© 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. 22831
    Prev
    Next
    INDEX
    Explanations

    ingredients and measurements

    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
    è¾ķ
    -0.09
    俯
    -0.09
    åĿ»
    -0.08
    Substring
    -0.08
    uais
    -0.08
    dür
    -0.08
    vere
    -0.08
    illet
    -0.08
    æĤĮ
    -0.08
     Subcommittee
    -0.08
    POSITIVE LOGITS
    ç¢
    0.09
    weak
    0.09
    çĪ¾
    0.08
     King
    0.08
     hypo
    0.08
     skeletons
    0.08
    ìĶ©
    0.08
    èĩªèº«çļĦ
    0.08
    å¹¢
    0.08
    rib
    0.08
    Activations Density 0.013%

    No Known Activations