© 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. 1680
    Prev
    Next
    INDEX
    Explanations

    number of times or iterations

    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
    åĬłå¼ºå¯¹
    -0.09
    åħ½
    -0.09
     ÑĥÑģи
    -0.08
    åīįä¸Ģ天
    -0.08
    æĽ´å¼º
    -0.08
    INGER
    -0.08
    ACS
    -0.08
    renc
    -0.08
     intensified
    -0.08
    顽强
    -0.08
    POSITIVE LOGITS
    è¿Ļæł·åģļ
    0.13
    è¿Ļä¹Īåģļ
    0.12
    累积
    0.11
    以æѤ
    0.10
    è¿Ļæł·çļĦ
    0.10
    缸继
    0.10
    åı¯æĮģç»Ń
    0.09
    累计
    0.09
    è¿Ļç§įäºĭæĥħ
    0.09
    opp
    0.09
    Activations Density 0.295%

    No Known Activations