© 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. 51951
    Prev
    Next
    INDEX
    Explanations

    dead horse, dead zone, dead letter, deadlocked

    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
    ilden
    -0.11
    ç¨įç¨į
    -0.10
    å·Ŀ
    -0.09
    å¤įå·¥å¤į
    -0.09
     Kendall
    -0.08
    æĿ¡ä¾ĭ
    -0.08
    IGHLIGHT
    -0.08
    å²Ĥ
    -0.08
     setResult
    -0.08
    éĩİ
    -0.08
    POSITIVE LOGITS
    (dead
    0.17
     dead
    0.17
    dead
    0.16
    æŃ»
    0.16
    Dead
    0.14
     Dead
    0.13
    .dead
    0.12
    æŃ»äºĨ
    0.12
    ä¸įçĶ¨æĭħå¿ĥ
    0.11
    _dead
    0.11
    Activations Density 0.043%

    No Known Activations