© 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. 21999
    Prev
    Next
    INDEX
    Explanations

    long term or longer duration

    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
    çĭĢ
    -0.10
    产éĩı
    -0.09
    è§ĦèĮĥ
    -0.08
     Bail
    -0.08
     pearl
    -0.08
    å¿«æĬ¥
    -0.08
    åħĪæĺ¯
    -0.08
    ä¿Ŀåģ¥åĵģ
    -0.08
    ени
    -0.08
    äºĽè®¸
    -0.08
    POSITIVE LOGITS
    è¾ĥéķ¿
    0.15
     Longer
    0.12
    -long
    0.11
     longer
    0.11
    éķ¿æľŁ
    0.11
    agos
    0.10
     ì§
    0.10
     dÃłi
    0.10
    éķ¿æķĪ
    0.10
    æľŁéĻIJ
    0.10
    Activations Density 0.084%

    No Known Activations