© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 38567
    Prev
    Next
    INDEX
    Explanations

    designed with

    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
    伺
    -0.09
    èħ³
    -0.09
    èĦļ
    -0.09
    å®ŀçī©
    -0.09
    åľ¨ä»ĸçļĦ
    -0.09
    精彩çļĦ
    -0.08
     Doll
    -0.08
    xes
    -0.08
    anging
    -0.08
    å¯Ħ
    -0.08
    POSITIVE LOGITS
    èĶij
    0.11
    çĭİ
    0.10
    ÑĢин
    0.09
    ä¸įä¿¡
    0.08
    ç»
    0.08
    宽
    0.08
    ãĥ£
    0.08
    éĨĴ
    0.08
    .getPort
    0.08
    èѬ
    0.08
    Activations Density 0.048%

    No Known Activations