© 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. 18949
    Prev
    Next
    INDEX
    Explanations

    people

    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
    of
    -0.42
     of
    -0.34
    Of
    -0.33
    -of
    -0.32
     Of
    -0.32
    .of
    -0.31
    _of
    -0.31
    /of
    -0.29
     OF
    -0.28
    cp
    -0.27
    POSITIVE LOGITS
     plagued
    0.27
    éģĹçķĻ
    0.26
     Coach
    0.25
     prefixed
    0.25
    ãģ¾ãģĹãģ¦
    0.25
    æŁ¥çľĭ详æĥħ
    0.24
     coach
    0.24
    éĺĹ
    0.24
    $class
    0.24
    Coach
    0.24
    Activations Density 0.172%

    No Known Activations