© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 8122
    Prev
    Next
    INDEX
    Explanations

    consider these

    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
    æļĮ
    -0.10
    æľīä¿¡å¿ĥ
    -0.09
    èĪĢ
    -0.09
    itta
    -0.09
    å¾Īå¤ļæŶåĢĻ
    -0.09
    女åıĭ
    -0.09
    .cam
    -0.08
    æ·Ŀ
    -0.08
    _TEX
    -0.08
    好åIJĹ
    -0.08
    POSITIVE LOGITS
    ably
    0.10
    éĹ®
    0.09
    对å¾ħ
    0.08
    ãģĤãģ®
    0.08
     how
    0.08
    éĹ®éĹ®
    0.08
    åħ¸åŀĭçļĦ
    0.08
     drain
    0.08
    nob
    0.08
     ÙĤÙĪÙĦÙĩ
    0.08
    Activations Density 0.070%

    No Known Activations