© 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. 15193
    Prev
    Next
    INDEX
    Explanations

    terminate, cancel, abort, stop, refund, close

    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
    éľĦ
    -0.22
    æ´ij
    -0.19
    é¢ł
    -0.19
     اÙĦØ°
    -0.19
    ecast
    -0.19
    ç¼®
    -0.18
     pand
    -0.18
    åĨįéĢł
    -0.18
    注è§Ĩ
    -0.18
    lin
    -0.18
    POSITIVE LOGITS
    éĢĢåĩº
    0.34
    ç»ĪæŃ¢
    0.33
    éĢĢ
    0.32
    æĴ¤
    0.32
    éĢĢè´§
    0.31
    .terminate
    0.31
     terminate
    0.30
     termination
    0.29
     terminating
    0.29
     cancel
    0.28
    Activations Density 0.189%

    No Known Activations