© 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. 8824
    Prev
    Next
    INDEX
    Explanations

    programming keywords

    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
     Kad
    -0.08
    atra
    -0.08
    имеÑĢ
    -0.08
    OAD
    -0.08
     ones
    -0.08
     via
    -0.08
    Associated
    -0.08
    óc
    -0.08
    æĸĻçIJĨ
    -0.08
     Ced
    -0.07
    POSITIVE LOGITS
    æ§
    0.10
    èĹķ
    0.10
    -mask
    0.09
    åĿij
    0.09
    çľĭ好
    0.09
    å¾Īæ·±
    0.09
    éķ¿æ±Ł
    0.09
    æĥ¹
    0.09
    arrera
    0.08
    æį»
    0.08
    Activations Density 1.156%

    No Known Activations