© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 9955
    Prev
    Next
    INDEX
    Explanations

    support or backup

    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
    年人
    -0.09
    atalog
    -0.09
    è¿ĻæĿ¡
    -0.09
    nip
    -0.09
    çŁ¥éģĵäºĨ
    -0.09
    两æĿ¡
    -0.09
    elpers
    -0.09
    annel
    -0.09
    ä¸įæĽ¾
    -0.08
    apons
    -0.08
    POSITIVE LOGITS
    æĶ¯æĮģ
    0.16
     support
    0.13
     evidence
    0.12
    support
    0.12
     supporting
    0.12
    æĶ¯æĴij
    0.12
    è¯ģæį®
    0.11
    çIJĨçĶ±
    0.11
     Supporting
    0.11
    Support
    0.11
    Activations Density 0.047%

    No Known Activations