© 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. 55772
    Prev
    Next
    INDEX
    Explanations

    response.end

    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
    ç³Ĭ
    -0.09
    raf
    -0.08
    åĩ¹
    -0.08
    è¿Ļ个è¯į
    -0.08
    adar
    -0.08
    åĽŀèIJ½
    -0.08
     referral
    -0.08
    èĸ°
    -0.08
    <?,
    -0.08
    çĸ®
    -0.08
    POSITIVE LOGITS
    æĸĩåĮĸèĬĤ
    0.10
    è°ł
    0.09
    ìĿij
    0.09
    æ¡£
    0.09
    è±ļ
    0.09
    çIJ¬
    0.09
    oire
    0.08
    пиÑĤ
    0.08
     chem
    0.08
    社ä¼ļåĮĸ
    0.08
    Activations Density 0.037%

    No Known Activations