© 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. 52665
    Prev
    Next
    INDEX
    Explanations

    easy or weak

    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
    èķĪ
    -0.09
    éļIJå½¢
    -0.09
     ÙĪÙĦÙĪ
    -0.09
    çľ¼çķĮ
    -0.09
    ãĤıãģĭãĤīãģªãģĦ
    -0.09
     Pest
    -0.09
    å¥ĭè¿Ľ
    -0.09
    èĨ³é£Ł
    -0.09
    ä½ĵéªĮ
    -0.09
    æŃ£å¼ı
    -0.08
    POSITIVE LOGITS
     easy
    0.17
    æĺĵ
    0.17
    æĺĵäºİ
    0.15
    容æĺĵ
    0.15
    easy
    0.15
    å¾Ī容æĺĵ
    0.15
    æľĢ容æĺĵ
    0.14
     easier
    0.14
     Easy
    0.14
     easiest
    0.14
    Activations Density 0.092%

    No Known Activations