© 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. 18574
    Prev
    Next
    INDEX
    Explanations

    left off last

    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
    æĭĽçīĮ
    -0.11
    би
    -0.10
    å©¿
    -0.10
    æĿ¥äºĨ
    -0.10
    大éĿ¢ç§¯
    -0.10
    .Unity
    -0.10
    åij½ä¸Ń
    -0.09
    _HAL
    -0.09
    åĬ¨æijĩ
    -0.09
    å±IJ
    -0.09
    POSITIVE LOGITS
    éģĹçķĻ
    0.17
    çķĻç»Ļ
    0.15
    åģľçķĻåľ¨
    0.13
    çķĻ
    0.12
    离å¼Ģ
    0.12
     leave
    0.12
    Last
    0.11
     leftover
    0.11
    Leave
    0.11
    æĪĽ
    0.11
    Activations Density 0.118%

    No Known Activations