© 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. 55455
    Prev
    Next
    INDEX
    Explanations

    various things offered

    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
    æĹģ
    -0.12
    åIJĦè¡Į
    -0.11
    åħ¸åŀĭæ¡Īä¾ĭ
    -0.11
    éģ½
    -0.10
    ä¸ĭ游
    -0.10
    æħµ
    -0.10
    æĪijçľģ
    -0.10
    è¡Ķ
    -0.10
    æĥħå½¢
    -0.10
    æĢĿ念
    -0.10
    POSITIVE LOGITS
     including
    0.08
     a
    0.07
    ï¼ļ
    0.07
    :
    0.07
     an
    0.07
    mob
    0.07
    .tpl
    0.07
    çļĦä¸Ģåijĺ
    0.07
    å¸Ī
    0.07
    mpp
    0.07
    Activations Density 0.060%

    No Known Activations