© 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. 53147
    Prev
    Next
    INDEX
    Explanations

    safety and security

    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
    oter
    -0.09
    à´Ļ
    -0.09
    称ä¹ĭ为
    -0.09
    ooter
    -0.09
    pedia
    -0.09
    operand
    -0.08
    çĥĻ
    -0.08
    ç®ķ
    -0.08
     Emer
    -0.08
    èĩªåıij
    -0.08
    POSITIVE LOGITS
     playbook
    0.09
    大å±Ģ
    0.09
    æĦģ
    0.09
     wel
    0.09
    itol
    0.09
    ç¿ĺ
    0.09
    人æ°ijæľįåĬ¡
    0.09
    åĨ³æĪĺ
    0.09
    åĴĮåıijå±ķ
    0.09
    lingen
    0.09
    Activations Density 0.043%

    No Known Activations