© 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. 31484
    Prev
    Next
    INDEX
    Explanations

    spread of information

    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
    år
    -0.10
    ä¸İä¸ŃåĽ½
    -0.09
    yen
    -0.09
    cken
    -0.09
    éŁ³
    -0.09
    çķĻç»Ļ
    -0.08
    à¦ĵ
    -0.08
     pudd
    -0.08
     Mes
    -0.08
    atak
    -0.08
    POSITIVE LOGITS
    ä½ľåĿĬ
    0.10
    ç±³é¥Ń
    0.10
     wireType
    0.10
    以åħį
    0.09
    åĽĽæĸ¹
    0.09
    ä¸ļåĨħ
    0.09
    -append
    0.09
    çŁ¥æĻĵ
    0.09
    è¾į
    0.09
     нам
    0.09
    Activations Density 0.031%

    No Known Activations