© 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. 4105
    Prev
    Next
    INDEX
    Explanations

    foundation or basis of something

    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
    igo
    -0.10
    fight
    -0.10
    æł·çļĦ
    -0.09
    озв
    -0.09
    clair
    -0.09
    è¯Ŀ说
    -0.08
    -être
    -0.08
    weed
    -0.08
    çķĻåѦçĶŁ
    -0.08
    akah
    -0.08
    POSITIVE LOGITS
    neath
    0.12
    lying
    0.12
    设æĸ½å»ºè®¾
    0.10
     underlying
    0.10
    most
    0.09
    ùng
    0.09
    rawer
    0.09
    ymous
    0.09
    mente
    0.09
    ndon
    0.09
    Activations Density 0.014%

    No Known Activations