© 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. 33416
    Prev
    Next
    INDEX
    Explanations

    predators and prey

    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
    çĩĥæĸĻ
    -0.09
    ä¸Ģè¶Ł
    -0.08
    çľ·
    -0.08
    æĥ³æĥ³
    -0.08
    vecs
    -0.08
    ync
    -0.08
    CTR
    -0.08
    æ³ķå¾ĭè§Ħå®ļ
    -0.08
    .disc
    -0.08
     disc
    -0.08
    POSITIVE LOGITS
    Bitcoin
    0.10
    ç¬ı
    0.09
    éĿĴå°ijå¹´
    0.09
    ä¸Ńå°ıåѦçĶŁ
    0.09
    å¸Ĩ
    0.09
    åIJŀ
    0.08
    åŁķ
    0.08
    invert
    0.08
    å°ıåŃ©
    0.08
    ä¸Ģåı£
    0.08
    Activations Density 0.196%

    No Known Activations