© 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. 57902
    Prev
    Next
    INDEX
    Explanations

    code structure delimiters (-|_)

    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
    ei
    -0.10
    纵
    -0.09
    èĥ½çľĭåĪ°
    -0.09
    峪
    -0.09
    echa
    -0.09
    è¯įæĿ¡
    -0.09
    rawler
    -0.09
    èĵĸ
    -0.09
    Responder
    -0.09
    çļĦåĪ°æĿ¥
    -0.08
    POSITIVE LOGITS
    0.10
    mar
    0.09
    onia
    0.09
    ATAB
    0.09
    olin
    0.08
    åľ¨å®¶
    0.08
    éĴĪ
    0.08
    ahan
    0.08
    olio
    0.08
    ẽ
    0.08
    Activations Density 0.092%

    No Known Activations