© 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. 30405
    Prev
    Next
    INDEX
    Explanations

    specific words and punctuation

    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
    éĤª
    -0.09
    éģ¿å¼Ģ
    -0.09
    约å®ļ
    -0.09
    é¢Į
    -0.09
    å»īä»·
    -0.09
    åħĭæľį
    -0.09
    æĦŁåı¹
    -0.09
    侪
    -0.08
    åħ´æĹº
    -0.08
    æŃ£åĽłä¸º
    -0.08
    POSITIVE LOGITS
    è¨Ģ论
    0.10
    åĪģ
    0.10
    éĿ¢åīį
    0.09
    ftar
    0.09
    quo
    0.09
    åIJı
    0.09
    нак
    0.09
    гон
    0.08
    çļĦæĶ¯æĮģ
    0.08
     quo
    0.08
    Activations Density 0.142%

    No Known Activations