© 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. 32818
    Prev
    Next
    INDEX
    Explanations

    security risk

    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
    _VERTEX
    -0.11
    æľīåIJįçļĦ
    -0.10
    олÑı
    -0.10
    èĥİ
    -0.09
    olest
    -0.09
     veto
    -0.09
    ç©¿ä¸Ĭ
    -0.09
    è¿ĺä¸įå¦Ĥ
    -0.09
     Mansion
    -0.09
    足足
    -0.09
    POSITIVE LOGITS
     hack
    0.12
    hack
    0.09
    æĻ¾
    0.08
     Hack
    0.08
    LLU
    0.08
    å²ŃåįĹ
    0.08
    azzi
    0.08
    éĶģå®ļ
    0.08
    kick
    0.08
    å°±æĬĬ
    0.08
    Activations Density 0.045%

    No Known Activations