© 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. 23318
    Prev
    Next
    INDEX
    Explanations

    free software license

    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
    ubar
    -0.08
     Downing
    -0.08
    linger
    -0.08
    ä½Ĺ
    -0.08
    æĭĽåĭŁ
    -0.08
    dbl
    -0.08
    æľķ
    -0.08
    åŁºå±Ĥ
    -0.08
    rames
    -0.08
    æĭĽ
    -0.08
    POSITIVE LOGITS
     yazılı
    0.10
    åĢĶ
    0.10
    ä¸įå¼Ģ
    0.09
    å´Ľ
    0.09
    è¿Ļ件
    0.09
    è¢
    0.09
    MERCHANTABILITY
    0.08
     بعÙĬد
    0.08
    麸
    0.08
    æ´ĭ
    0.08
    Activations Density 3.248%

    No Known Activations