© 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. 47755
    Prev
    Next
    INDEX
    Explanations

    length of stay or duration

    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
    iman
    -0.10
    çĤľ
    -0.09
    forme
    -0.09
    jie
    -0.09
     Bac
    -0.09
     sortOrder
    -0.09
    æ³ī
    -0.09
    Ñģад
    -0.09
    é»ĺé»ĺåľ°
    -0.09
    çݯå¢ĥä¸Ń
    -0.09
    POSITIVE LOGITS
    oose
    0.10
     duration
    0.09
    éķ¿çŁŃ
    0.09
     staying
    0.09
     stay
    0.09
     пÑĢоÑģ
    0.09
    ä¼Ĺ
    0.08
    æģ¢å¤į
    0.08
     Duration
    0.08
     durations
    0.08
    Activations Density 0.310%

    No Known Activations