© 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. 38755
    Prev
    Next
    INDEX
    Explanations

    gain knowledge or insight

    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
    fork
    -0.09
    æ¶Ī
    -0.09
    inal
    -0.08
    plug
    -0.08
     Marino
    -0.08
    elon
    -0.08
    choose
    -0.08
    æıĴ
    -0.08
     Tun
    -0.08
    è°ĥåĬ¨
    -0.08
    POSITIVE LOGITS
    究竣
    0.14
    究竣æĺ¯
    0.13
    究
    0.13
    æĥ³çŁ¥éģĵ
    0.11
    æĺ¯åIJ¦åŃĺåľ¨
    0.11
    rese
    0.10
    æĹ©çĤ¹
    0.10
    åĪ°åºķ
    0.10
    ä¸įæĺİçĻ½
    0.10
    ç¡®åĪĩ
    0.10
    Activations Density 0.246%

    No Known Activations