© 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. 58520
    Prev
    Next
    INDEX
    Explanations

    start application

    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
    è¿Ļä¸ĢåĪ»
    -0.10
    icles
    -0.10
    icle
    -0.10
    baum
    -0.09
    åįĥåı¤
    -0.09
    ool
    -0.09
    æ´ģåĩĢ
    -0.09
    头åıij
    -0.09
    被æī§è¡Į
    -0.09
    éĩįè¿Ķ
    -0.08
    POSITIVE LOGITS
    rita
    0.10
     sesión
    0.09
     Kidd
    0.08
    çĨı
    0.08
    漫
    0.08
    ä»»ä½ķä¸Ģ个
    0.08
     Pill
    0.08
    渡
    0.08
    à¸Ĭà¸Ļ
    0.08
    _BIND
    0.08
    Activations Density 0.089%

    No Known Activations