© 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. 61892
    Prev
    Next
    INDEX
    Explanations

    hours of activity

    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
    asers
    -0.09
    .resp
    -0.09
    MP
    -0.09
    ä¸Ģè·¯
    -0.09
    æĪª
    -0.08
    amp
    -0.08
    两项
    -0.08
    åįĬå¹´
    -0.08
    åħ¨æ°ij
    -0.08
    обÑĢаз
    -0.08
    POSITIVE LOGITS
    æĹłæķ°
    0.12
     countless
    0.12
    æĹłæķ°æ¬¡
    0.11
    åĩĿèģļ
    0.09
    äºĨå¤ļå°ij
    0.09
    æŀ¢çº½
    0.09
     hours
    0.08
     Relief
    0.08
    agna
    0.08
    acher
    0.08
    Activations Density 0.037%

    No Known Activations