© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 38334
    Prev
    Next
    INDEX
    Explanations

    file or site is

    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
    è¿Ļ个è¯į
    -0.13
    è¿Ļæł·çļĦ
    -0.12
    è¿Ļç§į
    -0.11
    è¿Ļ段
    -0.11
    éĢĻ種
    -0.11
    è¿Ļ两
    -0.11
    è¿Ļ两个
    -0.11
    è¿ĻåĿĹ
    -0.10
    éĤ£ä¹Īå¤ļ
    -0.10
    éĢĻ樣çļĦ
    -0.10
    POSITIVE LOGITS
    ä¹Łä¸įä¾ĭå¤ĸ
    0.14
    以åıĬåħ¶ä»ĸ
    0.12
    à¹ģหละ
    0.12
     ÙĪÙĩØ°Ùĩ
    0.11
    å°ıå°ıçļĦ
    0.11
    çī¹æ®ĬçļĦ
    0.10
    çªģå¦Ĥåħ¶æĿ¥
    0.10
     particular
    0.10
    ä¸ĸä¸Ĭ
    0.10
    夥
    0.10
    Activations Density 0.478%

    No Known Activations