© 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. 56321
    Prev
    Next
    INDEX
    Explanations

    larger than life

    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
    _MR
    -0.09
    ä¸Ģå°ģ
    -0.08
    encies
    -0.08
    chw
    -0.08
    ANEL
    -0.08
    äºŃ
    -0.08
     inc
    -0.08
    ÑĤок
    -0.08
    chia
    -0.08
    olars
    -0.08
    POSITIVE LOGITS
    丰满
    0.09
    æ¦Ħ
    0.09
    æĦı大
    0.09
    åŀŃ
    0.09
    纱
    0.09
    è¶´åľ¨
    0.09
    ä¸ĸçķĮåIJĦåĽ½
    0.08
    éħįå¥Ĺ
    0.08
    æ¦
    0.08
    ä¸Ģå±Ĥ
    0.08
    Activations Density 0.021%

    No Known Activations