© 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. 38258
    Prev
    Next
    INDEX
    Explanations

    more than numbers

    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
    å¾Ĺ好
    -0.10
    ä½ľèĢħæľ¬äºº
    -0.10
    odor
    -0.10
    ãģİ
    -0.10
     Rou
    -0.09
    æ¼³
    -0.09
    纶
    -0.09
    uly
    -0.09
    ë¡ł
    -0.09
     dobr
    -0.09
    POSITIVE LOGITS
    æĹ°
    0.09
    esting
    0.09
    ä¸Ģåı£æ°Ķ
    0.08
    深度
    0.08
    _HEADERS
    0.08
    ми
    0.08
    审
    0.08
    é¢Ģ
    0.08
     Spit
    0.08
    亶
    0.08
    Activations Density 0.063%

    No Known Activations