© 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. 5458
    Prev
    Next
    INDEX
    Explanations

    file paths and library names

    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
     thả
    -0.10
    å¿ĥåĬ¨
    -0.10
    Acts
    -0.10
    _sibling
    -0.09
    iali
    -0.09
    ial
    -0.09
     YM
    -0.09
    lion
    -0.09
    inia
    -0.08
    éĹŃ
    -0.08
    POSITIVE LOGITS
    åħ¨åĽ½äººæ°ij
    0.11
    éĵ¢
    0.10
     pó
    0.09
    oley
    0.09
    çIJĨæĻº
    0.09
    ç½ijåıĭ们
    0.09
    æĿijæ°ij们
    0.09
    à¹Ģà¸Ħล
    0.09
    è±ĨèħIJ
    0.09
    éĢłç¦ı
    0.09
    Activations Density 0.317%

    No Known Activations