© 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. 17047
    Prev
    Next
    INDEX
    Explanations

    killed, claiming, or leaving victims

    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
    èĢ½
    -0.10
    纯粹
    -0.09
     ÑģÑĢедне
    -0.09
     Dob
    -0.08
    çĶĺ
    -0.08
    подоб
    -0.08
    _sess
    -0.08
    serter
    -0.08
    Ñģка
    -0.08
    ä»ĸå·²ç»ı
    -0.08
    POSITIVE LOGITS
    ä¸įå°ij人
    0.10
    æľ¬æĬ¥
    0.09
    许å¤ļ人
    0.09
    ä»»ä½ķ人éĥ½
    0.09
    çĻ½éĽª
    0.09
    ä½Ļ人
    0.09
    大åıij
    0.09
    ä½Ļ家
    0.08
    个å¤ļæľĪ
    0.08
    ä¼Ĺ人
    0.08
    Activations Density 0.041%

    No Known Activations