© 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. 24-TRANSCODER-HP
    7. 150497
    Prev
    Next
    INDEX
    Explanations

    of

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_24.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.24.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    .locale
    -0.28
     spur
    -0.26
    (___
    -0.26
    ä¾ĭ
    -0.26
    ByExample
    -0.26
    æľĢåIJİä¸Ģ个
    -0.25
    ___
    -0.25
    æł¸å¿ĥæĬĢæľ¯
    -0.25
     Spend
    -0.25
    %%%
    -0.25
    POSITIVE LOGITS
    either
    0.40
     either
    0.40
    Either
    0.34
    æĪĸèĢħ
    0.32
     Either
    0.32
    æĪĸ
    0.32
    heads
    0.32
     или
    0.32
    hosts
    0.29
     æĪĸ
    0.29
    Activations Density 0.054%

    No Known Activations