© 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. 54420
    Prev
    Next
    INDEX
    Explanations

    .gdx file extensions

    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
    åĸ½
    -0.09
    åķ°
    -0.08
    iden
    -0.08
    åŬ
    -0.08
    ob
    -0.08
    ä¸ĢæĬĬ
    -0.08
    idor
    -0.08
    ائÙĤ
    -0.08
    APPER
    -0.08
    ло
    -0.07
    POSITIVE LOGITS
    æ¹¾åĮº
    0.12
    twig
    0.09
    å²ļ
    0.09
    ÐŁÐµÑĤ
    0.08
    å·
    0.08
    먹
    0.08
    è´´
    0.08
    .gdx
    0.08
    æŁŀ
    0.08
    _centers
    0.08
    Activations Density 0.113%

    No Known Activations