© 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. 54617
    Prev
    Next
    INDEX
    Explanations

    licenses and identifiers

    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
    reed
    -0.09
     bust
    -0.09
     Roller
    -0.08
    è¿ĺåİŁ
    -0.08
     olacaģı
    -0.08
    éĤ£ä¸ĢåĪ»
    -0.08
    çŁĽçĽ¾
    -0.07
    olin
    -0.07
     shedding
    -0.07
    rogen
    -0.07
    POSITIVE LOGITS
    ä¸įå½±åĵį
    0.11
    æĬ¥åIJį
    0.10
    Ñģен
    0.10
    ä»įæľī
    0.10
    赶ä¸Ĭ
    0.10
    _WRAP
    0.10
    æĮ¤
    0.10
    éĢĢç¨İ
    0.09
    èķī
    0.09
    INGER
    0.09
    Activations Density 0.020%

    No Known Activations