© 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. 5056
    Prev
    Next
    INDEX
    Explanations

    the rest of

    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
    èĮ¹
    -0.09
     اÙĦرØŃ
    -0.09
    ond
    -0.08
    zion
    -0.08
    次
    -0.08
     otherwise
    -0.08
    à¸Ļาม
    -0.08
    é²Ī
    -0.08
    æIJº
    -0.08
    aines
    -0.08
    POSITIVE LOGITS
    calar
    0.11
    oppers
    0.10
    çļĦåĬªåĬĽ
    0.09
    /new
    0.09
    stä
    0.09
    çļĦ人æĿ¥è¯´
    0.09
     Ø£ÙĨØŃاء
    0.09
    ä»»ä½ķæŶåĢĻ
    0.08
    ä»ĸå¦Ī
    0.08
    人éĥ½
    0.08
    Activations Density 0.009%

    No Known Activations