© 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. 19771
    Prev
    Next
    INDEX
    Explanations

    telling or distinguishing information

    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
    íĥĪ
    -0.11
    oose
    -0.10
    纪å½ķ
    -0.10
    oves
    -0.09
    emie
    -0.09
    ÑģÑĥ
    -0.09
    edom
    -0.09
    éİ
    -0.09
    zie
    -0.09
    çĵľ
    -0.09
    POSITIVE LOGITS
    æĺ¯åľ¨
    0.09
    æľīå¤ļå°ij
    0.09
     bahwa
    0.09
    æ°ĽåĽ´
    0.09
    纵åIJij
    0.09
    æĺ¯å¯¹
    0.08
    ine
    0.08
    ior
    0.08
    xing
    0.08
    wert
    0.08
    Activations Density 0.011%

    No Known Activations