© 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. 31247
    Prev
    Next
    INDEX
    Explanations

    Hugo or Hugh names

    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
    åĭĥåĭĥ
    -0.11
    vez
    -0.09
    ials
    -0.09
    al
    -0.09
    nya
    -0.09
    å¿ĥæĥ³
    -0.09
    ple
    -0.09
    æīª
    -0.09
    è´¨åľ°
    -0.08
    æĺĢ
    -0.08
    POSITIVE LOGITS
    andan
    0.11
    itarian
    0.11
    ossal
    0.10
    毳
    0.09
    alis
    0.09
    اÙĨÙĬ
    0.09
    andy
    0.09
    osity
    0.09
    altar
    0.09
    abar
    0.08
    Activations Density 0.008%

    No Known Activations