© 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. 4582
    Prev
    Next
    INDEX
    Explanations

    loss or accident

    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
    å½ĵåľ°äºº
    -0.10
    èĮ
    -0.09
    çķ¸
    -0.09
    åħĪè¡Į
    -0.08
    ä»ĺ
    -0.08
    æ¯ı次éĥ½
    -0.08
     shel
    -0.08
    ê»ĺ
    -0.08
    .MSG
    -0.08
    ETYPE
    -0.08
    POSITIVE LOGITS
    çŃīåİŁåĽł
    0.11
    æĴķ
    0.10
    çĪ±ä¸Ĭ
    0.10
    失åİ»äºĨ
    0.10
     mất
    0.09
    æīĵå¼ĢäºĨ
    0.09
    丢失
    0.09
    çĵ¢
    0.09
    è¿Ļæł·çļĦè¯Ŀ
    0.09
    è¿Ľåħ¥äºĨ
    0.09
    Activations Density 0.069%

    No Known Activations