© 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. 50280
    Prev
    Next
    INDEX
    Explanations

    dates

    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
    ä¸į说
    -0.10
    ÙĪغ
    -0.09
    ãģķãģĦ
    -0.09
    ónico
    -0.09
    oter
    -0.09
    éªĬ
    -0.09
    ende
    -0.08
    oner
    -0.08
    çŃīæĥħåĨµ
    -0.08
    åįģå¹´
    -0.08
    POSITIVE LOGITS
     behalf
    0.12
     occasion
    0.10
    çļĦä¸ĢäºĽ
    0.10
    lies
    0.09
    åºľ
    0.09
    çļĦæĺ¯
    0.08
     occasions
    0.08
    ãģŁãģı
    0.08
    cem
    0.08
    çļĦè¯Ŀé¢ĺ
    0.08
    Activations Density 0.075%

    No Known Activations