© 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. 19070
    Prev
    Next
    INDEX
    Explanations

    postponed or cancelled events

    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
    aille
    -0.10
     retro
    -0.10
    缴è¾ĸå¸Ĥ
    -0.09
    æĺı
    -0.09
    è¿ĽäºĨ
    -0.09
    è´±
    -0.09
    è·Łéļı
    -0.09
    ettes
    -0.09
    æ·Ļ
    -0.09
    çĪ¬ä¸Ĭ
    -0.08
    POSITIVE LOGITS
    æİ¨è¿Ł
    0.17
    å»¶æľŁ
    0.14
    延
    0.13
     postpon
    0.13
     postponed
    0.12
    延éķ¿
    0.12
     postpone
    0.12
    å»¶è¿Ł
    0.11
     cancell
    0.11
    åıĸæ¶Ī
    0.11
    Activations Density 0.086%

    No Known Activations