© 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. 17974
    Prev
    Next
    INDEX
    Explanations

    balance and statements

    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
     padd
    -0.09
    ãĥĻãĤ¹ãĥĪ
    -0.09
    ä¸Ńä»ĭ
    -0.08
    çĪ¬
    -0.08
    åıį转
    -0.08
    éĻIJæŶ
    -0.08
    éĶº
    -0.08
    è¦ģ害
    -0.08
    ðŁı¼
    -0.08
    lement
    -0.08
    POSITIVE LOGITS
    æ¶
    0.11
    ä½Ļé¢Ŀ
    0.10
    ç»ĵ
    0.10
     balance
    0.10
    ä¸Ģè¡Į
    0.10
    åĹĵåŃIJ
    0.09
     balances
    0.09
    å¯Ħ
    0.09
    éĢĥéģ¿
    0.09
    zas
    0.09
    Activations Density 0.092%

    No Known Activations