© 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. 51773
    Prev
    Next
    INDEX
    Explanations

    errors and accuracy

    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
    åĨ¤
    -0.10
    åĽŀæĿ¥äºĨ
    -0.09
    æ¯ı天éĥ½
    -0.09
    åķĨä¸ļåĮĸ
    -0.09
    çªį
    -0.08
    æ´ij
    -0.08
    æݧ
    -0.08
    é¦ı
    -0.08
    æ¨¡æł·
    -0.08
    çĮ¥
    -0.08
    POSITIVE LOGITS
     pragma
    0.09
    æģķ
    0.09
    è¯ļæĦı
    0.09
    线ä¸Ĭ线ä¸ĭ
    0.09
     Accuracy
    0.09
    æ½®
    0.09
     Lap
    0.09
    idding
    0.09
    -orders
    0.09
    GRE
    0.08
    Activations Density 0.024%

    No Known Activations