© 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. 26853
    Prev
    Next
    INDEX
    Explanations

    information appearing

    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
    oho
    -0.09
    è¸ı
    -0.09
    ä¸Ģè¾¹
    -0.09
     Gratis
    -0.08
    _COMPAT
    -0.08
    çł
    -0.08
    GLOBALS
    -0.08
    åĸĤ
    -0.08
    ìĽIJ
    -0.08
    IRA
    -0.08
    POSITIVE LOGITS
    æĪIJè´¥
    0.10
    å±ĢéĿ¢
    0.09
    åĽ½å®¶å®īåħ¨
    0.09
    箪
    0.09
    çļĦ责任
    0.09
    olen
    0.09
    æ³®
    0.09
    æªĦ
    0.09
     outcomes
    0.09
    åģľçķĻåľ¨
    0.09
    Activations Density 0.043%

    No Known Activations