© 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. 17668
    Prev
    Next
    INDEX
    Explanations

    being / that / is / was

    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
    ä¸Ńå¿ĥåŁİåĮº
    -0.09
    æ§İ
    -0.09
    æĺİæĹ¥
    -0.09
    ä¸ij
    -0.09
    åī©ä¸ĭ
    -0.08
    enis
    -0.08
    anes
    -0.08
    بت
    -0.08
    ayas
    -0.08
    ä¸Ģå®ļèĥ½
    -0.08
    POSITIVE LOGITS
    æŃ£åľ¨
    0.11
    ä¸įåıĹ
    0.10
     Äijang
    0.09
    é£ĵ
    0.09
    clide
    0.09
    æ·¡åĮĸ
    0.08
    æ¸ħéĨĴ
    0.08
    доб
    0.08
     being
    0.08
     pl
    0.08
    Activations Density 0.205%

    No Known Activations