© 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. 61662
    Prev
    Next
    INDEX
    Explanations

    attacks against

    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
    eln
    -0.09
    大家ä¸Ģèµ·
    -0.08
    åľºåĿĩ
    -0.08
     opportun
    -0.08
    æ¡Īä¾ĭ
    -0.08
    ç»ĻåĪ«äºº
    -0.08
    ä¸ĵ人
    -0.08
     Frem
    -0.08
    峡
    -0.07
    åĥ®
    -0.07
    POSITIVE LOGITS
    çĤ³
    0.10
    ì²Ń
    0.10
    çļĦ缮æłĩ
    0.10
    .Flag
    0.10
    zell
    0.09
    rail
    0.09
    åŃĹä½ĵ
    0.09
    åģ¥åº·çļĦ
    0.09
    æ¯įæł¡
    0.09
     discrepan
    0.08
    Activations Density 0.031%

    No Known Activations