© 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. 2979
    Prev
    Next
    INDEX
    Explanations

    be on, be lost, be broken, be giving, be able, be free, be unavailable

    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
    çļĦ姿æĢģ
    -0.09
    姿æĢģ
    -0.09
    姿
    -0.08
    _FD
    -0.08
    èIJ¥ä¸ļ
    -0.08
    èħ¹
    -0.08
     JD
    -0.08
    estate
    -0.08
    thouse
    -0.08
    ä»ģ
    -0.08
    POSITIVE LOGITS
    acci
    0.10
    缴éĢļ车
    0.10
    ãĤĭãĤĤãģ®
    0.09
    éĴĹ
    0.09
    uben
    0.08
    é³ĸ
    0.08
    两次
    0.08
    æĬ¢åħĪ
    0.08
    irect
    0.08
    å°½æĹ©
    0.08
    Activations Density 0.003%

    No Known Activations