© 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. 39478
    Prev
    Next
    INDEX
    Explanations

    luck

    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
    ä¾ĿéĿł
    -0.10
    mes
    -0.09
    lds
    -0.09
    hip
    -0.09
    è°İ
    -0.08
    .act
    -0.08
     AppModule
    -0.08
    çī¢åĽº
    -0.08
    -addons
    -0.08
    egen
    -0.08
    POSITIVE LOGITS
    fully
    0.14
    æĶ¯éħį
    0.10
    ily
    0.10
    鬼
    0.09
    åįģ足
    0.09
    anden
    0.09
    alse
    0.09
    0.09
    iest
    0.09
    ately
    0.09
    Activations Density 0.020%

    No Known Activations