© 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. 62498
    Prev
    Next
    INDEX
    Explanations

    playing a major role

    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
    éĥ½è§īå¾Ĺ
    -0.10
    弧
    -0.09
    (mt
    -0.08
    fect
    -0.08
    felt
    -0.08
    çŃīæĸ¹éĿ¢çļĦ
    -0.08
     subject
    -0.08
    coins
    -0.08
    æµ·æĭĶ
    -0.08
     sink
    -0.07
    POSITIVE LOGITS
    éĩįè¦ģä½ľçĶ¨
    0.20
     role
    0.19
     ÑĢолÑĮ
    0.17
     importance
    0.16
     Role
    0.16
    è§Ĵèī²
    0.16
    åľ°ä½į
    0.15
    ä½ľçĶ¨
    0.15
    æī®æ¼Ķ
    0.15
    çļĦè§Ĵèī²
    0.15
    Activations Density 0.131%

    No Known Activations