© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-14B
    5. 20-RESID-BATCHTOPK-65K__L0-80
    7. 10666
    Prev
    Next
    INDEX
    Explanations

    square wave, D, classes, brackets, root

    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
    esser
    -0.18
    entication
    -0.16
    ual
    -0.16
    andler
    -0.15
    osph
    -0.15
    åĭĥåĭĥ
    -0.15
     ÑģебÑı
    -0.15
    лем
    -0.15
    èħĪ
    -0.15
    auge
    -0.15
    POSITIVE LOGITS
    -minded
    0.23
     minded
    0.22
    åłĻ
    0.19
    ATURE
    0.19
    ever
    0.19
    mente
    0.19
    holders
    0.19
    à¹Ĩ
    0.18
    -ended
    0.17
    pants
    0.17
    Activations Density 0.040%

    No Known Activations