959%

    No Known Activations