© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 9094
    Prev
    Next
    INDEX
    Explanations

    available in various contexts

    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
    éļıå¤Ħåı¯è§ģ
    -0.09
    zo
    -0.08
     puzz
    -0.08
    PU
    -0.08
    Guest
    -0.08
    trimmed
    -0.08
     pam
    -0.08
    ana
    -0.08
     spoiler
    -0.08
    æ··ä¹±
    -0.07
    POSITIVE LOGITS
    æľīä¸įåIJĮçļĦ
    0.12
    å¾Ĺå¾Ī
    0.11
    æľī两ç§į
    0.10
    ä¸īç§į
    0.10
    æľīå¾Īå¤ļç§į
    0.10
    åĪĨ为
    0.10
    åı¯åĪĨ为
    0.09
    æĸ¯å¡Ķ
    0.09
    å¤ļç§įå½¢å¼ı
    0.09
    ç§ij室
    0.09
    Activations Density 0.049%

    No Known Activations