© 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. 9273
    Prev
    Next
    INDEX
    Explanations

    called, named, or referred to as

    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
    è¨Ĺ
    -0.10
    hower
    -0.10
     Visible
    -0.10
    Ы
    -0.09
    _SUPPLY
    -0.09
    åѦéĹ®
    -0.09
    åīĭ
    -0.09
    ÃĿ
    -0.09
    arella
    -0.09
    hana
    -0.08
    POSITIVE LOGITS
     contr
    0.09
     Vand
    0.09
     dab
    0.09
    èµ°åĩºåİ»
    0.09
    éķ¿å¯¿
    0.09
    åĦĨ
    0.09
     ones
    0.09
    家ä¼Ļ
    0.08
    aff
    0.08
    åľ¨æĪijåĽ½
    0.08
    Activations Density 0.023%

    No Known Activations