© 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. 26613
    Prev
    Next
    INDEX
    Explanations

    url parameters or path segments

    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
    åİ®
    -0.09
     Mattis
    -0.09
    æı
    -0.09
    doll
    -0.09
    PropertyChanged
    -0.09
    imap
    -0.08
    å®ĺç½ij
    -0.08
    _Util
    -0.08
    ä¸įåIJĪ
    -0.08
    kart
    -0.08
    POSITIVE LOGITS
     pres
    0.09
    æīĭèħķ
    0.09
    åľ¨äºİ
    0.09
    é£İåIJ¹
    0.08
    bn
    0.08
    åĪĿè¡·
    0.08
     hurry
    0.08
    å·±
    0.08
    éĥİ
    0.08
    olie
    0.08
    Activations Density 0.203%

    No Known Activations