© 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. 7566
    Prev
    Next
    INDEX
    Explanations

    football player positions

    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
    亶
    -0.10
    è¿ĺæĥ³
    -0.10
    æĥ³
    -0.09
     treasure
    -0.09
    itution
    -0.09
    ipation
    -0.09
    çľīæ¯Ľ
    -0.09
    _Params
    -0.09
    enty
    -0.09
    gebung
    -0.09
    POSITIVE LOGITS
    åĪ°ä½į
    0.10
     Berk
    0.09
    /mail
    0.08
    é«ĺé¢Ŀ
    0.08
    .vol
    0.08
    é¦ĸ次
    0.08
    们
    0.08
    楫
    0.08
    宸
    0.08
    éŨè¯Ĭ
    0.08
    Activations Density 0.046%

    No Known Activations