© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 20334
    Prev
    Next
    INDEX
    Explanations

    line of

    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
    ats
    -0.08
    åĩºåĬ¨
    -0.08
     slightest
    -0.08
    æ¦ķ
    -0.08
    å¤ļæĸ¹
    -0.08
    è¡Įä¸ļåıijå±ķ
    -0.08
    æij©
    -0.08
    lop
    -0.08
    ä¹IJåĽŃ
    -0.08
     Wash
    -0.08
    POSITIVE LOGITS
    é¾Ł
    0.10
    ç»ĵ
    0.10
    é¾
    0.09
    /slick
    0.09
    GED
    0.09
     ÑģеÑĢи
    0.09
    ãģĭãģĦ
    0.09
    è´¯
    0.09
    CESS
    0.09
    zech
    0.09
    Activations Density 0.023%

    No Known Activations