© 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. 22940
    Prev
    Next
    INDEX
    Explanations

    pastebin links and python code

    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
    èĮµ
    -0.09
    owi
    -0.09
    uron
    -0.09
    æĿŁ
    -0.08
     thr
    -0.08
    inoa
    -0.08
    éĢĢåĽŀ
    -0.08
    ons
    -0.08
    wipe
    -0.08
    éĿ¢åIJij
    -0.08
    POSITIVE LOGITS
    CONN
    0.10
    åı¤åŁİ
    0.10
     puck
    0.09
    ajor
    0.09
    .Fail
    0.09
    å®ļå¾ĭ
    0.09
    à¸Ħà¸Ń
    0.08
    EMON
    0.08
    arb
    0.08
    ä¿
    0.08
    Activations Density 0.006%

    No Known Activations