© 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. 3593
    Prev
    Next
    INDEX
    Explanations

    be liable, held liable, shall be included, be binding, not be deemed

    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
    天çĶŁ
    -0.10
    æĸ¯åŁº
    -0.09
    iales
    -0.09
    erez
    -0.08
    iams
    -0.08
    ç»Ī身
    -0.08
    lico
    -0.08
    ifo
    -0.08
    å°ļ
    -0.08
    screen
    -0.08
    POSITIVE LOGITS
    åĮ£
    0.11
    年第
    0.10
    æī¿
    0.10
    çłľ
    0.10
    说起
    0.10
    åĮħåĮħ
    0.10
    买车
    0.10
    åİ
    0.09
     ÑģÑĩиÑĤа
    0.09
    ä¸ĭ称
    0.09
    Activations Density 0.019%

    No Known Activations