© 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. 13764
    Prev
    Next
    INDEX
    Explanations

    saving, retaining, or reconciling

    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
    çľī头
    -0.21
    ä¸Ģç¬ij
    -0.20
    æĻ¯
    -0.20
    åħ¨éĿ¢æıIJåįĩ
    -0.20
     Finger
    -0.19
     Merry
    -0.19
    妥
    -0.19
    èĩªåľ¨
    -0.19
     Fay
    -0.19
     Elev
    -0.19
    POSITIVE LOGITS
    æĮ½æķij
    0.32
     salvage
    0.30
    æĮ½åĽŀ
    0.26
     salv
    0.25
    å¥ĩ迹
    0.25
    èĩªæķij
    0.25
     chance
    0.24
    侥幸
    0.24
     salvar
    0.23
    æĬ¢æķij
    0.23
    Activations Density 0.181%

    No Known Activations