© 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. 18353
    Prev
    Next
    INDEX
    Explanations

    number followed by punctuation or 'input'

    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¸Ģå®ļçļĦ
    -0.09
    .net
    -0.09
    ä¸Ģå®ļç¨ĭ度
    -0.09
    éĺµéĺµ
    -0.09
    ูà¹Ī
    -0.09
    atre
    -0.08
    unas
    -0.08
    ur
    -0.08
    ters
    -0.08
    odore
    -0.08
    POSITIVE LOGITS
    owing
    0.11
    ï¸ı
    0.10
    ãĤ£
    0.10
    бо
    0.09
    æıĸ
    0.09
    æĺ¯éĿŀ
    0.09
    owed
    0.08
    igits
    0.08
     SimpleName
    0.08
    phans
    0.08
    Activations Density 0.967%

    No Known Activations