© 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. 108670
    Prev
    Next
    INDEX
    Explanations

    Computer programming/software

    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
     virgin
    -0.30
    仪表
    -0.26
     <!--[
    -0.26
    å´ĩæĭľ
    -0.26
    Designer
    -0.26
    ÅĽcie
    -0.24
    å°±çŁ¥éģĵ
    -0.24
    åŀĵ
    -0.24
     +/-
    -0.24
    DES
    -0.24
    POSITIVE LOGITS
    .train
    0.28
    åľ¨ä¸ŃåĽ½
    0.26
     IH
    0.26
    Wy
    0.25
     scalp
    0.25
    å¾ĹåĪ°æľīæķĪ
    0.25
    è¿Ļ两
    0.25
     train
    0.24
    öne
    0.24
    èĢ³
    0.24
    Activations Density 0.105%

    No Known Activations