© 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. 23050
    Prev
    Next
    INDEX
    Explanations

    no wonder / not surprising

    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
     должноÑģÑĤ
    -0.10
    ilton
    -0.10
    Lorem
    -0.10
    æĺİç¡®è§Ħå®ļ
    -0.09
    urch
    -0.09
    çļĦåį±å®³
    -0.08
    ä¸įåı¯ç¼ºå°ij
    -0.08
    Callable
    -0.08
    éĢļè¿ĩ对
    -0.08
     ÑģпеÑĨиалÑĮно
    -0.08
    POSITIVE LOGITS
    å¦ĤæѤ
    0.12
    è¿ĻèĪ¬
    0.11
    éļ¾æĢª
    0.10
    assed
    0.09
    rox
    0.09
    顺åĬ¿
    0.09
    ä½Ĩè¿Ļ
    0.09
    ä¼ļè§īå¾Ĺ
    0.08
    ç»ķ
    0.08
     Ashe
    0.08
    Activations Density 0.042%

    No Known Activations