© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 10546
    Prev
    Next
    INDEX
    Explanations

    lock or locked

    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
    æªIJ
    -0.10
    å²ĺ
    -0.10
    æIJ¬è¿ģ
    -0.10
    ä¸įéĹ´æĸŃ
    -0.09
    roof
    -0.09
     bearer
    -0.09
    lint
    -0.09
    íĮIJ
    -0.09
    iros
    -0.09
    essen
    -0.09
    POSITIVE LOGITS
    éĶģå®ļ
    0.15
    -lock
    0.13
     Locked
    0.12
    ä½ıäºĨ
    0.12
    éĶģ
    0.12
     locked
    0.11
     Lock
    0.11
    locked
    0.11
     lock
    0.11
    解éĶģ
    0.11
    Activations Density 0.105%

    No Known Activations