© 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. 18734
    Prev
    Next
    INDEX
    Explanations

    ,

    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
    lint
    -0.32
    _coin
    -0.28
    ä¸Ĭ级
    -0.27
     lig
    -0.26
    液
    -0.26
    .accept
    -0.25
    gli
    -0.25
    æľįåĬ¡æ°´å¹³
    -0.25
    æģķ
    -0.24
    éķ«
    -0.24
    POSITIVE LOGITS
    æķ·
    0.29
    /channel
    0.29
     engagement
    0.27
    enaire
    0.27
     Vec
    0.26
    vecs
    0.26
    滩
    0.26
    ologically
    0.26
     pulses
    0.25
    enticator
    0.25
    Activations Density 2.160%

    No Known Activations