© 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. 4633
    Prev
    Next
    INDEX
    Explanations

    certainty and doubt

    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
    jej
    -0.10
     past
    -0.09
    æİ¥
    -0.09
    å¾Ģäºĭ
    -0.09
    è¿ĩåİ»
    -0.09
    adol
    -0.08
    æ¼ı
    -0.08
    èIJ¤
    -0.08
    夹
    -0.08
    éģİåİ»
    -0.08
    POSITIVE LOGITS
     certainty
    0.13
    ertainty
    0.12
     sure
    0.11
     sûr
    0.11
    åĿļå®ļ
    0.10
    èĤ¯å®ļæĺ¯
    0.10
     assured
    0.10
    ç¬ĥ
    0.10
    ä¸Ńæĸ°
    0.10
    ç¡®
    0.09
    Activations Density 0.062%

    No Known Activations