© 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. 20-TRANSCODER-HP
    7. 129613
    Prev
    Next
    INDEX
    Explanations

    probably

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_20.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.20.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ä¸Ģå¾ĭ
    -0.28
     preservation
    -0.27
     Lik
    -0.25
    Poss
    -0.25
    ç´¢åıĸ
    -0.25
     originals
    -0.25
    ä¸Ģç§į
    -0.25
    æľ¬èº«
    -0.24
    truth
    -0.23
     ÑĢегиÑģÑĤÑĢа
    -0.23
    POSITIVE LOGITS
    è¿ĺæľīå¾Īå¤ļ
    0.28
    chod
    0.26
    æĹ¥æĬ¥è®°èĢħ
    0.25
    åķĨæľº
    0.25
     econ
    0.25
    æ¡·
    0.25
    stre
    0.24
    çĤŃ
    0.24
     soon
    0.24
    ien
    0.24
    Activations Density 12.629%

    No Known Activations