© 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. 18864
    Prev
    Next
    INDEX
    Explanations

    in any event/case

    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
    çĶļèĩ³æĺ¯
    -0.10
    éĻŁ
    -0.10
    recover
    -0.09
    .pm
    -0.09
    zee
    -0.09
    ç¦ģ
    -0.08
    ĶåĽŀ
    -0.08
    _pm
    -0.08
     Recover
    -0.08
    Swift
    -0.08
    POSITIVE LOGITS
     nonetheless
    0.15
    æĹłè®ºå¦Ĥä½ķ
    0.15
    Anyway
    0.14
    è¿ĺæĺ¯è¦ģ
    0.13
    etheless
    0.13
     nevertheless
    0.13
     Nonetheless
    0.12
     anyway
    0.12
    ä¸įè¿ĩ
    0.12
    ä¸įéģİ
    0.12
    Activations Density 0.043%

    No Known Activations