© 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. 34253
    Prev
    Next
    INDEX
    Explanations

    sanctuary or refuge

    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
    ningen
    -0.09
    ä¸į满
    -0.08
    å¾Ĺèµ·
    -0.08
    ert
    -0.08
    rade
    -0.08
    aset
    -0.08
    ssi
    -0.08
    椴
    -0.08
    maf
    -0.08
     Hoch
    -0.08
    POSITIVE LOGITS
    /of
    0.10
    DataExchange
    0.09
    servers
    0.09
    çĩĶ
    0.09
    opies
    0.09
    odem
    0.09
    èĪ¬çļĦ
    0.09
    кли
    0.09
    éű
    0.08
    çħ¦
    0.08
    Activations Density 0.003%

    No Known Activations