© 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. 16448
    Prev
    Next
    INDEX
    Explanations

    in or of time contexts

    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
     BASIS
    -0.20
    è¿ĽæŃ¥
    -0.20
    ocus
    -0.20
     choice
    -0.20
    éĢīæĭ©
    -0.19
     Buk
    -0.19
    åĩºåľŁ
    -0.19
     $?
    -0.19
     dec
    -0.19
    缸åºĶçļĦ
    -0.18
    POSITIVE LOGITS
    åĩ¡æľ¬ç½ij
    0.21
    æľŁéĸĵ
    0.20
    è·¯ç¨ĭ
    0.20
    foreground
    0.19
    æĿijéķĩ
    0.19
    é²Ĥ
    0.18
    иÑı
    0.18
    漫éķ¿çļĦ
    0.18
    æĶ¶çĽĺ
    0.18
    æĿijæĿijæ°ij
    0.18
    Activations Density 0.005%

    No Known Activations