© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 21640
    Prev
    Next
    INDEX
    Explanations

    layout definitions and uses

    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
    nal
    -0.09
    åł
    -0.09
    oodle
    -0.09
    بخ
    -0.09
    bpp
    -0.09
     обÑĥÑĩ
    -0.09
    agma
    -0.09
    è¡Įéķ¿
    -0.09
    rous
    -0.09
    antor
    -0.08
    POSITIVE LOGITS
    .createSequentialGroup
    0.15
    .addWidget
    0.13
    .createParallelGroup
    0.11
    çħ
    0.09
    ASON
    0.09
    å¸ĥå±Ģ
    0.09
    uring
    0.09
    anes
    0.09
    ter
    0.09
    urations
    0.09
    Activations Density 0.029%

    No Known Activations