© 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. 53842
    Prev
    Next
    INDEX
    Explanations

    fair use copyright

    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
    çħŀ
    -0.09
     prag
    -0.08
    _logic
    -0.08
    ou
    -0.08
    Bush
    -0.08
    ãģļ
    -0.08
    ç°§
    -0.07
    cr
    -0.07
     Pars
    -0.07
    .crm
    -0.07
    POSITIVE LOGITS
    æľ¬é¢Ĩ
    0.10
    äºĮåŃĹ
    0.09
    æ¶Ī
    0.09
    å·´åŁº
    0.09
    à¸ĺรรม
    0.09
    æĬ»
    0.09
    åĸĩåıŃ
    0.09
    KIT
    0.09
    人家
    0.09
     Pey
    0.09
    Activations Density 0.008%

    No Known Activations