© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 26392
    Prev
    Next
    INDEX
    Explanations

    following a path or lead

    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¸Ģç¢Ĺ
    -0.10
    ally
    -0.09
    oster
    -0.09
    erate
    -0.09
    è¯
    -0.08
    ä¸Ńæĸ¹
    -0.08
    IDA
    -0.08
    _ste
    -0.08
     UPC
    -0.08
    ween
    -0.08
    POSITIVE LOGITS
     dõi
    0.13
     precedent
    0.11
    æĥ¯ä¾ĭ
    0.11
     Äijuá»ķi
    0.11
    踪
    0.10
     Tradition
    0.10
    足迹
    0.09
    çĹķ迹
    0.09
    queda
    0.09
    å°ıç¼ĸä¸Ģèµ·
    0.09
    Activations Density 0.031%

    No Known Activations