© 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. 4434
    Prev
    Next
    INDEX
    Explanations

    enter your information/email/dates

    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
    夫
    -0.10
    æ¢Ń
    -0.09
    è®·
    -0.08
    èĭĵ
    -0.08
    setDefault
    -0.08
    ç¼¢
    -0.08
    xDA
    -0.08
    CCC
    -0.08
    bjerg
    -0.08
     Antoine
    -0.08
    POSITIVE LOGITS
    æĦŁåħ´è¶£çļĦ
    0.10
    çĿ½
    0.10
     окол
    0.09
    æģķ
    0.09
    éľİ
    0.09
    æĺĶæĹ¥
    0.09
    æļĮ
    0.09
    edis
    0.08
    vil
    0.08
    åĮķ
    0.08
    Activations Density 0.013%

    No Known Activations