© 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. 31661
    Prev
    Next
    INDEX
    Explanations

    actions after "before"

    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
    欢è¿İ大家
    -0.09
    cheng
    -0.09
     давно
    -0.08
    ç®Ĺæĺ¯
    -0.08
    HAVE
    -0.08
    åıĸè¯ģ
    -0.08
    illon
    -0.08
    ayo
    -0.08
    quo
    -0.08
    inging
    -0.08
    POSITIVE LOGITS
    å¢ĥ
    0.10
    èij³
    0.09
    注æĺİæĿ¥æºIJ
    0.09
    æ³³
    0.08
    plat
    0.08
    濯
    0.08
    éĻħ
    0.08
    åıĤ
    0.08
    ophage
    0.08
    å°¢
    0.08
    Activations Density 0.013%

    No Known Activations