© 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. 21775
    Prev
    Next
    INDEX
    Explanations

    reporting opinions, feelings, or descriptions

    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
    è·³
    -0.10
    顺åĬ¿
    -0.09
    UME
    -0.09
    andom
    -0.09
    æĵĺ
    -0.09
    导
    -0.09
    è¿İ
    -0.09
     СейÑĩаÑģ
    -0.08
    hop
    -0.08
     chÃło
    -0.08
    POSITIVE LOGITS
    æĻ®éģį
    0.12
    åıįé¦Ī
    0.12
     оÑĤзÑĭв
    0.12
    满æĦı度
    0.11
    è¯Ħä»·
    0.11
    ä½ĵéªĮ
    0.10
     feedback
    0.10
    ä¸Ģèĩ´
    0.10
    ä¸Ģèĩ´å¥½è¯Ħ
    0.10
    称èµŀ
    0.09
    Activations Density 0.055%

    No Known Activations