© 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. 7554
    Prev
    Next
    INDEX
    Explanations

    years old

    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®Ģæ´ģ
    -0.09
    onom
    -0.09
    dio
    -0.08
    /operators
    -0.08
    chie
    -0.08
    Nx
    -0.08
    ä½ij
    -0.08
    æģĤ
    -0.08
     cyclists
    -0.08
    æĥ°
    -0.08
    POSITIVE LOGITS
    åIJģ
    0.08
    .sa
    0.08
     Synthetic
    0.08
    å¾Ģå¤ĸ
    0.08
    Òĵ
    0.08
    жа
    0.08
    /raw
    0.08
    .hxx
    0.08
    ä»ĸ们çļĦ
    0.08
    ä»Ĭ天å°ıç¼ĸ
    0.08
    Activations Density 0.071%

    No Known Activations