© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 7450
    Prev
    Next
    INDEX
    Explanations

    shelf life

    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
    éļ³
    -0.09
    ly
    -0.09
    fully
    -0.09
    è¿·æĥij
    -0.08
    ally
    -0.08
    izio
    -0.08
    åľºéĿ¢
    -0.08
    neh
    -0.08
    owo
    -0.08
    ernals
    -0.08
    POSITIVE LOGITS
    ä¸ĬçļĦ
    0.15
    ripsi
    0.10
    tones
    0.10
    -mounted
    0.09
    ridge
    0.09
    wright
    0.09
    KIT
    0.09
     Ñģебе
    0.09
    mates
    0.09
    ìĸ¼
    0.09
    Activations Density 0.009%

    No Known Activations