© 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. 56556
    Prev
    Next
    INDEX
    Explanations

    irreversible

    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
    omo
    -0.09
    WO
    -0.08
     virt
    -0.08
    è¾¾ä¸įåĪ°
    -0.08
     MCC
    -0.08
    ITS
    -0.08
    ä¸Ģå¤ľ
    -0.08
    3
    -0.07
     prolonged
    -0.07
    /cc
    -0.07
    POSITIVE LOGITS
    ä¸įåı¯
    0.23
    ä¸įèĥ½åĨį
    0.13
     irreversible
    0.13
    ä¸įåı¯ä»¥
    0.13
    éĻ¤éĿŀ
    0.12
    åıµ
    0.11
     нелÑĮзÑı
    0.11
    å¿ħä¸įåı¯
    0.11
    _CANNOT
    0.10
    éĢĨ转
    0.10
    Activations Density 0.047%

    No Known Activations