© 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. 17409
    Prev
    Next
    INDEX
    Explanations

    on/off states

    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
    _REFRESH
    -0.09
     ÑģÑĦ
    -0.09
    ieu
    -0.09
    asz
    -0.09
    æ¸İ
    -0.09
    åĪ·æĸ°
    -0.09
    .CustomButton
    -0.09
    panse
    -0.08
    è¯Ģ
    -0.08
    (IDC
    -0.08
    POSITIVE LOGITS
    ãģ«ãģĹãģ¦
    0.10
    ä¸İåIJ¦
    0.10
    ectors
    0.09
    åIJ¯çĶ¨
    0.09
    /off
    0.09
    æŃ¤é¡¹
    0.09
    该项
    0.09
     turned
    0.08
    好åĿı
    0.08
    ulls
    0.08
    Activations Density 0.177%

    No Known Activations