© 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. 27914
    Prev
    Next
    INDEX
    Explanations

    numbers often followed by a fraction

    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
    oir
    -0.09
    моÑĢ
    -0.09
     Cuomo
    -0.08
     squash
    -0.08
    shore
    -0.08
    å·¡åĽŀ
    -0.08
    çªģçł´
    -0.08
    çªģçł´åı£
    -0.08
    åľ¨äº¬
    -0.08
    å¸ħ
    -0.08
    POSITIVE LOGITS
    纪
    0.10
     пÑģ
    0.09
    åģĩæĹ¥
    0.09
    åºĶåıĬæŶ
    0.09
     Hick
    0.08
    é£İåIJ¹
    0.08
    ãģ°ãģĭãĤĬ
    0.08
    åĬ¿
    0.08
    ç¾İ丽çļĦ
    0.08
    è°ij
    0.08
    Activations Density 0.037%

    No Known Activations