© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 52902
    Prev
    Next
    INDEX
    Explanations

    vulnerability, weak, liability

    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
    串
    -0.09
     Tac
    -0.09
    èĩªçĦ¶èĢĮ
    -0.09
    ç§ĺ
    -0.08
    ç¼¢
    -0.08
    ç¥ķ
    -0.08
     Worlds
    -0.08
    æĢĿç»´
    -0.08
    åĽŀéģ¿
    -0.07
    éĺ»åĬĽ
    -0.07
    POSITIVE LOGITS
     vulnerability
    0.20
     Vulner
    0.20
     vulnerable
    0.18
     vulner
    0.17
     vulnerabilities
    0.17
    èĦĨå¼±
    0.16
    ulnerable
    0.14
    å¼±åĬ¿
    0.13
    å¼±çĤ¹
    0.13
    ulner
    0.13
    Activations Density 0.051%

    No Known Activations