© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 40022
    Prev
    Next
    INDEX
    Explanations

    rounding to nearest integer/decimal

    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
    æħ¢æħ¢
    -0.10
    hands
    -0.10
    éĿ¶
    -0.09
     вÑĭÑģоко
    -0.09
    ember
    -0.09
    å»ī
    -0.09
    ysi
    -0.09
    讼
    -0.09
    两项
    -0.09
     fund
    -0.08
    POSITIVE LOGITS
    精确
    0.10
    ноÑģÑĤ
    0.09
     exact
    0.09
    arda
    0.09
    ç»ĺ
    0.08
    度
    0.08
    个ä½ĵ
    0.08
    缺
    0.08
     exactly
    0.08
    è¿ĻçĤ¹
    0.08
    Activations Density 0.028%

    No Known Activations