© 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. 62607
    Prev
    Next
    INDEX
    Explanations

    upcoming publications, competitions, commentary, reviews

    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
    èŁ¾
    -0.09
    erta
    -0.09
    åĨįéĢł
    -0.09
    çī¢è®°ä½¿åij½
    -0.09
    atti
    -0.09
     anlaÅŁ
    -0.08
    å¢
    -0.08
     FEMA
    -0.08
    åĿŁ
    -0.08
    åĺ¶
    -0.08
    POSITIVE LOGITS
    ROC
    0.09
    chooser
    0.09
    instructions
    0.08
     Ñĩел
    0.08
    æŬ
    0.08
    éļľç¢į
    0.08
    ÑģоÑģ
    0.08
    æĭĴ
    0.08
    ä¿Ŀå®Ī
    0.08
     Mao
    0.08
    Activations Density 0.008%

    No Known Activations