© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-14B
    5. 20-RESID-BATCHTOPK-65K__L0-80
    7. 12336
    Prev
    Next
    INDEX
    Explanations

    lax, collapse, loopholes

    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
    表çĻ½
    -0.22
    è°
    -0.21
    nested
    -0.21
    åģľå·¥
    -0.20
    matching
    -0.20
     вÑģÑı
    -0.20
    éĶ£
    -0.19
    çıŃåŃIJ
    -0.19
     Shank
    -0.19
    æĬ¥åIJį
    -0.18
    POSITIVE LOGITS
    åĬ©éķ¿
    0.26
     lax
    0.23
    æºĸ
    0.21
    Weak
    0.21
    ä»»
    0.20
    æ´ĭæ´ĭ
    0.20
    宽æĿ¾
    0.19
    acer
    0.19
    nge
    0.19
    rella
    0.19
    Activations Density 0.113%

    No Known Activations