© 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. 50802
    Prev
    Next
    INDEX
    Explanations

    goals and targets

    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
    峤
    -0.11
    æīĭä¸ŃçļĦ
    -0.09
    pcs
    -0.09
    åįģ足
    -0.09
    åľ°äº§
    -0.08
    误解
    -0.08
    åī²
    -0.08
    _extractor
    -0.08
    obot
    -0.08
    æ»Ķ
    -0.08
    POSITIVE LOGITS
    缮æłĩ
    0.17
    å¥ĭæĸĹ缮æłĩ
    0.14
    缮æ¨Ļ
    0.14
    çļĦ缮æłĩ
    0.12
    ç»Ļèĩªå·±
    0.12
     goals
    0.11
     target
    0.11
    targets
    0.11
     targets
    0.11
     Target
    0.11
    Activations Density 0.027%

    No Known Activations