© 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. 63026
    Prev
    Next
    INDEX
    Explanations

    questions starting with "why"

    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
     ведÑĮ
    -0.12
    ../../../
    -0.11
    éħĭ
    -0.10
    hips
    -0.09
    ../../
    -0.09
    еÑģп
    -0.09
    缮çļĦæĺ¯
    -0.09
    aye
    -0.09
    INO
    -0.09
    oup
    -0.08
    POSITIVE LOGITS
    soever
    0.16
    /how
    0.13
    곡
    0.11
    äºĭåıij
    0.11
    è¿Ļä¹Īåģļ
    0.11
    cation
    0.11
    ymm
    0.09
    åģıåģı
    0.09
    zel
    0.09
    å¼Ģ设
    0.09
    Activations Density 0.039%

    No Known Activations