© 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. 1090
    Prev
    Next
    INDEX
    Explanations

    Wisconsin

    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
    ives
    -0.11
    åĽ½åĨħå¸Ĥåľº
    -0.09
     Schwarz
    -0.09
    ictions
    -0.09
    åĨĽäºĭ
    -0.09
    ì²ĺ
    -0.09
    acock
    -0.08
    è¦ģèĢĥèĻij
    -0.08
    ç»İ
    -0.08
    stone
    -0.08
    POSITIVE LOGITS
     Rapids
    0.11
    ResponseType
    0.10
     пÑĢид
    0.10
    ably
    0.09
    bell
    0.09
    ĭ
    0.09
    olle
    0.09
    åīįèĢħ
    0.09
    entai
    0.08
    Į
    0.08
    Activations Density 0.008%

    No Known Activations