© 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. 26816
    Prev
    Next
    INDEX
    Explanations

    competition

    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
     завод
    -0.10
    æķ£
    -0.09
    èĪī
    -0.09
    бег
    -0.09
    举
    -0.09
    SEQU
    -0.09
    è¿ľæ´ĭ
    -0.09
    èİ
    -0.09
    IDI
    -0.08
     Pee
    -0.08
    POSITIVE LOGITS
    ç«ŀäºī
    0.25
    çļĦç«ŀäºī
    0.23
    å¸Ĥåľºç«ŀäºī
    0.22
    æ¿ĢçĥĪçļĦ
    0.19
    競çĪŃ
    0.17
     competition
    0.17
    æ¿ĢçĥĪ
    0.16
     competitive
    0.15
    ç«ŀäºī对æīĭ
    0.14
    ç«ŀäºīä¼ĺåĬ¿
    0.13
    Activations Density 0.081%

    No Known Activations