© 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. 58826
    Prev
    Next
    INDEX
    Explanations

    dare or daring

    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
    ally
    -0.11
    omer
    -0.09
    atis
    -0.09
    å¾Ī满æĦı
    -0.09
    rpc
    -0.09
    race
    -0.09
    æ¯ĽçĹħ
    -0.08
    æľĢ好çļĦ
    -0.08
    rne
    -0.08
    ihad
    -0.08
    POSITIVE LOGITS
    0.11
    ingly
    0.11
    eration
    0.10
    waves
    0.10
    devil
    0.10
    thon
    0.09
    ctors
    0.09
    磾
    0.09
    说æĺ¯
    0.09
    PECT
    0.09
    Activations Density 0.002%

    No Known Activations