© 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. 48923
    Prev
    Next
    INDEX
    Explanations

    have to [verb]

    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
    oda
    -0.10
    诱人
    -0.10
    诱
    -0.10
    åŃĹåı·
    -0.09
    æīĵçĮİ
    -0.09
    æĭIJ
    -0.09
    hare
    -0.09
    æĪijå°±
    -0.09
    ä»įåľ¨
    -0.09
    _INTERFACE
    -0.09
    POSITIVE LOGITS
    èĩ¾
    0.09
    å¤±æľĽ
    0.09
    ä»ĺåĩº
    0.08
    idis
    0.08
     spindle
    0.08
    çIJĨæĥ³çļĦ
    0.08
    缣
    0.08
    ç´¢åıĸ
    0.08
    ellen
    0.08
    å®īæİĴ
    0.08
    Activations Density 0.077%

    No Known Activations