© 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. 22554
    Prev
    Next
    INDEX
    Explanations

    then, eventually, finally

    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
    æĪIJç«ĭ以æĿ¥
    -0.11
    以æĿ¥
    -0.09
    ç«ĭåį³
    -0.09
    èĨĬ
    -0.09
     immediately
    -0.09
    æĿ¥ä¸įåıĬ
    -0.09
    太é«ĺ
    -0.09
    èį¯ä¸ļ
    -0.09
    åı¦æľī
    -0.09
    è¿ħ
    -0.09
    POSITIVE LOGITS
    æľĢç»Ī
    0.24
     eventually
    0.24
    ç»Īäºİ
    0.23
     Eventually
    0.22
     finally
    0.21
     eventual
    0.21
    æľĢçµĤ
    0.20
    Eventually
    0.19
    æ¸IJæ¸IJ
    0.19
    æľĢåIJİ
    0.18
    Activations Density 0.144%

    No Known Activations