/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 0	score: 0.7286
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 1	score: 0.7611
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 2	score: 0.5908
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 3	score: 0.7780
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 4	score: 0.4624
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 5	score: 0.7926
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 6	score: 0.6980
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 7	score: 0.4432
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 8	score: 0.6825
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 9	score: 0.7898
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 10	score: 0.6102
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 11	score: 0.6014
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 12	score: 0.6535
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 13	score: 0.6479
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 14	score: 0.3338
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 15	score: 0.2711
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 16	score: 0.6196
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 17	score: 0.8722
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 18	score: 0.7435
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 19	score: 0.8331
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 20	score: 0.8088
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 21	score: 0.5982
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 22	score: 0.3944
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 23	score: 0.5955
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 24	score: 0.8681
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 25	score: 0.8675
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 26	score: 0.7839
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 27	score: 0.6261
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 28	score: 0.4301
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 29	score: 0.6691
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 30	score: 0.5108
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 31	score: 0.4097
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 32	score: 0.4866
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 33	score: 0.6570
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 34	score: 0.7765
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 35	score: 0.9077
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 36	score: 0.4161
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 37	score: 0.6168
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 38	score: 0.6386
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 39	score: 0.5475
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 40	score: 0.5508
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 41	score: 0.9097
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 42	score: 0.8371
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 43	score: 0.7882
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 44	score: 0.6504
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 45	score: 0.6707
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 46	score: 0.5610
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 47	score: 0.6244
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 48	score: 0.5506
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 49	score: 0.7337
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 50	score: 0.8026
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 51	score: 0.4762
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 52	score: 0.7794
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 53	score: 0.6809
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 54	score: 0.7146
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 55	score: 0.6792
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 56	score: 0.7567
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 57	score: 0.3578
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 58	score: 0.6684
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 59	score: 0.8507
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 60	score: 0.6236
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 61	score: 0.8010
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 62	score: 0.7271
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 63	score: 0.7046
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 64	score: 0.7659
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 65	score: 0.6842
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 66	score: 0.5975
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 67	score: 0.6728
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 68	score: 0.9193
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 69	score: 0.6286
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 70	score: 0.7185
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 71	score: 0.2623
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 72	score: 0.4233
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 73	score: 0.6910
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 74	score: 0.5538
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 75	score: 0.7563
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 76	score: 0.5694
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 77	score: 0.6385
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 78	score: 0.8678
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 79	score: 0.3202
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 80	score: 0.8053
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 81	score: 0.8103
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 82	score: 0.3838
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 83	score: 0.6964
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 84	score: 0.4064
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 85	score: 0.8378
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 86	score: 0.6273
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 87	score: 0.7114
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 88	score: 0.7035
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 89	score: 0.5757
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 90	score: 0.5151
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 91	score: 0.3312
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 92	score: 0.3806
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 93	score: 0.6300
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 94	score: 0.2541
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 95	score: 0.8020
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 96	score: 0.4326
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 97	score: 0.6086
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 98	score: 0.7415
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 99	score: 0.7600
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 100	score: 0.6790
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 101	score: 0.7938
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 102	score: 0.7993
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 103	score: 0.6895
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 104	score: 0.6918
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 105	score: 0.5911
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 106	score: 0.7845
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 107	score: 0.7199
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 108	score: 0.6004
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 109	score: 0.5612
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 110	score: 0.6721
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 111	score: 0.7038
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 112	score: 0.7459
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 113	score: 0.5359
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 114	score: 0.5549
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 115	score: 0.6687
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 116	score: 0.3794
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 117	score: 0.8022
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 118	score: 0.6641
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 119	score: 0.6383
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 120	score: 0.6818
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 121	score: 0.7237
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 122	score: 0.8909
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 123	score: 0.6521
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 124	score: 0.8508
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 125	score: 0.4892
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 126	score: 0.8172
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 127	score: 0.6701
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 128	score: 0.7820
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 129	score: 0.7253
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 130	score: 0.5825
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 131	score: 0.5885
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 132	score: 0.6456
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 133	score: 0.6890
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 134	score: 0.6104
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 135	score: 0.6074
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 136	score: 0.5304
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 137	score: 0.6460
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 138	score: 0.6886
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 139	score: 0.4530
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 140	score: 0.6686
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 141	score: 0.7201
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 142	score: 0.6734
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 143	score: 0.6411
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 144	score: 0.6399
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 145	score: 0.7731
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 146	score: 0.7539
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 147	score: 0.2443
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 148	score: 0.6326
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 149	score: 0.6712
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 150	score: 0.7565
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 151	score: 0.6410
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 152	score: 0.7111
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 153	score: 0.3337
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 154	score: 0.4831
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 155	score: 0.6913
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 156	score: 0.6673
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 157	score: 0.6111
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 158	score: 0.8020
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 159	score: 0.6657
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 160	score: 0.6465
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 161	score: 0.3399
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 162	score: 0.3958
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 163	score: 0.6609
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 164	score: 0.7102
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 165	score: 0.5949
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 166	score: 0.6198
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 167	score: 0.6730
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 168	score: 0.7670
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 169	score: 0.4004
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 170	score: 0.7461
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 171	score: 0.8221
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 172	score: 0.7790
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 173	score: 0.6247
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 174	score: 0.7095
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 175	score: 0.8020
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 176	score: 0.7244
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 177	score: 0.6314
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 178	score: 0.4374
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 179	score: 0.8045
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 180	score: 0.7221
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 181	score: 0.8620
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 182	score: 0.7374
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 183	score: 0.8384
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 184	score: 0.7628
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 185	score: 0.5859
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 186	score: 0.8244
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 187	score: 0.5669
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 188	score: 0.8143
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 189	score: 0.6895
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 190	score: 0.6418
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 191	score: 0.7699
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 192	score: 0.5506
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 193	score: 0.6729
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 194	score: 0.8101
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 195	score: 0.6700
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 196	score: 0.6029
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 197	score: 0.3765
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 198	score: 0.6702
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 199	score: 0.5981
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 200	score: 0.7912
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 201	score: 0.7662
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 202	score: 0.6953
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 203	score: 0.7179
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 204	score: 0.8093
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 205	score: 0.7327
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 206	score: 0.8418
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 207	score: 0.7669
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 208	score: 0.7398
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 209	score: 0.7409
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 210	score: 0.8142
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 211	score: 0.6990
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 212	score: 0.8394
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 213	score: 0.9082
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 214	score: 0.6416
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 215	score: 0.5840
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 216	score: 0.7692
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 217	score: 0.7855
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 218	score: 0.7163
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 219	score: 0.5727
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 220	score: 0.9085
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 221	score: 0.6710
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 222	score: 0.4173
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 223	score: 0.7981
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 224	score: 0.5831
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 225	score: 0.7892
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 226	score: 0.6287
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 227	score: 0.3755
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 228	score: 0.7873
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 229	score: 0.7075
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 230	score: 0.8669
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 231	score: 0.7095
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 232	score: 0.6397
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 233	score: 0.6791
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 234	score: 0.4210
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 235	score: 0.6798
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 236	score: 0.8540
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 237	score: 0.4015
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 238	score: 0.7461
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 239	score: 0.5777
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 240	score: 0.6017
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 241	score: 0.6191
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 242	score: 0.7146
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 243	score: 0.6071
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 244	score: 0.7511
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 245	score: 0.7633
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 246	score: 0.7956
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 247	score: 0.3229
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 248	score: 0.5654
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 249	score: 0.1994
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 250	score: 0.8200
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 251	score: 0.8906
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 252	score: 0.6773
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 253	score: 0.7140
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 254	score: 0.7669
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 255	score: 0.7505
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 256	score: 0.3556
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 257	score: 0.6259
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 258	score: 0.6571
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 259	score: 0.7284
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 260	score: 0.6454
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 261	score: 0.5182
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 262	score: 0.8264
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 263	score: 0.5420
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 264	score: 0.7123
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 265	score: 0.4570
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 266	score: 0.7797
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 267	score: 0.5873
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 268	score: 0.7585
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 269	score: 0.7343
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 270	score: 0.4729
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 271	score: 0.5522
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 272	score: 0.7915
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 273	score: 0.6435
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 274	score: 0.7328
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 275	score: 0.3218
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 276	score: 0.7177
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 277	score: 0.7743
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 278	score: 0.6233
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 279	score: 0.9263
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 280	score: 0.7873
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 281	score: 0.6242
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 282	score: 0.7542
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 283	score: 0.6555
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 284	score: 0.8475
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 285	score: 0.8959
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 286	score: 0.7815
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 287	score: 0.6093
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 288	score: 0.5568
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 289	score: 0.6756
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 290	score: 0.4463
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 291	score: 0.7483
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 292	score: 0.7614
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 293	score: 0.6530
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 294	score: 0.6196
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 295	score: 0.6702
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 296	score: 0.7023
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 297	score: 0.6706
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 298	score: 0.8243
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 299	score: 0.4018
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 300	score: 0.7547
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 301	score: 0.8404
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 302	score: 0.6013
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 303	score: 0.8648
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 304	score: 0.7696
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 305	score: 0.6976
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 306	score: 0.8352
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 307	score: 0.8132
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 308	score: 0.7738
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 309	score: 0.7831
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 310	score: 0.6921
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 311	score: 0.5655
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 312	score: 0.7289
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 313	score: 0.6302
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 314	score: 0.5771
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 315	score: 0.7090
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 316	score: 0.7764
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 317	score: 0.7634
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 318	score: 0.5748
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 319	score: 0.6145
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 320	score: 0.6707
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 321	score: 0.6427
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 322	score: 0.7785
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 323	score: 0.8219
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 324	score: 0.5253
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 325	score: 0.7672
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 326	score: 0.7426
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 327	score: 0.6555
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 328	score: 0.3451
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 329	score: 0.7703
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 330	score: 0.6996
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 331	score: 0.7193
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 332	score: 0.5986
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 333	score: 0.7032
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 334	score: 0.8686
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 335	score: 0.2508
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 336	score: 0.4401
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 337	score: 0.4345
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 338	score: 0.2767
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 339	score: 0.4189
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 340	score: 0.6340
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 341	score: 0.6675
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 342	score: 0.3161
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 343	score: 0.4483
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 344	score: 0.7143
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 345	score: 0.3794
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 346	score: 0.8063
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 347	score: 0.7168
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 348	score: 0.3085
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 349	score: 0.5696
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 350	score: 0.6601
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 351	score: 0.6469
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 352	score: 0.4628
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 353	score: 0.6295
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 354	score: 0.4043
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 355	score: 0.6861
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 356	score: 0.6691
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 357	score: 0.6797
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 358	score: 0.6213
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 359	score: 0.2799
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 360	score: 0.4302
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 361	score: 0.7424
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 362	score: 0.5790
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 363	score: 0.4235
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 364	score: 0.4885
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 365	score: 0.4867
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 366	score: 0.3005
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 367	score: 0.7609
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 368	score: 0.4348
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 369	score: 0.5454
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 370	score: 0.5898
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 371	score: 0.3913
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 372	score: 0.6390
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 373	score: 0.5983
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 374	score: 0.6385
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 375	score: 0.7349
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 376	score: 0.5368
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 377	score: 0.6306
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 378	score: 0.5655
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 379	score: 0.6708
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 380	score: 0.6869
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 381	score: 0.5942
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 382	score: 0.3440
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 383	score: 0.5138
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 384	score: 0.6624
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 385	score: 0.4708
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 386	score: 0.4039
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 387	score: 0.7488
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 388	score: 0.5519
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 389	score: 0.7104
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 390	score: 0.5523
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 391	score: 0.4179
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 392	score: 0.7324
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 393	score: 0.7680
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 394	score: 0.5454
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 395	score: 0.8695
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 396	score: 0.6242
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 397	score: 0.6895
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 398	score: 0.2804
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 399	score: 0.7839
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 400	score: 0.7933
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 401	score: 0.6817
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 402	score: 0.6462
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 403	score: 0.7033
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 404	score: 0.5734
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 405	score: 0.4187
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 406	score: 0.8565
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 407	score: 0.7854
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 408	score: 0.7471
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 409	score: 0.7365
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 410	score: 0.5852
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 411	score: 0.8006
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 412	score: 0.5631
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 413	score: 0.3862
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 414	score: 0.6454
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 415	score: 0.7224
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 416	score: 0.6597
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 417	score: 0.7836
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 418	score: 0.8169
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 419	score: 0.6976
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 420	score: 0.7523
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 421	score: 0.8490
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 422	score: 0.7742
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 423	score: 0.7782
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 424	score: 0.3895
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 425	score: 0.7902
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 426	score: 0.8215
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 427	score: 0.6833
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 428	score: 0.7642
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 429	score: 0.7299
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 430	score: 0.7614
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 431	score: 0.3401
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 432	score: 0.6659
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 433	score: 0.4227
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 434	score: 0.6237
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 435	score: 0.2382
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 436	score: 0.6087
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 437	score: 0.7209
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 438	score: 0.7532
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 439	score: 0.5653
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 440	score: 0.6614
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 441	score: 0.6547
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 442	score: 0.5136
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 443	score: 0.7223
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 444	score: 0.6721
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 445	score: 0.6583
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 446	score: 0.6852
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 447	score: 0.3895
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 448	score: 0.7917
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 449	score: 0.6087
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 450	score: 0.5327
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 451	score: 0.5455
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 452	score: 0.7615
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 453	score: 0.6950
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 454	score: 0.6247
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 455	score: 0.5980
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 456	score: 0.6130
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 457	score: 0.6068
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 458	score: 0.2643
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 459	score: 0.7142
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 460	score: 0.6230
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 461	score: 0.6621
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 462	score: 0.3105
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 463	score: 0.6336
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 464	score: 0.7593
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 465	score: 0.7896
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 466	score: 0.7720
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 467	score: 0.4343
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 468	score: 0.4776
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 469	score: 0.3796
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 470	score: 0.3345
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 471	score: 0.7269
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 472	score: 0.2728
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 473	score: 0.3664
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 474	score: 0.7295
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 475	score: 0.6550
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 476	score: 0.5764
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 477	score: 0.6848
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 478	score: 0.7143
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 479	score: 0.4341
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 480	score: 0.3763
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 481	score: 0.4128
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 482	score: 0.7029
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 483	score: 0.6885
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 484	score: 0.7476
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 485	score: 0.7170
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 486	score: 0.8126
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 487	score: 0.6459
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 488	score: 0.6704
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 489	score: 0.6615
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 490	score: 0.2993
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 491	score: 0.7562
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 492	score: 0.7878
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 493	score: 0.6408
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 494	score: 0.7096
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 495	score: 0.6586
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 496	score: 0.5323
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 497	score: 0.4164
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 498	score: 0.6185
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 499	score: 0.4747
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 500	score: 0.4182
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 501	score: 0.7146
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 502	score: 0.7354
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 503	score: 0.5218
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 504	score: 0.6247
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 505	score: 0.5446
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 506	score: 0.6364
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 507	score: 0.5491
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 508	score: 0.6972
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 509	score: 0.4094
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 510	score: 0.6081
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 511	score: 0.5647
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 512	score: 0.6350
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 513	score: 0.5986
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 514	score: 0.7496
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 515	score: 0.6449
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 516	score: 0.7133
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 517	score: 0.7356
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 518	score: 0.6940
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 519	score: 0.7645
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 520	score: 0.6689
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 521	score: 0.6387
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 522	score: 0.3734
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 523	score: 0.7346
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 524	score: 0.7539
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 525	score: 0.6958
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 526	score: 0.6484
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 527	score: 0.5682
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 528	score: 0.7450
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 529	score: 0.3501
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 530	score: 0.1731
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 531	score: 0.5714
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 532	score: 0.6231
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 533	score: 0.4620
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 534	score: 0.3139
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 535	score: 0.5387
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 536	score: 0.4930
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 537	score: 0.5143
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 538	score: 0.4268
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 539	score: 0.6607
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 540	score: 0.8521
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 541	score: 0.6532
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 542	score: 0.6727
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 543	score: 0.7071
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 544	score: 0.7623
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 545	score: 0.6528
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 546	score: 0.7805
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 547	score: 0.5943
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 548	score: 0.3599
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 549	score: 0.4723
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 550	score: 0.6617
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 551	score: 0.7505
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 552	score: 0.7257
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 553	score: 0.6903
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 554	score: 0.8192
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 555	score: 0.2886
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 556	score: 0.3274
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 557	score: 0.7697
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 558	score: 0.6006
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 559	score: 0.5890
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 560	score: 0.4006
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 561	score: 0.5510
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 562	score: 0.7433
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 563	score: 0.6459
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 564	score: 0.6546
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 565	score: 0.4078
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 566	score: 0.7011
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 567	score: 0.6913
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 568	score: 0.6133
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 569	score: 0.6639
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 570	score: 0.7008
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 571	score: 0.6853
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 572	score: 0.5367
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 573	score: 0.6135
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 574	score: 0.5991
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 575	score: 0.8031
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 576	score: 0.3951
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 577	score: 0.6932
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 578	score: 0.3999
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 579	score: 0.5581
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 580	score: 0.4181
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 581	score: 0.5977
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 582	score: 0.5351
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 583	score: 0.6572
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 584	score: 0.4542
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 585	score: 0.6779
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 586	score: 0.6666
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 587	score: 0.3548
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 588	score: 0.5653
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 589	score: 0.7130
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 590	score: 0.6122
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 591	score: 0.7787
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 592	score: 0.5215
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 593	score: 0.8199
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 594	score: 0.7284
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 595	score: 0.6851
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 596	score: 0.6740
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 597	score: 0.6679
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 598	score: 0.3018
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 599	score: 0.8648
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 600	score: 0.8180
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 601	score: 0.8090
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 602	score: 0.4211
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 603	score: 0.5753
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 604	score: 0.4806
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 605	score: 0.6580
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 606	score: 0.6866
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 607	score: 0.6456
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 608	score: 0.5033
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 609	score: 0.6481
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 610	score: 0.7957
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 611	score: 0.7462
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 612	score: 0.3454
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 613	score: 0.7517
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 614	score: 0.6404
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 615	score: 0.5512
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 616	score: 0.5410
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 617	score: 0.6033
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 618	score: 0.4683
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 619	score: 0.5622
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 620	score: 0.7457
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 621	score: 0.5207
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 622	score: 0.7321
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 623	score: 0.7016
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 624	score: 0.8003
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 625	score: 0.8061
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 626	score: 0.3875
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 627	score: 0.9145
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 628	score: 0.7497
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 629	score: 0.6451
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 630	score: 0.6431
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 631	score: 0.7903
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 632	score: 0.8939
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 633	score: 0.8108
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 634	score: 0.9112
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 635	score: 0.5044
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 636	score: 0.8382
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 637	score: 0.6081
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 638	score: 0.5963
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 639	score: 0.3835
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 640	score: 0.4689
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 641	score: 0.7380
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 642	score: 0.6219
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 643	score: 0.6895
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 644	score: 0.4662
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 645	score: 0.3328
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 646	score: 0.6738
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 647	score: 0.3026
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 648	score: 0.4488
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 649	score: 0.5812
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 650	score: 0.5297
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 651	score: 0.4863
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 652	score: 0.4764
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 653	score: 0.5603
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 654	score: 0.6398
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 655	score: 0.7637
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 656	score: 0.5230
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 657	score: 0.6843
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 658	score: 0.5922
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 659	score: 0.4183
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 660	score: 0.5421
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 661	score: 0.7535
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 662	score: 0.7986
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 663	score: 0.7033
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 664	score: 0.7956
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 665	score: 0.7429
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 666	score: 0.7983
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 667	score: 0.6610
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 668	score: 0.6136
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 669	score: 0.5491
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 670	score: 0.7271
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 671	score: 0.5491
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 672	score: 0.7501
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 673	score: 0.5089
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 674	score: 0.8499
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 675	score: 0.3700
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 676	score: 0.7692
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 677	score: 0.6152
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 678	score: 0.7248
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 679	score: 0.7350
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 680	score: 0.4593
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 681	score: 0.8192
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 682	score: 0.7931
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 683	score: 0.7293
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 684	score: 0.5705
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 685	score: 0.6030
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 686	score: 0.3096
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 687	score: 0.5938
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 688	score: 0.7206
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 689	score: 0.6834
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 690	score: 0.8093
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 691	score: 0.6739
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 692	score: 0.7377
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 693	score: 0.3260
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 694	score: 0.7459
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 695	score: 0.7262
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 696	score: 0.6899
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 697	score: 0.7577
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 698	score: 0.6734
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 699	score: 0.7748
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 700	score: 0.7469
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 701	score: 0.8339
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 702	score: 0.8062
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 703	score: 0.6180
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 704	score: 0.5818
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 705	score: 0.3966
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 706	score: 0.5921
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 707	score: 0.6859
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 708	score: 0.6991
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 709	score: 0.8372
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 710	score: 0.8194
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 711	score: 0.3169
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 712	score: 0.8877
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 713	score: 0.7798
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 714	score: 0.6638
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 715	score: 0.7592
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 716	score: 0.7791
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 717	score: 0.5991
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 718	score: 0.7895
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 719	score: 0.6954
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 720	score: 0.8637
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 721	score: 0.6044
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 722	score: 0.7870
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 723	score: 0.6179
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 724	score: 0.8868
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 725	score: 0.6224
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 726	score: 0.7427
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 727	score: 0.7456
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 728	score: 0.6518
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 729	score: 0.7285
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 730	score: 0.3892
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 731	score: 0.7832
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 732	score: 0.7837
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 733	score: 0.8239
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 734	score: 0.3516
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 735	score: 0.8799
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 736	score: 0.6993
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 737	score: 0.7918
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 738	score: 0.5375
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 739	score: 0.7598
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 740	score: 0.7312
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 741	score: 0.7296
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 742	score: 0.6834
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 743	score: 0.7456
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 744	score: 0.7360
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 745	score: 0.4057
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 746	score: 0.6182
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 747	score: 0.6423
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 748	score: 0.6632
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 749	score: 0.6438
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 750	score: 0.8720
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 751	score: 0.8898
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 752	score: 0.8103
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 753	score: 0.7157
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 754	score: 0.6505
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 755	score: 0.7504
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 756	score: 0.7735
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 757	score: 0.6324
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 758	score: 0.7355
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 759	score: 0.6665
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 760	score: 0.7804
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 761	score: 0.5691
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 762	score: 0.5791
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 763	score: 0.6262
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 764	score: 0.6307
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 765	score: 0.6911
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 766	score: 0.6538
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 767	score: 0.5087
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 768	score: 0.9041
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 769	score: 0.8072
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 770	score: 0.7211
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 771	score: 0.7579
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 772	score: 0.2738
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 773	score: 0.6696
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 774	score: 0.8161
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 775	score: 0.6128
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 776	score: 0.6837
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 777	score: 0.4797
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 778	score: 0.7520
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 779	score: 0.6341
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 780	score: 0.7754
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 781	score: 0.7560
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 782	score: 0.5944
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 783	score: 0.6275
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 784	score: 0.6533
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 785	score: 0.6046
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 786	score: 0.5594
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 787	score: 0.6880
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 788	score: 0.6667
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 789	score: 0.6755
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 790	score: 0.4383
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 791	score: 0.6087
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 792	score: 0.6110
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 793	score: 0.6608
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 794	score: 0.6615
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 795	score: 0.7331
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 796	score: 0.7321
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 797	score: 0.7110
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 798	score: 0.7047
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 799	score: 0.6658
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 800	score: 0.3599
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 801	score: 0.7384
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 802	score: 0.4763
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 803	score: 0.3742
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 804	score: 0.3839
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 805	score: 0.4480
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 806	score: 0.6430
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 807	score: 0.7688
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 808	score: 0.6965
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 809	score: 0.7131
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 810	score: 0.7304
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 811	score: 0.7769
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 812	score: 0.8395
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 813	score: 0.8698
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 814	score: 0.7669
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 815	score: 0.6387
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 816	score: 0.6900
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 817	score: 0.2907
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 818	score: 0.6064
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 819	score: 0.4315
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 820	score: 0.6742
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 821	score: 0.6945
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 822	score: 0.6490
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 823	score: 0.6892
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 824	score: 0.3234
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 825	score: 0.5715
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 826	score: 0.6002
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 827	score: 0.7516
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 828	score: 0.7136
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 829	score: 0.6287
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 830	score: 0.4739
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 831	score: 0.7201
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 832	score: 0.4089
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 833	score: 0.4866
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 834	score: 0.3886
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 835	score: 0.7606
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 836	score: 0.4560
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 837	score: 0.7067
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 838	score: 0.3792
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 839	score: 0.6525
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 840	score: 0.7549
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 841	score: 0.2975
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 842	score: 0.4074
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 843	score: 0.7256
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 844	score: 0.4655
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 845	score: 0.2863
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 846	score: 0.5339
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 847	score: 0.7399
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 848	score: 0.6293
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 849	score: 0.5033
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 850	score: 0.5809
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 851	score: 0.7400
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 852	score: 0.7185
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 853	score: 0.4287
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 854	score: 0.6384
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 855	score: 0.5401
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 856	score: 0.3526
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 857	score: 0.7821
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 858	score: 0.3488
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 859	score: 0.6859
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 860	score: 0.7594
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 861	score: 0.7121
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 862	score: 0.6157
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 863	score: 0.6141
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 864	score: 0.5793
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 865	score: 0.3740
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 866	score: 0.6059
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 867	score: 0.7832
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 868	score: 0.6397
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 869	score: 0.7596
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 870	score: 0.7732
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 871	score: 0.7787
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 872	score: 0.2150
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 873	score: 0.4304
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 874	score: 0.7392
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 875	score: 0.3893
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 876	score: 0.5770
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 877	score: 0.6531
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 878	score: 0.7511
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 879	score: 0.6958
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 880	score: 0.5280
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 881	score: 0.6801
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 882	score: 0.7145
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 883	score: 0.7455
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 884	score: 0.6205
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 885	score: 0.5706
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 886	score: 0.7852
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 887	score: 0.5328
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 888	score: 0.6901
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 889	score: 0.5040
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 890	score: 0.7189
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 891	score: 0.6454
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 892	score: 0.7405
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 893	score: 0.3417
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 894	score: 0.7698
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 895	score: 0.7236
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 896	score: 0.7671
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 897	score: 0.6858
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 898	score: 0.8806
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 899	score: 0.6667
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 900	score: 0.6767
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 901	score: 0.7791
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 902	score: 0.6565
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 903	score: 0.7497
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 904	score: 0.8044
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 905	score: 0.4006
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 906	score: 0.8112
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 907	score: 0.6443
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 908	score: 0.7131
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 909	score: 0.6481
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 910	score: 0.4836
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 911	score: 0.6376
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 912	score: 0.7217
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 913	score: 0.5111
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 914	score: 0.5577
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 915	score: 0.6857
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 916	score: 0.7021
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 917	score: 0.2624
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 918	score: 0.5934
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 919	score: 0.7074
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 920	score: 0.7146
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 921	score: 0.6727
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 922	score: 0.5695
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 923	score: 0.8131
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 924	score: 0.7867
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 925	score: 0.2925
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 926	score: 0.5801
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 927	score: 0.7645
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 928	score: 0.7077
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 929	score: 0.5470
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 930	score: 0.8470
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 931	score: 0.6139
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 932	score: 0.6117
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 933	score: 0.6496
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 934	score: 0.5731
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 935	score: 0.6500
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 936	score: 0.4788
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 937	score: 0.8074
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 938	score: 0.6234
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 939	score: 0.7657
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 940	score: 0.8483
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 941	score: 0.9325
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 942	score: 0.8102
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 943	score: 0.5391
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 944	score: 0.6081
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 945	score: 0.7861
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 946	score: 0.6408
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 947	score: 0.6482
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 948	score: 0.5985
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 949	score: 0.6981
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 950	score: 0.7516
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 951	score: 0.6421
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 952	score: 0.2510
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 953	score: 0.5470
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 954	score: 0.6335
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 955	score: 0.7479
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 956	score: 0.7368
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 957	score: 0.6746
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 958	score: 0.7134
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 959	score: 0.4919
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 960	score: 0.5752
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 961	score: 0.6604
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 962	score: 0.6443
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 963	score: 0.4663
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 964	score: 0.7191
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 965	score: 0.2980
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 966	score: 0.5192
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 967	score: 0.5802
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 968	score: 0.3610
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 969	score: 0.5212
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 970	score: 0.7451
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 971	score: 0.4744
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 972	score: 0.6603
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 973	score: 0.6939
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 974	score: 0.6186
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 975	score: 0.7721
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 976	score: 0.6641
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 977	score: 0.6740
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 978	score: 0.6987
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 979	score: 0.6257
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 980	score: 0.6125
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 981	score: 0.7112
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 982	score: 0.6722
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 983	score: 0.6264
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 984	score: 0.3639
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 985	score: 0.8824
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 986	score: 0.7132
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 987	score: 0.3548
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 988	score: 0.6008
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 989	score: 0.6436
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 990	score: 0.2395
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 991	score: 0.5741
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 992	score: 0.6593
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 993	score: 0.7187
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 994	score: 0.7239
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 995	score: 0.6836
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 996	score: 0.7027
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 997	score: 0.7920
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 998	score: 0.7757
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 999	score: 0.2358
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 1000	score: 0.7816
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 1001	score: 0.3855
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 1002	score: 0.5889
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 1003	score: 0.5610
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 1004	score: 0.4274
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 1005	score: 0.6894
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 1006	score: 0.6804
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 1007	score: 0.2958
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 1008	score: 0.4521
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 1009	score: 0.7394
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 1010	score: 0.7064
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	Segment 1011	score: 0.5027
/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-4-max-tokens-512//test-en-mg	score: 0.6379