© 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. 38374
    Prev
    Next
    INDEX
    Explanations

    coming together

    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
    çĶŁæ¶¯
    -0.09
    kie
    -0.08
    acon
    -0.08
    cept
    -0.08
    ajor
    -0.08
    磨
    -0.08
    ophone
    -0.08
    æ´İ
    -0.08
    è¿ŀ
    -0.08
    ernel
    -0.08
    POSITIVE LOGITS
    欢
    0.10
    zá
    0.09
    åī©ä¸ĭ
    0.09
     CONDITION
    0.09
    BOR
    0.09
    æĭįæĭį
    0.09
    æĸ°çļĦä¸Ģ
    0.08
    鸷
    0.08
    人为
    0.08
    į
    0.08
    Activations Density 0.005%

    No Known Activations