© 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. 53509
    Prev
    Next
    INDEX
    Explanations

    nobody or one

    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
    ç¦ıå¾·
    -0.09
    æĢı
    -0.09
    æŁIJç§į
    -0.09
    ä¸įåħ¬å¹³
    -0.08
    ucci
    -0.08
    干线
    -0.08
    海峡
    -0.08
    imas
    -0.07
     Oak
    -0.07
    ror
    -0.07
    POSITIVE LOGITS
    theless
    0.15
    åĴĮ社ä¼ļ
    0.11
    crets
    0.11
    osate
    0.10
    agara
    0.10
    ìĹŃ
    0.10
    اص
    0.10
    åIJ¬è¯´è¿ĩ
    0.10
    没人
    0.10
    æ·
    0.10
    Activations Density 0.020%

    No Known Activations