© 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. 4340
    Prev
    Next
    INDEX
    Explanations

    "a" followed by location or description

    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
    å¦Ĥæŀľä¸įæĺ¯
    -0.09
    ä¸įåĥı
    -0.09
    ä¸įä¼ļ
    -0.09
    ä¸įåıĹ
    -0.08
    rive
    -0.08
    ä¹İ
    -0.08
    atic
    -0.08
    nee
    -0.08
    nees
    -0.08
    atical
    -0.08
    POSITIVE LOGITS
     indeed
    0.12
    ç¡®
    0.10
    æĮīè§Ħå®ļ
    0.10
    ç¡®å®ŀ
    0.09
    unga
    0.09
    確實
    0.09
    ä¹ŁçŁ¥éģĵ
    0.09
    è¿Ļä¹Īåģļ
    0.09
     Indeed
    0.09
    è´£
    0.09
    Activations Density 0.059%

    No Known Activations