© 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. 4982
    Prev
    Next
    INDEX
    Explanations

    impossibility, completeness, leverage, tipping, cyclic, opportunity

    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
    quirer
    -0.10
    ä¹Łä¸įçŁ¥éģĵ
    -0.10
    åŃĹåı·
    -0.10
    ften
    -0.09
    æ³ķ人
    -0.09
    ¬¸
    -0.09
    éĥ½è§īå¾Ĺ
    -0.09
    essel
    -0.09
    åħ¨æĹ¥åĪ¶
    -0.09
    éĤ£æŶåĢĻ
    -0.09
    POSITIVE LOGITS
    ly
    0.16
    oriously
    0.10
    adays
    0.10
    viously
    0.09
     “
    0.08
    们
    0.08
    ful
    0.08
    &nbsp
    0.08
    ematics
    0.08
    los
    0.08
    Activations Density 0.253%

    No Known Activations