© 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. 54576
    Prev
    Next
    INDEX
    Explanations

    again and another

    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
    é¦ĸå±Ĭ
    -0.10
    ä¸Ģ缴
    -0.09
    å¼ĢåıijåĮº
    -0.09
    ä¸ĵ
    -0.09
    é¦ĸæī¹
    -0.09
    æѤåIJİ
    -0.09
    ä»İæѤ
    -0.09
    è¿İåIJĪ
    -0.08
    æĬĵä½ı
    -0.08
    ä¸ĢåIJij
    -0.08
    POSITIVE LOGITS
    åıĪä¸Ģ次
    0.13
    åĨįä¸Ģ次
    0.12
     оÑĩеÑĢед
    0.11
    again
    0.10
     AGAIN
    0.10
    AGAIN
    0.09
    滥
    0.09
     wieder
    0.09
     again
    0.09
    å¦ĤæľŁ
    0.09
    Activations Density 0.096%

    No Known Activations