© 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. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 2859
    Prev
    Next
    INDEX
    Explanations

    best, earliest, est, most, highest, lowest

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    adamkarvonen/qwen3-32b-saes/saes_Qwen_Qwen3-32B_batch_top_k/resid_post_layer_32
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    享çĶ¨
    -0.09
    åłµ
    -0.09
    æĸ°åŁİ
    -0.09
    å¾Īå°ij
    -0.09
    å¢ĥ
    -0.09
    ä¸ĭåįĬ
    -0.09
    樯
    -0.09
    çī©çIJĨåѦ
    -0.08
    roles
    -0.08
    åľ¨æľªæĿ¥
    -0.08
    POSITIVE LOGITS
     available
    0.12
     currently
    0.10
    currently
    0.10
     disponible
    0.09
    æīĢèĥ½
    0.09
    å°ļ
    0.09
    available
    0.09
    æĥħåĨµä¸ĭ
    0.08
    .available
    0.08
     disponibles
    0.08
    Activations Density 0.073%

    No Known Activations