© 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. 57169
    Prev
    Next
    INDEX
    Explanations

    flash, flashback, flashcard, flash bomb, flash.net

    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
    è§ĦåĪĴ建设
    -0.11
     Winchester
    -0.09
    standing
    -0.09
    å½Ģ
    -0.09
    psilon
    -0.09
    UCH
    -0.08
    ãĥªãĤ¢
    -0.08
     Äijá»±ng
    -0.08
    omi
    -0.08
    aceous
    -0.08
    POSITIVE LOGITS
    iness
    0.10
    ness
    0.10
    backs
    0.10
    ../../../
    0.10
    ëģĶ
    0.09
    engers
    0.09
    point
    0.09
    ÑĩаÑĤ
    0.09
    frog
    0.09
    idious
    0.09
    Activations Density 0.013%

    No Known Activations