© 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. 32454
    Prev
    Next
    INDEX
    Explanations

    call to action or challenge

    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
    éłĨ
    -0.10
    顺
    -0.10
    adio
    -0.09
    顺çĿĢ
    -0.09
    就说
    -0.08
    æĪIJçĨŁçļĦ
    -0.08
    krit
    -0.08
    èĩªä¹ł
    -0.08
    éģ´
    -0.08
    ons
    -0.08
    POSITIVE LOGITS
    åIJ©åĴIJ
    0.10
    åı·åı¬
    0.10
    åĺ±
    0.09
    éĩįä»»
    0.09
    OLA
    0.09
    æħ·æħ¨
    0.09
    DataTask
    0.09
    åįı
    0.09
    éŀŃ
    0.09
    çIJĨäºĭä¼ļ
    0.09
    Activations Density 0.062%

    No Known Activations