© 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. 50347
    Prev
    Next
    INDEX
    Explanations

    look up, lookup

    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
    临è¿ij
    -0.09
    angep
    -0.09
    åī½
    -0.08
    鼷éľĨ
    -0.08
    imator
    -0.08
    ä¾
    -0.08
    ä¸Ģä½ĵ
    -0.08
    åĿIJä¸ĭ
    -0.08
    .mm
    -0.07
    ksen
    -0.07
    POSITIVE LOGITS
    æĬ¬èµ·å¤´
    0.13
    天èĬ±æĿ¿
    0.12
    æĬ¬å¤´
    0.11
    ä»°
    0.10
    å¾ĩ
    0.09
    æĬ¬èµ·
    0.09
    éķ£
    0.09
    èĵ®
    0.09
    unning
    0.09
    ä»¥äººä¸ºæľ¬
    0.09
    Activations Density 0.016%

    No Known Activations