© 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. 35110
    Prev
    Next
    INDEX
    Explanations

    ever

    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
    竦
    -0.11
    ä¸įæŃ»
    -0.10
    theless
    -0.09
    å½»
    -0.09
    æĺ¯éŁ©åĽ½
    -0.09
    åħħ满
    -0.09
    iram
    -0.09
    åĪ¶çº¦
    -0.09
    èĪĤ
    -0.09
    _SCL
    -0.08
    POSITIVE LOGITS
     ever
    0.14
     jamais
    0.11
    æľ¬ç½ijç«Ļ
    0.10
    æ¯ı次éĥ½
    0.09
    ä»İæĿ¥
    0.09
    ä»»ä½ķæŶåĢĻ
    0.09
    å§ĭç»Īä¿ĿæĮģ
    0.09
    æ¯ı次
    0.09
     always
    0.08
    AME
    0.08
    Activations Density 0.100%

    No Known Activations