© 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. 15539
    Prev
    Next
    INDEX
    Explanations

    ensure, prevent, avoid

    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
    aal
    -0.20
    æ·±åĪ»çļĦ
    -0.19
    	RTLR
    -0.19
    éĢ¡
    -0.18
    èµ°åIJij
    -0.18
     Vig
    -0.18
    iked
    -0.18
    æ·±åĪ»
    -0.17
     Jordan
    -0.17
     Wal
    -0.17
    POSITIVE LOGITS
    éĺ²æŃ¢
    0.35
    éģ¿åħį
    0.31
    为äºĨéģ¿åħį
    0.29
     avoid
    0.28
    ç¡®ä¿Ŀ
    0.27
    以éĺ²
    0.27
    以åħį
    0.26
     ensure
    0.24
     prevent
    0.24
    Avoid
    0.24
    Activations Density 0.140%

    No Known Activations