© 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. 4709
    Prev
    Next
    INDEX
    Explanations

    register

    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
    çĨı
    -0.08
    -ranked
    -0.08
    4
    -0.08
    å¤ĸç±į
    -0.08
     гоÑĤов
    -0.08
    3
    -0.08
    1
    -0.08
    èĨĬ
    -0.07
    0
    -0.07
    _PD
    -0.07
    POSITIVE LOGITS
    eneration
    0.09
    大æ°Ķ
    0.09
    ulate
    0.09
    Sy
    0.09
     Stall
    0.09
    lund
    0.09
    _endian
    0.09
    ä¸Ŀ
    0.08
    éķĽ
    0.08
    æģ©
    0.08
    Activations Density 0.015%

    No Known Activations