© 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. 49183
    Prev
    Next
    INDEX
    Explanations

    how many times

    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
    æľº
    -0.09
    ä¸ĢæĬ¹
    -0.08
    æĶ¹è£ħ
    -0.08
     пÑĢож
    -0.08
    azi
    -0.08
     mai
    -0.08
    å¦Ĥæŀľæľī
    -0.08
    æĺ¯åIJ¦åŃĺåľ¨
    -0.08
    å¾ĴæŃ¥
    -0.08
    ç©°
    -0.08
    POSITIVE LOGITS
    æŶåĢĻ
    0.11
    licer
    0.11
    æĻĤåĢĻ
    0.09
    æĪijåĴĮ
    0.09
     ways
    0.09
     vez
    0.09
    åľ°æĸ¹
    0.08
    _gateway
    0.08
    æĥħåĨµä¸ĭ
    0.08
    /how
    0.08
    Activations Density 0.037%

    No Known Activations