© 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. 55711
    Prev
    Next
    INDEX
    Explanations

    described herein

    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
    转åĮĸ
    -0.10
    æ¯Ļ
    -0.09
     Naw
    -0.09
    jack
    -0.09
    çļĦç§ĺå¯Ĩ
    -0.09
    å¾Ģå¤ĸ
    -0.09
    ç»Ī
    -0.08
    è¦ĨçĽĸçİĩ
    -0.08
    å²Ń
    -0.08
    åĢĶ
    -0.08
    POSITIVE LOGITS
    æķĻ导
    0.09
    åĽ¾çº¸
    0.09
     descri
    0.08
    lobals
    0.08
     indefinite
    0.08
     Pon
    0.08
    _HDR
    0.08
    ä»İä¸ļ
    0.08
    BD
    0.08
    æııè¿°
    0.08
    Activations Density 0.685%

    No Known Activations