© 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. 38835
    Prev
    Next
    INDEX
    Explanations

    technical documents

    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
    é«ĺæ°´å¹³
    -0.10
    æĽ´å¤ļ精彩
    -0.10
    æŀķ
    -0.09
    è´Ŀ壳
    -0.09
     gal
    -0.09
    åħĭæľį
    -0.09
    许
    -0.09
    è¿Ļåĩłå¤©
    -0.09
    iful
    -0.09
    èĥ½æ»¡è¶³
    -0.09
    POSITIVE LOGITS
    deen
    0.09
    surf
    0.08
    alse
    0.08
    ç¢Ľ
    0.08
    人身
    0.08
    è¢į
    0.08
    æ®·
    0.08
    æ¯ħåĬĽ
    0.08
    æĬij
    0.08
    /md
    0.08
    Activations Density 0.343%

    No Known Activations