© 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. 53879
    Prev
    Next
    INDEX
    Explanations

    appear on or in

    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
    ä¾Ī
    -0.12
    arro
    -0.11
    æ¯ĵ
    -0.10
    inke
    -0.10
    驽
    -0.09
    ç»ıéªĮåĴĮ
    -0.09
    ir
    -0.09
    éĵĻ
    -0.08
    æ¯ĶæĪij
    -0.08
    gun
    -0.08
    POSITIVE LOGITS
    allet
    0.09
     ÑĥÑĩеб
    0.09
    èĪĴ
    0.09
    rometer
    0.08
    aled
    0.08
    çīµ
    0.08
    ût
    0.08
     Espresso
    0.08
    adies
    0.08
     Mir
    0.08
    Activations Density 0.029%

    No Known Activations