© 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. 55306
    Prev
    Next
    INDEX
    Explanations

    numbers and associated punctuation

    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
    åĪ¸
    -0.09
    reau
    -0.09
    inkel
    -0.09
    èł¹
    -0.09
    bum
    -0.08
    åŁ¹è®ŃçıŃ
    -0.08
    åĪ·
    -0.08
    âĺī
    -0.08
    gart
    -0.08
    bite
    -0.08
    POSITIVE LOGITS
    æĻºåºĵ
    0.09
    éŀį
    0.09
    佶
    0.09
    åIJ¬åIJ¬
    0.09
    以便
    0.08
    äºĨè§£æĽ´å¤ļ
    0.08
    延伸
    0.08
    æĹ¥å¸¸å·¥ä½ľ
    0.08
    æİ¥åıĹéĩĩ访
    0.08
    ç§ijåѦåıijå±ķ
    0.08
    Activations Density 0.005%

    No Known Activations