© 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. 56060
    Prev
    Next
    INDEX
    Explanations

    enter the

    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
    itori
    -0.10
    åĩºéŨ
    -0.09
    åıŁ
    -0.09
    gra
    -0.09
    åĽłä¸ºä»ĸ们
    -0.09
    ulta
    -0.09
    åĽłä¸ºæĪij
    -0.09
    åŁŃ
    -0.09
    æīĭå¥Ĺ
    -0.09
    éĴ±å¸ģ
    -0.08
    POSITIVE LOGITS
    å°Ĭ严
    0.10
     Schwartz
    0.09
     blows
    0.08
    iton
    0.08
     Blow
    0.08
    æĢĿ绪
    0.08
    交ä¼ļ
    0.08
     trav
    0.08
    ä½ı
    0.08
    ãģĭãģ¤
    0.08
    Activations Density 0.001%

    No Known Activations