© 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. Qwen3-14B
    5. 20-RESID-BATCHTOPK-65K__L0-80
    7. 13880
    Prev
    Next
    INDEX
    Explanations

    inter-

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    adamkarvonen/qwen3-14b-saes/saes_Qwen_Qwen3-14B_batch_top_k/resid_post_layer_20
    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
    ãģ»ãģĨ
    -0.20
    -0.19
    /her
    -0.19
    skin
    -0.17
     latter
    -0.16
    ê¼´
    -0.16
    busters
    -0.15
    ¿
    -0.15
     trÃŃ
    -0.15
    etime
    -0.15
    POSITIVE LOGITS
    lude
    0.20
     ÑģебÑı
    0.18
    clude
    0.18
    æĢ§å¼º
    0.18
    iors
    0.17
    esa
    0.17
    eing
    0.17
    halb
    0.17
    è°į
    0.17
    æ¼ī
    0.16
    Activations Density 0.066%

    No Known Activations