© 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. 23927
    Prev
    Next
    INDEX
    Explanations

    nominees or award candidates

    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¿Ļä½į
    -0.10
    æĸ°å¼Ģ
    -0.09
    æ¯ıä½į
    -0.09
    对人ä½ĵ
    -0.09
    (permission
    -0.09
     perm
    -0.08
    æĭĽæĶ¶
    -0.08
    èıı
    -0.08
    å¼Ģä¸ļ
    -0.08
     Franken
    -0.08
    POSITIVE LOGITS
    æıIJåIJį
    0.11
    çĶ³æĬ¥
    0.11
    åIJįé¢Ŀ
    0.11
    åħ¥åĽ´
    0.11
    èĢĥèĻij
    0.10
    å§Ķå±Ī
    0.10
     candidacy
    0.10
    符åIJĪæĿ¡ä»¶
    0.09
    ä¸įåħ¬å¹³
    0.09
     eligibility
    0.09
    Activations Density 0.060%

    No Known Activations