© 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. 13287
    Prev
    Next
    INDEX
    Explanations

    word parts forming names/words

    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
    讣
    -0.21
    èĪį
    -0.20
    é²ł
    -0.20
    çĸijéļ¾
    -0.20
    çłļ
    -0.20
    éĻĮ
    -0.20
     cra
    -0.20
    å§Ĭ
    -0.19
    äºĨä¸Ģåı£æ°Ķ
    -0.19
     mutually
    -0.19
    POSITIVE LOGITS
    ç½ļ
    0.35
     punishment
    0.35
    æĥ©ç½ļ
    0.33
    å¤Ħç½ļ
    0.33
    ç½°
    0.32
    责令
    0.30
    à¹Ĥà¸Ĺษ
    0.26
     phạt
    0.26
     punishments
    0.26
    æĥ©æĪĴ
    0.25
    Activations Density 0.029%

    No Known Activations