© 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. 35179
    Prev
    Next
    INDEX
    Explanations

    adsorption and binding

    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
    è·¨
    -0.09
     Ngh
    -0.09
     patched
    -0.09
    ReadStream
    -0.08
    åĮĸ
    -0.08
    æİĴåºı
    -0.08
    ä¼Ļ
    -0.08
    Thunk
    -0.08
    æİ¥ç§į
    -0.08
    çģĮ
    -0.08
    POSITIVE LOGITS
    isoft
    0.12
    åīĤ
    0.11
    åIJ¸éĻĦ
    0.11
     binding
    0.09
    çķĻä½ı
    0.09
    /bind
    0.09
     Capacity
    0.09
    èĥ½åĬĽå¼º
    0.09
    满äºĨ
    0.09
    çī¢åĽº
    0.09
    Activations Density 0.535%

    No Known Activations