© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 36842
    Prev
    Next
    INDEX
    Explanations

    single post, array, instance, app, linked

    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
     scre
    -0.09
    .mj
    -0.08
    antan
    -0.08
    Encoded
    -0.08
    å¾ĹçĽĬ
    -0.08
    ãģįãģŁ
    -0.08
     servic
    -0.08
    alog
    -0.08
    建ç«ĭåģ¥åħ¨
    -0.07
    omat
    -0.07
    POSITIVE LOGITS
    isci
    0.10
    ä¸Ģ举
    0.10
    /single
    0.10
    .single
    0.09
    åĹ¡
    0.09
    绫
    0.09
    ä¸įçIJĨ
    0.09
     célib
    0.09
    ä¸į说
    0.09
    åij
    0.08
    Activations Density 0.010%

    No Known Activations