© 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. 48611
    Prev
    Next
    INDEX
    Explanations

    correspondence

    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
    astic
    -0.10
    饮æĸĻ
    -0.10
    èħ°
    -0.09
    ãģ³
    -0.09
    bear
    -0.09
    çݰ身
    -0.09
    çĶ·å¥³
    -0.09
    ptr
    -0.09
    IMUM
    -0.08
    омеÑĢ
    -0.08
    POSITIVE LOGITS
    _pdu
    0.09
    å¢ħ
    0.09
     correspondence
    0.09
    æľ¬ç½ijç«Ļ
    0.08
     localization
    0.08
    _NT
    0.08
     Localization
    0.08
    åŁı
    0.08
     localize
    0.08
     Correspond
    0.08
    Activations Density 0.013%

    No Known Activations