© 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. 12057
    Prev
    Next
    INDEX
    Explanations

    been a [adjective] [time]

    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
    ä¸Ģåı£
    -0.11
    ä¸Ģåıª
    -0.09
    ize
    -0.08
    ä¼ĹæīĢ
    -0.08
    éĤ£æŶ
    -0.08
    agua
    -0.08
    éĤ£æŶåĢĻ
    -0.08
    éĴı
    -0.08
    大å¹ħ度
    -0.08
    ä¸ĢåıĮ
    -0.08
    POSITIVE LOGITS
    нÑĮ
    0.11
    Ñģон
    0.10
    imd
    0.10
     bá»ģ
    0.09
    PFN
    0.09
    chwitz
    0.09
    .son
    0.09
    âĻļ
    0.09
    getDb
    0.09
    æĹħéĢĶ
    0.09
    Activations Density 0.037%

    No Known Activations