© 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. 12118
    Prev
    Next
    INDEX
    Explanations

    names or name parts

    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
    æĶ¾åѦ
    -0.20
    å°ijå¹´
    -0.19
    ift
    -0.19
    绿è±Ĩ
    -0.19
    è¿ĽæŃ¥
    -0.18
     Pine
    -0.18
    è¾ĺ
    -0.18
    plib
    -0.18
    åĬ¿å¤´
    -0.18
    ä½ĵè´´
    -0.18
    POSITIVE LOGITS
     Cer
    0.18
    hea
    0.18
    amac
    0.18
    å¥łåŁº
    0.18
    çĩ¥
    0.18
    peak
    0.17
    alom
    0.17
    rzy
    0.17
    ãģ©ãĤĮ
    0.17
    spl
    0.17
    Activations Density 0.003%

    No Known Activations