© 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. 60972
    Prev
    Next
    INDEX
    Explanations

    work 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
    åīij
    -0.11
    èł¢
    -0.09
    æĿ¥è¯´
    -0.08
    åĬį
    -0.08
    æ´Ĺ
    -0.08
    æŁĦ
    -0.08
    åĿļ
    -0.08
    åĽ½åľŁ
    -0.08
    èĭĵ
    -0.08
     IV
    -0.08
    POSITIVE LOGITS
    寡
    0.10
    å¼Ģåĩº
    0.09
    å°ıå¿ĥ
    0.09
     rever
    0.09
    .tie
    0.09
     inf
    0.09
    å«Ĵ
    0.09
     Ñĩел
    0.08
    logged
    0.08
    æ¾į
    0.08
    Activations Density 0.053%

    No Known Activations