© 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. 4960
    Prev
    Next
    INDEX
    Explanations

    will until must wait check

    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
    æŃ¹
    -0.10
    Ñģли
    -0.09
    çģ¾
    -0.09
    人éĥ½
    -0.08
    isci
    -0.08
    å¾Ģäºĭ
    -0.08
     ÙĥØ«ÙĬر
    -0.08
    /sbin
    -0.08
    alter
    -0.08
    hi
    -0.08
    POSITIVE LOGITS
    åĽŀåİ»
    0.16
    ä¸ĭ次
    0.16
    æľīå¾ħ
    0.14
    åIJİç»Ń
    0.14
    æĹ¥åIJİ
    0.12
    åı¦è¡Į
    0.12
    ä»ĬåIJİ
    0.12
    åĨįè§ģ
    0.11
    æļĤ
    0.11
     Continued
    0.11
    Activations Density 0.053%

    No Known Activations