© 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. 48672
    Prev
    Next
    INDEX
    Explanations

    sed, vari, administr, fasc, sucess, pun endings

    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
    tics
    -0.10
    ष
    -0.08
    /ion
    -0.08
    AIT
    -0.07
    ##↵↵
    -0.07
     wur
    -0.07
    ieber
    -0.07
     BEST
    -0.07
    ãģªãĤĵãģ¦
    -0.07
     wart
    -0.07
    POSITIVE LOGITS
    åı¯çĪ±çļĦ
    0.11
    æĥħåĨµæĿ¥çľĭ
    0.10
    åľ¨æĪijçľĭæĿ¥
    0.09
    ãĤ
    0.09
    itude
    0.09
    个å¤ļ
    0.09
    ness
    0.09
    è¿ľæĻ¯
    0.08
    马äºij
    0.08
    ita
    0.08
    Activations Density 0.335%

    No Known Activations