© 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. 53864
    Prev
    Next
    INDEX
    Explanations

    final part or last occurrence

    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
    ä¸Ńå°ıåŀĭ
    -0.11
    ä¼łç»ŁçļĦ
    -0.10
    ধ
    -0.09
    åİŁæľīçļĦ
    -0.09
    åIJĦæľī
    -0.09
    ä¸ŀ
    -0.09
    çķªèĮĦ
    -0.08
    itus
    -0.08
    upe
    -0.08
    æľªç»ı
    -0.08
    POSITIVE LOGITS
    imos
    0.10
    è¿Łè¿Ł
    0.10
    ì§Ģ를
    0.09
    çªĿ
    0.08
    ByExample
    0.08
    agers
    0.08
    айн
    0.08
    лÑıем
    0.07
    ters
    0.07
    æĬļ
    0.07
    Activations Density 0.007%

    No Known Activations