© 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. 29733
    Prev
    Next
    INDEX
    Explanations

    template files and usage

    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
    æĹ¶æľº
    -0.10
    wig
    -0.09
    aco
    -0.09
    èѬ
    -0.09
    Ñĥк
    -0.09
    Æ°ng
    -0.09
    uate
    -0.09
    uator
    -0.09
    .pic
    -0.08
    fic
    -0.08
    POSITIVE LOGITS
    模æĿ¿
    0.17
    (template
    0.16
     template
    0.15
     templates
    0.14
    /template
    0.13
     TEMPLATE
    0.13
    template
    0.12
    .Template
    0.11
     Template
    0.11
    templates
    0.11
    Activations Density 0.082%

    No Known Activations