© 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. 31345
    Prev
    Next
    INDEX
    Explanations

    I think / I was

    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
    åIJ¹
    -0.09
    dept
    -0.09
    ç¤ģ
    -0.08
    esimal
    -0.08
    泡
    -0.08
    zes
    -0.08
    éĻª
    -0.08
    ’n
    -0.08
    dez
    -0.08
    brain
    -0.08
    POSITIVE LOGITS
    æ¯Ķå¦Ĥ说
    0.09
    éĢĢåĩº
    0.09
    lingen
    0.09
    åĤ¢
    0.09
    è¾Ľ
    0.09
    çķ
    0.09
    åħĪçĶŁ
    0.09
    ä¸İä¸ŃåĽ½
    0.09
    æĭħä¿Ŀ
    0.08
    ä¸ŃåĽ½å¤§éĻĨ
    0.08
    Activations Density 0.044%

    No Known Activations