© 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. 21631
    Prev
    Next
    INDEX
    Explanations

    mis- prefix indicating error or incorrectness

    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
     Crunch
    -0.09
    ilor
    -0.09
     vic
    -0.09
    æķĻè®Ń
    -0.09
    gom
    -0.09
    oleon
    -0.09
    izer
    -0.09
    ëĿ½
    -0.08
    asca
    -0.08
    ossa
    -0.08
    POSITIVE LOGITS
    ellaneous
    0.10
    çĶļèĩ³æĺ¯
    0.10
    åıijæĺİ
    0.10
    /un
    0.09
    æĪĸèĢħ说
    0.09
    çĶļèĩ³
    0.09
    CU
    0.08
    åºĶä»ĺ
    0.08
    åĽŀäºĭ
    0.08
    éĥ¢
    0.08
    Activations Density 0.056%

    No Known Activations