© 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. 51083
    Prev
    Next
    INDEX
    Explanations

    Sumerian

    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
    离
    -0.10
    å¸ĮèħĬ
    -0.09
    çĨ
    -0.09
    棱
    -0.09
    çĩĥ
    -0.09
    æ¸ħæľĿ
    -0.09
    身
    -0.09
    aż
    -0.09
     scratch
    -0.08
    æĶ¾æĿ¾
    -0.08
    POSITIVE LOGITS
    ientes
    0.10
    пон
    0.10
    èĩ³ä¸Ĭ
    0.09
    宥
    0.09
    æĴħ
    0.09
    udad
    0.08
    æĶ¶åĽŀ
    0.08
    æĶ¿åĬ¡
    0.08
     recess
    0.08
     enth
    0.08
    Activations Density 0.042%

    No Known Activations