© 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. 53386
    Prev
    Next
    INDEX
    Explanations

    trouble with

    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
     goog
    -0.10
    èµŁ
    -0.09
    è°Ī
    -0.09
    å¼ĢæĿ¥
    -0.09
    人人éĥ½
    -0.09
    对ä¼ģä¸ļ
    -0.08
    è¡Įäºĭ
    -0.08
    对æѤ
    -0.08
    æĪij对
    -0.08
    对ä¸ŃåĽ½
    -0.08
    POSITIVE LOGITS
    è¿ĩçļĦ
    0.09
    inka
    0.09
    zia
    0.08
    å³Ļ
    0.08
    æľ´å®ŀ
    0.08
    lessly
    0.08
    -sama
    0.08
    äºĭæĥħ
    0.08
    à¹ģà¸Ĺ
    0.08
     vũ
    0.08
    Activations Density 0.033%

    No Known Activations