© 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. 58149
    Prev
    Next
    INDEX
    Explanations

    looks like

    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
     himself
    -0.09
    jack
    -0.08
    NT
    -0.08
     Th
    -0.08
     nor
    -0.08
     NT
    -0.08
    æľīä¿¡å¿ĥ
    -0.08
    ç»·
    -0.08
    .calls
    -0.08
    nt
    -0.08
    POSITIVE LOGITS
    ä¼ij
    0.11
    èĢĮæĿ¥
    0.10
    archs
    0.10
    好åĥı
    0.10
    èĴ¿
    0.10
    ãĤ»ãĥĥãĥĪ
    0.09
    ernal
    0.09
    ENTIC
    0.09
    _MINUS
    0.09
    uset
    0.09
    Activations Density 0.027%

    No Known Activations