© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 36533
    Prev
    Next
    INDEX
    Explanations

    jump or fall from heights

    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
    æĭ¿æĿ¥
    -0.11
    ä¸įèĥ½åĨį
    -0.11
    çIJĨä¼ļ
    -0.10
    umed
    -0.10
    /slick
    -0.10
    å½ķ
    -0.10
    èįŁ
    -0.09
    è¡Įä¸ļåĨħ
    -0.09
    éĹ²
    -0.09
    åĭ¤
    -0.09
    POSITIVE LOGITS
    æĬķ
    0.11
    è·Į
    0.11
    é£ŀè·ĥ
    0.10
     leap
    0.10
    è·³è·ĥ
    0.10
    è·ĥ
    0.10
    åĸĬ
    0.10
    é£Ľãģ³
    0.10
    åĢĴåħ¥
    0.10
    Leap
    0.10
    Activations Density 0.053%

    No Known Activations