© 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. 61512
    Prev
    Next
    INDEX
    Explanations

    low Earth, network seen

    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
    lick
    -0.10
    _currency
    -0.10
    ä¸Ģè·¯
    -0.09
    conde
    -0.09
    åľ°ä¸ĭ
    -0.09
    .Bounds
    -0.09
    Currency
    -0.09
    çĶ¨åĬĽ
    -0.08
    çĹĽ
    -0.08
    ught
    -0.08
    POSITIVE LOGITS
    åĿŀ
    0.12
    ä¸Ģé¢Ĺ
    0.10
    交ä¼ļ
    0.09
    è¿IJ转
    0.09
    олоÑģ
    0.09
    æĿĤ
    0.09
    è¡Ľ
    0.09
    åİŁåĪĻä¸Ĭ
    0.09
    å±ĬæŶ
    0.08
    åį«æĺŁ
    0.08
    Activations Density 0.059%

    No Known Activations