© 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. 50094
    Prev
    Next
    INDEX
    Explanations

    code is licensed

    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
    rance
    -0.09
    符
    -0.08
    æĿ¯
    -0.08
    åIJįæ°Ķ
    -0.08
    ä¸ĢæĿ¯
    -0.08
    大ä¸ĵ
    -0.08
    поÑĩ
    -0.08
    è¥
    -0.08
    пон
    -0.08
     stør
    -0.07
    POSITIVE LOGITS
    æİ°
    0.09
    izzazione
    0.09
    å¢Ł
    0.09
    _probe
    0.09
     gri
    0.09
    æģŃ
    0.09
    éĥ¨åĪĨåĨħ容
    0.08
    åĬ¨èĦī
    0.08
    odos
    0.08
    iske
    0.08
    Activations Density 0.034%

    No Known Activations