© 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. 12309
    Prev
    Next
    INDEX
    Explanations

    unrestrained or volatile energy

    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
    çIJĨçĶ±
    -0.20
    ä½ĵçİ°
    -0.19
    æľį
    -0.19
     rationale
    -0.19
    TODO
    -0.19
    ä¸ºè¿Ľä¸ĢæŃ¥
    -0.19
    ä¼ģåĽ¾
    -0.19
     бÑĥд
    -0.19
     contr
    -0.18
    è¿Ļæĺ¯åĽłä¸º
    -0.18
    POSITIVE LOGITS
     alone
    0.23
    èIJ½åĪ°
    0.22
    èIJ½åľ¨
    0.20
     suddenly
    0.19
    aro
    0.19
    ses
    0.19
    è¿Ļä¹Īåģļ
    0.19
    .actions
    0.19
     investments
    0.18
    æīĢ带æĿ¥çļĦ
    0.18
    Activations Density 0.458%

    No Known Activations