© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 21107
    Prev
    Next
    INDEX
    Explanations

    should followed by an action

    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»Ļä½łä»¬
    -0.10
    ä½łä»¥ä¸º
    -0.10
    ORMAL
    -0.09
    人æ°ijæľįåĬ¡
    -0.09
    strup
    -0.09
    è¿¢
    -0.09
    éĩįå·¥
    -0.09
    åıĪ被
    -0.09
    ManagerInterface
    -0.09
    为é¦ĸ
    -0.09
    POSITIVE LOGITS
     dok
    0.08
     Brilliant
    0.08
    æİ¨è¿Ł
    0.08
    acct
    0.08
    带ä¸Ĭ
    0.08
    èĢĥèĻij
    0.08
    æľŁå¾ħ
    0.08
    è§Ħéģ¿
    0.08
    æĹ¥å¸¸
    0.08
    aoke
    0.07
    Activations Density 0.028%

    No Known Activations