© 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. 31225
    Prev
    Next
    INDEX
    Explanations

    classification and classes

    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
    çīĩ段
    -0.11
    å¼ĺ
    -0.10
    åģĮ
    -0.10
    æ£ļ
    -0.10
    лев
    -0.09
    avid
    -0.09
    æ´»
    -0.09
    гÑĢан
    -0.09
    éľ¾
    -0.09
    æĤ¨çļĦåŃ©åŃIJ
    -0.09
    POSITIVE LOGITS
     Keller
    0.10
    pret
    0.08
    ummies
    0.08
    æī¶
    0.08
    features
    0.08
     Votes
    0.08
    eller
    0.08
    åīĬ
    0.08
    ån
    0.08
    _BINARY
    0.07
    Activations Density 0.303%

    No Known Activations