© 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. 11381
    Prev
    Next
    INDEX
    Explanations

    past tense verbs followed by context

    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
     itself
    -0.12
    ä¼ī
    -0.10
    æľ¬èº«
    -0.10
    å¤Ļ
    -0.10
    istar
    -0.09
    igram
    -0.09
    éĩįéĩı
    -0.08
     Oliv
    -0.08
    oral
    -0.08
    èIJ½åħ¥
    -0.08
    POSITIVE LOGITS
    æīĭä¸ŃçļĦ
    0.10
    宪
    0.09
    æĭ¥
    0.09
    æľīç§į
    0.09
    èĤļ
    0.09
    çݳ
    0.09
    ä»Ļå¢ĥ
    0.09
     дÑĥÑħ
    0.09
    æ¥
    0.08
    溢
    0.08
    Activations Density 0.193%

    No Known Activations