© 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. 2652
    Prev
    Next
    INDEX
    Explanations

    Lo Green, derivative, Macross, Anni

    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
    åĪ»çĶ»
    -0.09
    بÙĥ
    -0.08
    aten
    -0.08
     MyApp
    -0.08
    jac
    -0.08
    wall
    -0.08
    èį£
    -0.08
    èĥĨ
    -0.08
    çľģæĶ¿åºľ
    -0.08
    èĬ½
    -0.07
    POSITIVE LOGITS
    è¿ĽåĨĽ
    0.09
    umes
    0.09
    åŦ
    0.09
    cheid
    0.09
    åĴĮå°ļ
    0.08
    azzo
    0.08
    æĵįä½ľç³»ç»Ł
    0.08
    ä¸Ĭå²Ĺ
    0.08
     Veranst
    0.08
    eways
    0.08
    Activations Density 0.011%

    No Known Activations