© 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. 48419
    Prev
    Next
    INDEX
    Explanations

    takes some time

    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
    ç¢į
    -0.10
    illet
    -0.09
    dataProvider
    -0.09
    hoff
    -0.09
    .Library
    -0.09
    zeigen
    -0.09
    å«Į
    -0.08
    éģįåıĬ
    -0.08
    åĪ¶
    -0.08
    ź
    -0.08
    POSITIVE LOGITS
    _scalar
    0.09
    åѦåĪ°
    0.09
    ä¸Ģ个å¤ļæľĪ
    0.09
    ä¸ļä½Ļ
    0.09
    æĬĬèĩªå·±çļĦ
    0.08
    éĢĶ
    0.08
    æIJ¶
    0.08
    éĹ¼
    0.08
    çħ½
    0.08
     ot
    0.08
    Activations Density 0.035%

    No Known Activations