© 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. 1665
    Prev
    Next
    INDEX
    Explanations

    end of clause/question punctuation

    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
    eer
    -0.10
    æªĶ
    -0.10
    ted
    -0.09
    ackets
    -0.09
    zig
    -0.08
    een
    -0.08
    typ
    -0.08
    lobber
    -0.08
    eenth
    -0.08
    ifecycle
    -0.08
    POSITIVE LOGITS
    ï¸ı
    0.11
    kus
    0.10
    หาà¸ģ
    0.09
    orary
    0.09
    èĢĮ对äºİ
    0.09
    alley
    0.09
    åīįæīĢ
    0.09
    åĽłä¸ºä»ĸ们
    0.09
    ughter
    0.09
    èĥ½åģļåĪ°
    0.09
    Activations Density 0.063%

    No Known Activations