© 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. 4743
    Prev
    Next
    INDEX
    Explanations

    git repository operations

    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
    gment
    -0.10
    ç©¿æĪ´
    -0.09
    åīįè¿°
    -0.09
    æĻ®
    -0.09
     oppos
    -0.09
    临
    -0.09
    .payload
    -0.09
    _MALLOC
    -0.09
    ç©¿
    -0.08
    ç±į
    -0.08
    POSITIVE LOGITS
    åĴ«
    0.09
     trunk
    0.09
    å¼Ģåıij
    0.09
     airl
    0.09
    ests
    0.08
    åıijå±ķ模å¼ı
    0.08
    æľīä¸Ģ个
    0.08
    æĪIJåĵģ
    0.08
    åıijå±ķ
    0.08
    ä¸įå®ļ
    0.08
    Activations Density 0.542%

    No Known Activations