© 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. 53058
    Prev
    Next
    INDEX
    Explanations

    places and towns

    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
    æµ·æ·Ģ
    -0.10
     humming
    -0.09
    éĺĸ
    -0.09
    pectrum
    -0.09
     Broncos
    -0.09
    åµ
    -0.09
    碰
    -0.08
    æī¿å¾·
    -0.08
    çįIJ
    -0.08
    avana
    -0.08
    POSITIVE LOGITS
     Mississippi
    0.11
    olls
    0.10
    é»Ħæ²³
    0.10
     Arkansas
    0.10
    èĻ¢
    0.10
    èĬĪ
    0.10
     кÑĢÑĥ
    0.09
    .flink
    0.09
    ä¸įå½ĵ
    0.09
    æģ¸
    0.09
    Activations Density 0.380%

    No Known Activations