- GRB found by the ground process (failed event data). - The event data are only available for part of the burst duration. - DETECTION mask with pcodethresh = 0.01 is used for everything except spectral analysis. - Spectral analysis is not available. - Burst durations are found using quad-rate data from T0_GCN-70 s to T0_GCN+100 s - Trigger ID is different than the one reported in the GCN (617313) - The low signal-to-noise ratio could be due to the duration (found by the standard pipeline using the non-maskweighted quad-rate data) is not correct. We can get a much higher signal-to-noise ratio (11 sigma) with manually chosen burst interval from T0+1.0 s to T0+2.0 s in 25-350 keV. However, due to the hard-to-quantify false detection rate from manual trails, we do not report this result. - Tentative detection.