© 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. 58315
    Prev
    Next
    INDEX
    Explanations

    gaps, lines, streaks, slack, thick

    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
    ç»ĿæľĽ
    -0.10
    ób
    -0.08
    åĶĶ
    -0.08
    ä¹Łå¥½
    -0.08
    çĤ¯
    -0.08
    ller
    -0.08
    çĿĽ
    -0.08
    ä¸įå¤ļ
    -0.08
    æĻ´
    -0.08
    [arg
    -0.08
    POSITIVE LOGITS
    太å¤ļ
    0.13
    太å¤ļçļĦ
    0.11
    äºĨä¸ĢäºĽ
    0.10
    è¿ĩå¤ļ
    0.10
    è¿ĩäºİ
    0.10
    éļ¼
    0.09
    è¾ĥå¤ļ
    0.09
    è¿ĩ大
    0.09
    åĩ°
    0.09
     ÙĪغÙĬر
    0.09
    Activations Density 0.241%

    No Known Activations