© 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. 50764
    Prev
    Next
    INDEX
    Explanations

    range

    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
    bern
    -0.10
    ä¸Ģç«Ļå¼ı
    -0.09
    addle
    -0.09
     Bing
    -0.09
    å´Ľ
    -0.09
    轸
    -0.09
    agger
    -0.09
    æľ¬ç«Ļ
    -0.08
     Tiên
    -0.08
    é¦ĸéĥ½
    -0.08
    POSITIVE LOGITS
    èĮĥåĽ´
    0.25
     range
    0.25
    range
    0.23
    èĮĥåĽ´åĨħ
    0.21
     Range
    0.19
    _range
    0.19
    Range
    0.19
    .range
    0.19
    (range
    0.18
    -range
    0.18
    Activations Density 0.206%

    No Known Activations