© 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. 48577
    Prev
    Next
    INDEX
    Explanations

    not only

    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
    纤
    -0.09
    æīģ
    -0.09
     Lub
    -0.09
    èĢĮä¸Ķè¿ĺ
    -0.08
    pell
    -0.08
    olate
    -0.08
    assoc
    -0.08
    çѱ
    -0.08
     Associ
    -0.08
    为ä»Ģä¹Īä¸į
    -0.08
    POSITIVE LOGITS
    大åıij
    0.11
    ä¸įæĺ¯
    0.09
    æİĸ
    0.09
    éĥ½ä¸įæķ¢
    0.09
    обÑĭ
    0.09
    ius
    0.09
    æıIJåĢ¡
    0.09
    ä¸įæĪIJ
    0.09
     Gret
    0.09
    æĥ³åĪ°
    0.09
    Activations Density 0.039%

    No Known Activations