© 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. 10062
    Prev
    Next
    INDEX
    Explanations

    damaged or corruption

    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
    arella
    -0.09
    olo
    -0.09
    ardin
    -0.09
    çķĻä¸ĭæĿ¥
    -0.09
    ä¸ºæľŁ
    -0.08
    æĽ´é«ĺ
    -0.08
    æĮģèĤ¡
    -0.08
     blind
    -0.08
    æĽ´å¼º
    -0.08
    æĤ©ãģ¿
    -0.08
    POSITIVE LOGITS
    污æŁĵ
    0.13
    严éĩį
    0.13
    æįŁåĿı
    0.13
    åıĹåĪ°äºĨ
    0.13
    æ±ļ
    0.13
    已被
    0.12
    åĿıäºĨ
    0.12
     corrupted
    0.12
    èħIJ
    0.12
    åıĹåĪ°
    0.11
    Activations Density 0.167%

    No Known Activations