© 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. 52929
    Prev
    Next
    INDEX
    Explanations

    spelling words that end with 'ire', 'ney', 'ck', 'ill', or 'vell'

    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
    æĢ§å¼º
    -0.10
    æĢ§çļĦ
    -0.09
    çķĮçļĦ
    -0.09
    橱
    -0.09
    ario
    -0.09
    AIL
    -0.09
    _rq
    -0.09
    æ°ijçĶ¨
    -0.09
    æĸ°ä¸ī
    -0.08
    mall
    -0.08
    POSITIVE LOGITS
    åıĬåħ¶
    0.12
    /her
    0.10
    çķĶ
    0.09
    (Char
    0.09
    夫å¦ĩ
    0.09
    åĴĮä»ĸçļĦ
    0.09
    ä¸Ģæĸ¹
    0.09
    以åıĬ
    0.08
    çļĦ妻åŃIJ
    0.08
    PCS
    0.08
    Activations Density 0.103%

    No Known Activations