© 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. 61845
    Prev
    Next
    INDEX
    Explanations

    dielectrics

    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
    èĻı
    -0.11
    oes
    -0.09
    ny
    -0.09
    onto
    -0.09
    ä½ij
    -0.09
    æµ·æ¹¾
    -0.08
    åĽ½æľīèµĦ产
    -0.08
    çIJ¥
    -0.08
    usta
    -0.08
    -Ta
    -0.08
    POSITIVE LOGITS
    建åįİ
    0.11
    lectric
    0.10
    bolt
    0.09
    è¿Ļæł·ä¸ĢæĿ¥
    0.08
    羶
    0.08
    ä¼ģ
    0.08
    æ¸Ń
    0.08
    _exchange
    0.08
    UTION
    0.08
    è§īæĤŁ
    0.08
    Activations Density 0.026%

    No Known Activations