© 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. 57495
    Prev
    Next
    INDEX
    Explanations

    not yet completed

    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
    ÙħÙĨت
    -0.09
    ä¸įæŃ»
    -0.09
    urat
    -0.09
    око
    -0.09
     обоÑĢ
    -0.09
    جب
    -0.08
    è·º
    -0.08
    igar
    -0.08
    iolet
    -0.08
    .enc
    -0.08
    POSITIVE LOGITS
    çĶ¨æĪ·æıIJä¾Ľ
    0.10
    ufs
    0.10
    è¯ļ
    0.09
    ä¸ĢåĪĩéĥ½
    0.09
    èĮ§
    0.09
    ummy
    0.09
    ä¼ģä¸ļæıIJä¾Ľ
    0.09
    umm
    0.09
    ly
    0.09
    éķĽ
    0.09
    Activations Density 0.017%

    No Known Activations