© 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. 31850
    Prev
    Next
    INDEX
    Explanations

    previously or formerly

    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
     ÑĢеÑĪа
    -0.09
    pard
    -0.09
    ä¾Ŀæŧ
    -0.09
    ForRow
    -0.09
    acen
    -0.09
     Bris
    -0.09
    å®ĺæĸ¹ç½ijç«Ļ
    -0.09
    æ§Ľ
    -0.08
    贯穿
    -0.08
    ä¸ĭåij¨
    -0.08
    POSITIVE LOGITS
     previously
    0.14
    以åīį
    0.12
    ä¸Ģ度
    0.12
    严éĩįå½±åĵį
    0.11
     previous
    0.11
     Previously
    0.11
     formerly
    0.10
    Previously
    0.10
     ÑĢанее
    0.10
    æĽ¾ç»ı
    0.09
    Activations Density 0.339%

    No Known Activations