© 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. 48428
    Prev
    Next
    INDEX
    Explanations

    hook or loop

    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
    бе
    -0.09
    èĦļ
    -0.08
    æŁı
    -0.08
     MU
    -0.08
    _DOM
    -0.08
    ume
    -0.08
    ç¡ķ
    -0.08
    apers
    -0.08
    åıĤ
    -0.08
     ÙĪاÙĦØŃ
    -0.08
    POSITIVE LOGITS
    å¡ĺ
    0.10
    åĮĸè¿Ľç¨ĭ
    0.10
    YTE
    0.10
    æ»ŀ
    0.10
    .signature
    0.09
    çĮİ
    0.09
     pron
    0.09
    SES
    0.09
    uttle
    0.09
    ä¸Ģ个èģĮä¸ļ
    0.09
    Activations Density 0.041%

    No Known Activations