© 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. 27205
    Prev
    Next
    INDEX
    Explanations

    am, is, are, 're

    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
    무
    -0.09
    æĶĺ
    -0.08
    åħŃ大
    -0.08
     завод
    -0.08
    å²ij
    -0.08
    å¾Ĺ好
    -0.08
    åīij
    -0.08
    ä¸ĸçķĮç»ıæµİ
    -0.08
    MEA
    -0.08
    IFI
    -0.08
    POSITIVE LOGITS
    ãģ¨ãģįãģ¯
    0.10
    case
    0.09
    éļıæīĭ
    0.09
    Parms
    0.09
    ills
    0.09
    å´ĩ
    0.09
    æķĻè®Ń
    0.08
    primitive
    0.08
     gerade
    0.08
    stud
    0.08
    Activations Density 0.048%

    No Known Activations