© 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. 58518
    Prev
    Next
    INDEX
    Explanations

    frames and mats

    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
    RV
    -0.10
    æ³¢çī¹
    -0.09
    èĩªæĦ¿
    -0.09
    æķ£å¸ĥ
    -0.09
    rv
    -0.09
    ÙĨد
    -0.09
    ä¼ı
    -0.09
    åĶ¤
    -0.09
    çİĭåĽ½
    -0.09
     pitched
    -0.09
    POSITIVE LOGITS
     bor
    0.09
     اÙĦتÙĤ
    0.09
     Mars
    0.08
    éªĭ
    0.08
     processing
    0.08
    ä¸Ģç¢Ĺ
    0.08
    åĴĮæĬĢæľ¯
    0.08
    ä¸Ģæł¹
    0.08
    å¢ĥ
    0.08
    mer
    0.08
    Activations Density 0.075%

    No Known Activations