© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 8142
    Prev
    Next
    INDEX
    Explanations

    apache commons libraries

    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
    ivity
    -0.11
    zelf
    -0.09
    çļ¿
    -0.09
    avor
    -0.09
    etro
    -0.08
    ç¬ijè¯Ŀ
    -0.08
    ä½£
    -0.08
    erman
    -0.08
    æĬĵç´§
    -0.08
     hundred
    -0.08
    POSITIVE LOGITS
    èĵĿ天
    0.10
    .wik
    0.09
    æ¦Ķ
    0.09
    latex
    0.08
    entes
    0.08
    unken
    0.08
     collapse
    0.08
    Collapse
    0.08
     collapsed
    0.08
    truth
    0.08
    Activations Density 0.016%

    No Known Activations