© 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. 34563
    Prev
    Next
    INDEX
    Explanations

    the population

    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
    çļĦä¸Ģåijĺ
    -0.10
    æ´¾
    -0.09
    çļĦä¸Ģç§į
    -0.09
     sos
    -0.09
    å¤ļæł·åĮĸ
    -0.09
    åIJĦèĩªçļĦ
    -0.09
    è¿ĩåħ³
    -0.09
     ì¢Ģ
    -0.09
    缸åħ³
    -0.09
     spectator
    -0.09
    POSITIVE LOGITS
    åĬŁå¤«
    0.10
    ç²¾åĬĽ
    0.09
    erta
    0.09
     Rena
    0.09
    å¿ĥæĢĿ
    0.09
    身åŃIJ
    0.09
     blame
    0.09
    çĥŃéĩı
    0.09
    工夫
    0.09
    ertas
    0.09
    Activations Density 0.044%

    No Known Activations