© 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. 31302
    Prev
    Next
    INDEX
    Explanations

    review content and ratings

    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
     sched
    -0.10
    åĩºå¤Ħ
    -0.09
    è¡°èĢģ
    -0.09
    èµŀåĬ©
    -0.09
    æłĩå¿ĹæĢ§
    -0.08
    AtIndex
    -0.08
    çģµæķı
    -0.08
    æĸĩä¸Ń
    -0.08
    ITU
    -0.08
    æľ¬æĿ¥å°±
    -0.08
    POSITIVE LOGITS
    è¿Ļæł·è¯´
    0.11
    è¯Ħä»·
    0.11
    éħ·
    0.10
    è¿Ļä¹Ī说
    0.10
    ä¹°çļĦ
    0.10
    æļ
    0.10
    åı£æ°´
    0.10
    诤
    0.10
    è´¡çĮ®
    0.10
    æŀģ端
    0.09
    Activations Density 0.099%

    No Known Activations