© 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. 30588
    Prev
    Next
    INDEX
    Explanations

    juggler, magician, pelican, polo

    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¦ģ好好
    -0.10
    ä¸įå¿ĺ
    -0.09
    LESS
    -0.08
    stalk
    -0.08
    _ENCODING
    -0.08
    è¦ģæ±Ĥ
    -0.08
    forget
    -0.08
    abad
    -0.08
    æľīæľº
    -0.08
    ç¼Ĥ
    -0.08
    POSITIVE LOGITS
    ä¹ĭåŁİ
    0.09
    ster
    0.09
    åħķ
    0.09
    åħļå§Ķ书记
    0.09
    InputChange
    0.09
    åĿ¤
    0.09
    sters
    0.08
    avit
    0.08
     poczÄħ
    0.08
     Äijiá»ĥn
    0.08
    Activations Density 0.004%

    No Known Activations