© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 21910
    Prev
    Next
    INDEX
    Explanations

    linked to

    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
    fo
    -0.08
    èħ¹
    -0.08
    ces
    -0.08
     correspondence
    -0.08
     smart
    -0.07
    ance
    -0.07
    æĺ¥å¤ı
    -0.07
    Ñħи
    -0.07
     sex
    -0.07
    ter
    -0.07
    POSITIVE LOGITS
    ç©¿è¿ĩ
    0.11
    ç´§å¯Ĩç»ĵåIJĪ
    0.10
    说çļĦæĺ¯
    0.10
    åı¦ä¸Ģä½į
    0.10
    >Error
    0.10
    æ·±åĪĩ
    0.10
    åįĪé¥Ń
    0.10
    åħľ
    0.10
    åİĦ
    0.10
    HN
    0.09
    Activations Density 0.013%

    No Known Activations