© 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. 33737
    Prev
    Next
    INDEX
    Explanations

    said in an interview

    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
     ÑģобÑģÑĤвенно
    -0.09
    ERE
    -0.09
    æ¡Įä¸Ĭ
    -0.09
    羣çļĦ
    -0.09
    æģ°å¥½
    -0.09
    .Sync
    -0.09
    ssue
    -0.09
    æĤ¨å¥½
    -0.08
    é»Ħéĩij
    -0.08
    éļ¨æĻĤ
    -0.08
    POSITIVE LOGITS
     SD
    0.09
    ä¸įäºĨ
    0.08
    asar
    0.08
    ::::::::
    0.08
    SD
    0.08
     rằng
    0.08
     Ø£ÙĨÙĩا
    0.07
     cÄĥ
    0.07
    楹
    0.07
    utting
    0.07
    Activations Density 0.077%

    No Known Activations