© 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. 61790
    Prev
    Next
    INDEX
    Explanations

    unexpected

    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
     окол
    -0.09
    rock
    -0.09
    eries
    -0.09
    æĸľ
    -0.08
    ayed
    -0.08
     Dane
    -0.08
    ÑĢен
    -0.08
    пÑĢоÑģ
    -0.08
    acre
    -0.08
    åĪ°åºķ
    -0.08
    POSITIVE LOGITS
     Unexpected
    0.14
    æĦıå¤ĸ
    0.14
    æĥĬåĸľ
    0.14
    Unexpected
    0.14
    unexpected
    0.14
     unexpected
    0.13
     unexpectedly
    0.12
    æĦıæĥ³ä¸įåĪ°
    0.11
    没æľīæĥ³åĪ°
    0.10
     surprising
    0.10
    Activations Density 0.020%

    No Known Activations