© 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. 55720
    Prev
    Next
    INDEX
    Explanations

    get away with

    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
    æĸĩåĮĸä¼łæĴŃ
    -0.10
     пÑĢид
    -0.10
    èŀ³
    -0.10
    ãģ©ãģĵãĤį
    -0.09
    åĦ¿
    -0.09
    elay
    -0.09
    .TryParse
    -0.09
    Ñıвление
    -0.09
    dos
    -0.09
    大è±Ĩ
    -0.09
    POSITIVE LOGITS
     underway
    0.09
     hull
    0.09
    é¦ĸä½į
    0.09
    åĩºäºİ
    0.08
    å¼Ģå§ĭäºĨ
    0.08
    hana
    0.08
    è¿ĩåİ»äºĨ
    0.08
    çŃ®
    0.08
    Ñĥз
    0.08
    /tiny
    0.08
    Activations Density 0.047%

    No Known Activations