© 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. 17023
    Prev
    Next
    INDEX
    Explanations

    game scores

    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
    ฯ
    -0.09
    los
    -0.08
    enate
    -0.08
    arov
    -0.08
    ä¸įä¸ĭ
    -0.08
    çªĿ
    -0.07
    /manual
    -0.07
    e
    -0.07
    äºĨä¸Ģ个
    -0.07
    代表大ä¼ļ
    -0.07
    POSITIVE LOGITS
    /of
    0.10
    åĪ«çļĦ
    0.10
    éķ¿çĽ¸
    0.10
    åij¼ãģ°
    0.09
    esser
    0.09
    edly
    0.09
    é¢ĨåħĪçļĦ
    0.09
    说æĺ¯
    0.09
    çıŃ
    0.09
    /from
    0.09
    Activations Density 0.014%

    No Known Activations