© 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. 33219
    Prev
    Next
    INDEX
    Explanations

    I thought/expect

    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
    åĽĽæĸ¹
    -0.09
    æĩĬ
    -0.08
    伦
    -0.08
    ä¸įåIJĮäºİ
    -0.08
    ä¸įæĢķ
    -0.08
    ä¸įæĪIJ
    -0.08
     Hav
    -0.08
     ÑĤебе
    -0.08
    ä¹į
    -0.07
    è´»
    -0.07
    POSITIVE LOGITS
    è¿Ļä¹Īå¤ļ
    0.11
    yles
    0.10
    .persistent
    0.10
    æĪ´
    0.09
    panse
    0.09
    éĤ£ä¹Īå¤ļ
    0.09
    è°±
    0.09
    лон
    0.09
    ynes
    0.09
    DataExchange
    0.09
    Activations Density 0.015%

    No Known Activations