© 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. 21736
    Prev
    Next
    INDEX
    Explanations

    define or defining

    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
    ç¬ı
    -0.11
    大åİħ
    -0.10
    umn
    -0.10
    åİħ
    -0.09
    åIJĮæľŁ
    -0.09
    .isBlank
    -0.09
    åħ¼
    -0.09
    çļĦæĥ³æ³ķ
    -0.09
    .effects
    -0.09
    adge
    -0.09
    POSITIVE LOGITS
    ä»Ģä¹Īåı«
    0.10
     boundaries
    0.09
    å¦Ĥä½ķåİ»
    0.09
     contours
    0.09
    è½®å»ĵ
    0.09
     defines
    0.08
    æ¸ħæ¥ļ
    0.08
     define
    0.08
    æĻ°
    0.08
     redefine
    0.08
    Activations Density 0.156%

    No Known Activations