© 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. 39658
    Prev
    Next
    INDEX
    Explanations

    often, usually, sometimes

    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
    ynom
    -0.10
    羣æŃ£åģļåĪ°
    -0.09
    ä½ĵä¼ļåĪ°
    -0.09
    è¿Ļ件
    -0.09
    éĢŀ
    -0.09
    /vendor
    -0.09
    åĬŁå¾·
    -0.08
    éķĽ
    -0.08
     Bund
    -0.08
    tat
    -0.08
    POSITIVE LOGITS
    éĴĪ
    0.11
     xuyên
    0.10
    éĩĿ
    0.10
    aspers
    0.09
    è¯ķéªĮåĮº
    0.09
    /span
    0.09
    深度
    0.09
    æīĩ
    0.09
    éĩijéĵ¶
    0.09
    Rare
    0.08
    Activations Density 0.188%

    No Known Activations