© 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. 31475
    Prev
    Next
    INDEX
    Explanations

    select a, an, data, object

    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
    æĥ³æ³ķ
    -0.10
    æĿ¥ä¸įåıĬ
    -0.10
    使åij½æĦŁ
    -0.10
    åįļ士åѦä½į
    -0.09
    åĤ
    -0.09
    ounder
    -0.09
    ató
    -0.09
    unos
    -0.08
    ahl
    -0.08
    æīĵçĿĢ
    -0.08
    POSITIVE LOGITS
    åIJĪéĢĤçļĦ
    0.11
     wisely
    0.10
     Wis
    0.09
    _IMP
    0.09
     appropriate
    0.08
    çļĦéģĵè·¯
    0.08
    FromString
    0.08
    æĿ¥è®²
    0.08
    .Operator
    0.08
     option
    0.08
    Activations Density 0.165%

    No Known Activations