© 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. 61148
    Prev
    Next
    INDEX
    Explanations

    article metadata

    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
     Invent
    -0.07
     Ingen
    -0.07
    æĹ©æĻ¨
    -0.07
    æĺ§
    -0.07
    ä¸īåįĥ
    -0.07
    ä¹Łå¥½
    -0.06
    ä¹¾åĿ¤
    -0.06
    失åİ»
    -0.06
    eming
    -0.06
    å°¸
    -0.06
    POSITIVE LOGITS
     мил
    0.10
    é¹³
    0.10
    én
    0.10
    .piece
    0.10
    irit
    0.10
    .Selenium
    0.10
    产åĮº
    0.10
    egl
    0.10
    ANDOM
    0.10
    .experimental
    0.09
    Activations Density 3.742%

    No Known Activations