© 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. 14392
    Prev
    Next
    INDEX
    Explanations

    Initial letters stand alone

    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
     ÐŁÐ¾Ñĩем
    -0.19
    ä¸İåIJ¦
    -0.18
    formatter
    -0.18
    ilty
    -0.18
    Strict
    -0.17
     strict
    -0.17
     spare
    -0.17
    _every
    -0.17
     every
    -0.17
    æ¯ıä¸Ģ天
    -0.17
    POSITIVE LOGITS
    à¹Į
    0.18
     spep
    0.18
     creampie
    0.18
     занима
    0.18
     Fra
    0.17
     bigot
    0.17
    (coder
    0.17
    _HW
    0.17
    PointerException
    0.17
    ï¸ı
    0.17
    Activations Density 0.117%

    No Known Activations