© 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. 37902
    Prev
    Next
    INDEX
    Explanations

    events and statements

    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
    ir
    -0.08
    æŀ¯
    -0.08
     Sick
    -0.07
    æ¶Īè´¹éľĢæ±Ĥ
    -0.07
     desc
    -0.07
    MMC
    -0.07
     ...
    -0.07
     N
    -0.07
     pil
    -0.07
     pilot
    -0.07
    POSITIVE LOGITS
    大éĿ¢ç§¯
    0.11
    .telegram
    0.11
    æĮł
    0.11
    羯
    0.10
    ainment
    0.10
     Jed
    0.10
    ạc
    0.10
    æĻļé¥Ń
    0.10
    çļĦ家åºŃ
    0.10
    _STAGE
    0.10
    Activations Density 0.276%

    No Known Activations