© 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. 6505
    Prev
    Next
    INDEX
    Explanations

    selection and choice

    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
    综åIJĪæ²»çIJĨ
    -0.08
    пÑĥÑĤ
    -0.08
    æĬ±
    -0.08
    _identity
    -0.08
     Projects
    -0.07
    LD
    -0.07
    GEST
    -0.07
    ä¸įå°ı
    -0.07
    èĩªæ²»
    -0.07
    èIJ¥éĢł
    -0.07
    POSITIVE LOGITS
    ä¼ĺè¶Ĭ
    0.12
    æĬ¢
    0.11
    æĮij
    0.11
    å°±ä¸ļå²Ĺä½į
    0.11
    éĵĢ
    0.10
    好çļĦ
    0.10
    ä¼ĺèī¯
    0.10
    æį·
    0.10
    .ro
    0.09
    è¾ĥ好çļĦ
    0.09
    Activations Density 0.063%

    No Known Activations