© 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. 34170
    Prev
    Next
    INDEX
    Explanations

    time periods and cycles

    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
    åѤ
    -0.11
     Fut
    -0.10
    urger
    -0.09
    大åݦ
    -0.09
    ersist
    -0.09
    etest
    -0.09
    urge
    -0.09
    æijĨ
    -0.08
    usted
    -0.08
    产éĩı
    -0.08
    POSITIVE LOGITS
    nev
    0.10
     Ub
    0.09
    è¡Ķ
    0.09
    ç´Ĭä¹±
    0.08
    ีà¹ī
    0.08
    à¸Ħำ
    0.08
    åIJįæł¡
    0.08
    顺
    0.08
    lists
    0.08
     Mint
    0.08
    Activations Density 0.038%

    No Known Activations