© 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. 18597
    Prev
    Next
    INDEX
    Explanations

    code snippets/reports

    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
    份
    -0.27
     khúc
    -0.27
    cross
    -0.26
    çĸµ
    -0.26
    ç©¿
    -0.26
    å°½
    -0.26
    ogh
    -0.26
     extrem
    -0.26
    交åıī
    -0.25
    交
    -0.25
    POSITIVE LOGITS
    æİ¢
    0.26
    opathy
    0.26
    zik
    0.25
    FieldType
    0.25
    references
    0.24
    ervoir
    0.24
    æ³°åĽ½
    0.24
    idents
    0.23
    omat
    0.23
     luaL
    0.23
    Activations Density 0.185%

    No Known Activations