diff --git "a/test-mr-en.comet" "b/test-mr-en.comet"
new file mode 100644--- /dev/null
+++ "b/test-mr-en.comet"
@@ -0,0 +1,1013 @@
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 0	score: 0.5443
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 1	score: 0.8244
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 2	score: 0.7622
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 3	score: 0.8059
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 4	score: 0.8256
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 5	score: 0.7536
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 6	score: 0.8853
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 7	score: 0.7451
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 8	score: 0.8676
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 9	score: 0.8037
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 10	score: 0.8299
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 11	score: 0.7526
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 12	score: 0.8976
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 13	score: 0.6681
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 14	score: 0.7803
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 15	score: 0.7467
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 16	score: 0.8862
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 17	score: 0.9352
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 18	score: 0.7515
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 19	score: 0.8510
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 20	score: 0.6614
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 21	score: 0.8938
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 22	score: 0.8588
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 23	score: 0.8385
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 24	score: 0.6802
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 25	score: 0.9267
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 26	score: 0.7571
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 27	score: 0.8225
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 28	score: 0.7869
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 29	score: 0.5371
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 30	score: 0.7037
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 31	score: 0.5960
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 32	score: 0.6200
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 33	score: 0.8936
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 34	score: 0.8834
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 35	score: 0.6051
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 36	score: 0.8054
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 37	score: 0.5487
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 38	score: 0.6560
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 39	score: 0.5440
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 40	score: 0.7850
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 41	score: 0.9333
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 42	score: 0.7732
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 43	score: 0.8953
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 44	score: 0.7926
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 45	score: 0.7264
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 46	score: 0.8920
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 47	score: 0.7547
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 48	score: 0.7974
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 49	score: 0.7660
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 50	score: 0.9554
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 51	score: 0.8393
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 52	score: 0.5695
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 53	score: 0.6953
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 54	score: 0.8023
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 55	score: 0.8638
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 56	score: 0.7085
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 57	score: 0.8431
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 58	score: 0.6500
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 59	score: 0.8072
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 60	score: 0.7306
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 61	score: 0.7615
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 62	score: 0.7748
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 63	score: 0.8806
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 64	score: 0.8316
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 65	score: 0.7783
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 66	score: 0.8014
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 67	score: 0.7843
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 68	score: 0.8992
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 69	score: 0.8510
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 70	score: 0.7159
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 71	score: 0.6675
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 72	score: 0.5875
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 73	score: 0.7687
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 74	score: 0.7564
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 75	score: 0.8169
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 76	score: 0.8101
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 77	score: 0.6505
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 78	score: 0.9784
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 79	score: 0.7960
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 80	score: 0.8119
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 81	score: 0.8954
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 82	score: 0.7713
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 83	score: 0.6958
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 84	score: 0.8216
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 85	score: 0.7099
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 86	score: 0.6398
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 87	score: 0.8057
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 88	score: 0.8248
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 89	score: 0.6083
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 90	score: 0.7597
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 91	score: 0.7249
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 92	score: 0.6671
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 93	score: 0.5877
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 94	score: 0.8974
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 95	score: 0.6290
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 96	score: 0.8584
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 97	score: 0.8355
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 98	score: 0.8792
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 99	score: 0.7221
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 100	score: 0.9422
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 101	score: 0.7237
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 102	score: 0.8859
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 103	score: 0.8538
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 104	score: 0.8460
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 105	score: 0.7146
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 106	score: 0.8201
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 107	score: 0.7985
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 108	score: 0.6979
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 109	score: 0.5406
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 110	score: 0.8431
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 111	score: 0.9354
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 112	score: 0.9312
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 113	score: 0.8151
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 114	score: 0.6786
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 115	score: 0.7418
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 116	score: 0.4899
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 117	score: 0.8001
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 118	score: 0.5958
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 119	score: 0.6898
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 120	score: 0.8142
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 121	score: 0.8307
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 122	score: 0.9128
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 123	score: 0.9231
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 124	score: 0.7815
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 125	score: 0.8429
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 126	score: 0.6889
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 127	score: 0.8316
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 128	score: 0.8771
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 129	score: 0.5869
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 130	score: 0.9055
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 131	score: 0.8449
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 132	score: 0.6934
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 133	score: 0.8180
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 134	score: 0.7336
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 135	score: 0.6995
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 136	score: 0.8497
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 137	score: 0.8177
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 138	score: 0.6408
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 139	score: 0.6930
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 140	score: 0.8684
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 141	score: 0.8529
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 142	score: 0.6991
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 143	score: 0.5077
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 144	score: 0.5948
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 145	score: 0.9025
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 146	score: 0.9083
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 147	score: 0.8697
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 148	score: 0.7499
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 149	score: 0.9004
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 150	score: 0.8031
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 151	score: 0.8197
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 152	score: 0.6846
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 153	score: 0.7290
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 154	score: 0.6968
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 155	score: 0.8754
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 156	score: 0.8279
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 157	score: 0.6130
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 158	score: 0.7713
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 159	score: 0.8839
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 160	score: 0.7920
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 161	score: 0.7454
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 162	score: 0.8074
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 163	score: 0.7742
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 164	score: 0.7652
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 165	score: 0.7702
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 166	score: 0.8403
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 167	score: 0.6192
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 168	score: 0.6991
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 169	score: 0.8747
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 170	score: 0.8696
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 171	score: 0.8352
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 172	score: 0.7203
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 173	score: 0.6445
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 174	score: 0.8148
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 175	score: 0.7882
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 176	score: 0.7962
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 177	score: 0.7326
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 178	score: 0.5156
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 179	score: 0.5930
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 180	score: 0.8639
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 181	score: 0.8961
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 182	score: 0.9062
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 183	score: 0.9209
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 184	score: 0.8159
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 185	score: 0.7710
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 186	score: 0.9236
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 187	score: 0.7417
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 188	score: 0.8037
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 189	score: 0.7790
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 190	score: 0.5859
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 191	score: 0.8074
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 192	score: 0.7954
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 193	score: 0.6105
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 194	score: 0.5439
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 195	score: 0.8483
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 196	score: 0.7911
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 197	score: 0.7478
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 198	score: 0.8487
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 199	score: 0.7867
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 200	score: 0.6969
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 201	score: 0.7340
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 202	score: 0.7862
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 203	score: 0.7442
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 204	score: 0.8322
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 205	score: 0.6608
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 206	score: 0.7632
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 207	score: 0.7774
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 208	score: 0.7377
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 209	score: 0.9237
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 210	score: 0.8005
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 211	score: 0.7043
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 212	score: 0.6381
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 213	score: 0.9013
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 214	score: 0.7869
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 215	score: 0.8541
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 216	score: 0.8708
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 217	score: 0.8551
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 218	score: 0.8398
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 219	score: 0.7846
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 220	score: 0.8831
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 221	score: 0.8713
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 222	score: 0.8347
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 223	score: 0.7291
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 224	score: 0.7753
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 225	score: 0.9016
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 226	score: 0.8714
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 227	score: 0.8344
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 228	score: 0.8755
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 229	score: 0.7839
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 230	score: 0.9715
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 231	score: 0.8268
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 232	score: 0.8517
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 233	score: 0.8113
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 234	score: 0.6115
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 235	score: 0.8207
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 236	score: 0.8580
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 237	score: 0.7665
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 238	score: 0.8804
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 239	score: 0.6702
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 240	score: 0.6821
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 241	score: 0.7518
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 242	score: 0.7271
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 243	score: 0.5668
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 244	score: 0.8308
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 245	score: 0.7686
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 246	score: 0.7358
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 247	score: 0.8043
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 248	score: 0.6753
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 249	score: 0.8282
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 250	score: 0.8664
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 251	score: 0.7408
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 252	score: 0.8923
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 253	score: 0.9527
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 254	score: 0.9039
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 255	score: 0.9037
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 256	score: 0.8515
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 257	score: 0.6863
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 258	score: 0.8217
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 259	score: 0.7678
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 260	score: 0.8170
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 261	score: 0.5278
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 262	score: 0.7310
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 263	score: 0.5992
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 264	score: 0.9156
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 265	score: 0.8545
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 266	score: 0.8243
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 267	score: 0.7428
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 268	score: 0.9344
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 269	score: 0.8694
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 270	score: 0.7150
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 271	score: 0.6747
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 272	score: 0.7705
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 273	score: 0.8023
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 274	score: 0.6925
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 275	score: 0.6354
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 276	score: 0.8403
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 277	score: 0.8670
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 278	score: 0.8688
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 279	score: 0.9258
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 280	score: 0.8076
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 281	score: 0.8060
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 282	score: 0.6968
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 283	score: 0.7018
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 284	score: 0.8982
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 285	score: 0.9300
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 286	score: 0.8590
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 287	score: 0.7695
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 288	score: 0.7914
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 289	score: 0.8312
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 290	score: 0.8597
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 291	score: 0.7486
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 292	score: 0.8564
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 293	score: 0.6747
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 294	score: 0.6265
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 295	score: 0.8948
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 296	score: 0.8653
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 297	score: 0.8907
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 298	score: 0.9007
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 299	score: 0.8542
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 300	score: 0.8841
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 301	score: 0.7340
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 302	score: 0.7971
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 303	score: 0.9231
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 304	score: 0.8251
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 305	score: 0.7621
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 306	score: 0.8228
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 307	score: 0.6857
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 308	score: 0.8913
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 309	score: 0.8430
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 310	score: 0.9263
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 311	score: 0.8735
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 312	score: 0.6228
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 313	score: 0.6760
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 314	score: 0.6290
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 315	score: 0.8983
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 316	score: 0.7301
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 317	score: 0.7155
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 318	score: 0.7635
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 319	score: 0.7906
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 320	score: 0.7360
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 321	score: 0.8178
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 322	score: 0.7666
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 323	score: 0.7916
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 324	score: 0.8034
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 325	score: 0.7925
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 326	score: 0.7876
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 327	score: 0.7868
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 328	score: 0.8705
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 329	score: 0.7342
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 330	score: 0.7243
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 331	score: 0.8113
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 332	score: 0.7986
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 333	score: 0.8374
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 334	score: 0.9239
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 335	score: 0.8749
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 336	score: 0.7999
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 337	score: 0.8332
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 338	score: 0.7181
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 339	score: 0.7744
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 340	score: 0.7719
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 341	score: 0.5732
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 342	score: 0.5488
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 343	score: 0.6352
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 344	score: 0.8938
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 345	score: 0.7780
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 346	score: 0.8425
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 347	score: 0.8803
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 348	score: 0.5932
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 349	score: 0.6112
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 350	score: 0.8631
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 351	score: 0.7365
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 352	score: 0.8411
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 353	score: 0.9698
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 354	score: 0.5790
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 355	score: 0.8106
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 356	score: 0.8586
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 357	score: 0.5954
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 358	score: 0.7596
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 359	score: 0.7368
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 360	score: 0.7949
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 361	score: 0.6969
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 362	score: 0.6148
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 363	score: 0.5231
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 364	score: 0.8201
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 365	score: 0.7179
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 366	score: 0.5645
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 367	score: 0.7011
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 368	score: 0.6991
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 369	score: 0.8049
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 370	score: 0.8136
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 371	score: 0.6995
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 372	score: 0.6964
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 373	score: 0.8268
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 374	score: 0.7378
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 375	score: 0.8385
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 376	score: 0.8846
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 377	score: 0.8713
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 378	score: 0.7109
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 379	score: 0.8343
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 380	score: 0.9278
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 381	score: 0.8674
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 382	score: 0.7116
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 383	score: 0.7301
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 384	score: 0.8488
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 385	score: 0.7565
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 386	score: 0.6115
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 387	score: 0.8016
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 388	score: 0.8012
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 389	score: 0.8270
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 390	score: 0.8760
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 391	score: 0.6514
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 392	score: 0.7437
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 393	score: 0.8670
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 394	score: 0.7602
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 395	score: 0.6118
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 396	score: 0.7673
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 397	score: 0.7078
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 398	score: 0.7082
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 399	score: 0.8474
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 400	score: 0.8656
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 401	score: 0.7809
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 402	score: 0.7106
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 403	score: 0.7800
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 404	score: 0.6550
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 405	score: 0.8571
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 406	score: 0.8648
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 407	score: 0.8089
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 408	score: 0.9338
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 409	score: 0.8772
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 410	score: 0.8362
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 411	score: 0.4305
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 412	score: 0.5696
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 413	score: 0.7277
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 414	score: 0.8312
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 415	score: 0.8492
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 416	score: 0.7658
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 417	score: 0.7673
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 418	score: 0.9178
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 419	score: 0.9429
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 420	score: 0.8527
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 421	score: 0.7545
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 422	score: 0.7083
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 423	score: 0.7868
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 424	score: 0.8725
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 425	score: 0.8856
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 426	score: 0.8979
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 427	score: 0.8429
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 428	score: 0.7778
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 429	score: 0.8044
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 430	score: 0.8557
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 431	score: 0.6586
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 432	score: 0.5234
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 433	score: 0.6549
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 434	score: 0.7237
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 435	score: 0.6682
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 436	score: 0.5952
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 437	score: 0.9039
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 438	score: 0.8534
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 439	score: 0.5811
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 440	score: 0.8697
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 441	score: 0.8829
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 442	score: 0.7904
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 443	score: 0.7587
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 444	score: 0.7657
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 445	score: 0.8011
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 446	score: 0.8992
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 447	score: 0.8968
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 448	score: 0.9026
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 449	score: 0.7619
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 450	score: 0.8596
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 451	score: 0.6869
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 452	score: 0.7541
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 453	score: 0.6248
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 454	score: 0.8091
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 455	score: 0.7660
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 456	score: 0.6374
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 457	score: 0.6343
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 458	score: 0.7048
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 459	score: 0.7241
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 460	score: 0.5999
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 461	score: 0.7109
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 462	score: 0.7773
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 463	score: 0.9069
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 464	score: 0.8444
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 465	score: 0.8438
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 466	score: 0.8375
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 467	score: 0.6256
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 468	score: 0.6260
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 469	score: 0.7790
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 470	score: 0.7144
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 471	score: 0.7442
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 472	score: 0.6732
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 473	score: 0.6889
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 474	score: 0.6616
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 475	score: 0.8241
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 476	score: 0.6424
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 477	score: 0.8283
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 478	score: 0.8084
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 479	score: 0.6296
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 480	score: 0.8903
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 481	score: 0.6332
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 482	score: 0.8289
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 483	score: 0.9146
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 484	score: 0.8406
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 485	score: 0.7901
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 486	score: 0.7692
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 487	score: 0.7107
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 488	score: 0.8010
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 489	score: 0.6935
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 490	score: 0.6142
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 491	score: 0.5846
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 492	score: 0.9240
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 493	score: 0.8788
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 494	score: 0.6255
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 495	score: 0.8209
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 496	score: 0.8769
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 497	score: 0.6845
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 498	score: 0.7467
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 499	score: 0.7419
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 500	score: 0.7843
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 501	score: 0.6837
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 502	score: 0.7372
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 503	score: 0.7796
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 504	score: 0.7968
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 505	score: 0.6705
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 506	score: 0.7820
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 507	score: 0.6255
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 508	score: 0.7655
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 509	score: 0.7753
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 510	score: 0.7484
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 511	score: 0.6903
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 512	score: 0.8173
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 513	score: 0.8267
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 514	score: 0.9161
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 515	score: 0.9020
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 516	score: 0.7782
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 517	score: 0.7065
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 518	score: 0.6825
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 519	score: 0.8546
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 520	score: 0.8711
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 521	score: 0.8429
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 522	score: 0.7501
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 523	score: 0.8332
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 524	score: 0.7902
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 525	score: 0.7200
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 526	score: 0.6870
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 527	score: 0.4889
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 528	score: 0.8019
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 529	score: 0.8877
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 530	score: 0.6828
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 531	score: 0.7337
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 532	score: 0.7305
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 533	score: 0.5357
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 534	score: 0.6973
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 535	score: 0.5789
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 536	score: 0.6030
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 537	score: 0.6993
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 538	score: 0.8621
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 539	score: 0.8595
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 540	score: 0.8757
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 541	score: 0.8784
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 542	score: 0.8345
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 543	score: 0.8296
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 544	score: 0.7951
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 545	score: 0.8314
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 546	score: 0.7693
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 547	score: 0.7743
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 548	score: 0.8412
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 549	score: 0.9003
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 550	score: 0.4842
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 551	score: 0.8564
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 552	score: 0.7487
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 553	score: 0.6480
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 554	score: 0.8387
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 555	score: 0.6668
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 556	score: 0.6119
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 557	score: 0.3796
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 558	score: 0.6784
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 559	score: 0.8007
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 560	score: 0.5164
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 561	score: 0.4599
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 562	score: 0.8666
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 563	score: 0.8645
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 564	score: 0.8011
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 565	score: 0.8835
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 566	score: 0.8147
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 567	score: 0.8517
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 568	score: 0.8037
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 569	score: 0.7440
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 570	score: 0.8723
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 571	score: 0.4985
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 572	score: 0.5104
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 573	score: 0.7258
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 574	score: 0.9233
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 575	score: 0.5495
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 576	score: 0.4685
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 577	score: 0.7002
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 578	score: 0.9072
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 579	score: 0.6874
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 580	score: 0.5278
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 581	score: 0.5025
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 582	score: 0.8387
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 583	score: 0.7894
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 584	score: 0.5259
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 585	score: 0.8127
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 586	score: 0.9005
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 587	score: 0.9192
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 588	score: 0.6964
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 589	score: 0.6220
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 590	score: 0.8092
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 591	score: 0.9086
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 592	score: 0.8123
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 593	score: 0.5823
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 594	score: 0.8646
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 595	score: 0.7415
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 596	score: 0.7053
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 597	score: 0.5479
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 598	score: 0.5030
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 599	score: 0.5649
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 600	score: 0.6068
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 601	score: 0.6521
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 602	score: 0.6162
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 603	score: 0.5319
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 604	score: 0.8861
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 605	score: 0.8706
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 606	score: 0.6727
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 607	score: 0.6620
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 608	score: 0.5560
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 609	score: 0.8098
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 610	score: 0.6789
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 611	score: 0.5417
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 612	score: 0.5880
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 613	score: 0.8577
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 614	score: 0.6065
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 615	score: 0.6174
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 616	score: 0.8775
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 617	score: 0.9346
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 618	score: 0.8307
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 619	score: 0.7298
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 620	score: 0.6716
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 621	score: 0.6480
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 622	score: 0.6648
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 623	score: 0.7381
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 624	score: 0.5875
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 625	score: 0.8648
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 626	score: 0.8985
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 627	score: 0.8810
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 628	score: 0.9096
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 629	score: 0.7565
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 630	score: 0.8386
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 631	score: 0.8781
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 632	score: 0.8962
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 633	score: 0.8404
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 634	score: 0.9723
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 635	score: 0.7262
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 636	score: 0.9003
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 637	score: 0.8644
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 638	score: 0.5559
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 639	score: 0.7390
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 640	score: 0.5574
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 641	score: 0.4928
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 642	score: 0.5360
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 643	score: 0.5943
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 644	score: 0.8465
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 645	score: 0.7887
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 646	score: 0.5906
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 647	score: 0.8348
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 648	score: 0.6344
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 649	score: 0.7999
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 650	score: 0.6661
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 651	score: 0.5389
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 652	score: 0.6186
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 653	score: 0.6466
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 654	score: 0.8387
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 655	score: 0.7466
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 656	score: 0.4782
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 657	score: 0.8813
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 658	score: 0.6943
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 659	score: 0.7258
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 660	score: 0.7650
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 661	score: 0.8263
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 662	score: 0.6240
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 663	score: 0.7443
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 664	score: 0.7704
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 665	score: 0.8034
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 666	score: 0.8045
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 667	score: 0.8821
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 668	score: 0.6590
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 669	score: 0.7934
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 670	score: 0.7204
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 671	score: 0.7139
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 672	score: 0.6743
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 673	score: 0.8199
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 674	score: 0.7583
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 675	score: 0.7367
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 676	score: 0.8565
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 677	score: 0.8303
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 678	score: 0.8487
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 679	score: 0.6948
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 680	score: 0.7898
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 681	score: 0.7012
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 682	score: 0.8869
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 683	score: 0.7866
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 684	score: 0.8378
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 685	score: 0.7717
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 686	score: 0.8443
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 687	score: 0.8163
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 688	score: 0.8379
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 689	score: 0.8056
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 690	score: 0.9503
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 691	score: 0.8203
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 692	score: 0.8864
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 693	score: 0.6114
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 694	score: 0.7500
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 695	score: 0.8881
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 696	score: 0.8438
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 697	score: 0.5998
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 698	score: 0.7335
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 699	score: 0.7911
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 700	score: 0.8511
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 701	score: 0.9138
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 702	score: 0.8196
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 703	score: 0.7187
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 704	score: 0.4829
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 705	score: 0.6068
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 706	score: 0.7477
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 707	score: 0.8340
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 708	score: 0.8226
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 709	score: 0.7198
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 710	score: 0.9404
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 711	score: 0.6872
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 712	score: 0.7203
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 713	score: 0.8599
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 714	score: 0.7633
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 715	score: 0.6110
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 716	score: 0.9092
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 717	score: 0.6894
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 718	score: 0.8457
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 719	score: 0.5901
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 720	score: 0.8299
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 721	score: 0.8560
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 722	score: 0.8696
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 723	score: 0.8990
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 724	score: 0.9008
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 725	score: 0.8816
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 726	score: 0.8325
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 727	score: 0.5809
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 728	score: 0.5405
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 729	score: 0.7006
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 730	score: 0.8323
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 731	score: 0.7775
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 732	score: 0.7696
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 733	score: 0.8843
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 734	score: 0.7858
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 735	score: 0.8104
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 736	score: 0.8291
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 737	score: 0.6761
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 738	score: 0.6858
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 739	score: 0.8361
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 740	score: 0.7622
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 741	score: 0.7870
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 742	score: 0.7022
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 743	score: 0.8054
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 744	score: 0.8155
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 745	score: 0.5766
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 746	score: 0.7663
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 747	score: 0.7414
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 748	score: 0.5121
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 749	score: 0.8277
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 750	score: 0.9044
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 751	score: 0.8641
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 752	score: 0.7215
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 753	score: 0.7535
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 754	score: 0.8808
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 755	score: 0.6456
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 756	score: 0.5887
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 757	score: 0.8301
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 758	score: 0.5110
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 759	score: 0.5297
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 760	score: 0.7174
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 761	score: 0.9078
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 762	score: 0.5270
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 763	score: 0.8173
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 764	score: 0.8710
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 765	score: 0.6683
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 766	score: 0.6855
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 767	score: 0.8293
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 768	score: 0.8111
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 769	score: 0.8669
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 770	score: 0.8297
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 771	score: 0.8460
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 772	score: 0.8209
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 773	score: 0.7145
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 774	score: 0.7156
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 775	score: 0.6454
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 776	score: 0.7947
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 777	score: 0.7611
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 778	score: 0.8925
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 779	score: 0.8548
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 780	score: 0.9010
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 781	score: 0.8421
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 782	score: 0.7014
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 783	score: 0.5793
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 784	score: 0.4501
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 785	score: 0.7335
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 786	score: 0.8698
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 787	score: 0.8166
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 788	score: 0.9423
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 789	score: 0.8615
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 790	score: 0.8369
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 791	score: 0.8000
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 792	score: 0.8729
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 793	score: 0.7535
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 794	score: 0.5746
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 795	score: 0.8037
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 796	score: 0.9269
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 797	score: 0.6427
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 798	score: 0.7709
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 799	score: 0.7205
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 800	score: 0.5622
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 801	score: 0.8278
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 802	score: 0.8906
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 803	score: 0.8434
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 804	score: 0.7409
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 805	score: 0.8627
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 806	score: 0.7678
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 807	score: 0.8321
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 808	score: 0.7885
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 809	score: 0.7160
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 810	score: 0.7159
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 811	score: 0.7971
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 812	score: 0.9067
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 813	score: 0.8839
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 814	score: 0.8395
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 815	score: 0.7829
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 816	score: 0.8849
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 817	score: 0.6660
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 818	score: 0.7963
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 819	score: 0.6193
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 820	score: 0.8964
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 821	score: 0.6557
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 822	score: 0.7644
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 823	score: 0.6650
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 824	score: 0.8370
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 825	score: 0.8353
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 826	score: 0.7332
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 827	score: 0.7926
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 828	score: 0.7491
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 829	score: 0.7098
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 830	score: 0.7194
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 831	score: 0.8247
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 832	score: 0.6763
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 833	score: 0.7091
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 834	score: 0.8086
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 835	score: 0.6768
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 836	score: 0.5887
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 837	score: 0.5636
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 838	score: 0.7250
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 839	score: 0.8278
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 840	score: 0.4907
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 841	score: 0.6673
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 842	score: 0.6884
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 843	score: 0.7838
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 844	score: 0.6844
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 845	score: 0.4460
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 846	score: 0.6840
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 847	score: 0.5493
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 848	score: 0.7573
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 849	score: 0.8846
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 850	score: 0.8178
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 851	score: 0.8999
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 852	score: 0.8766
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 853	score: 0.8531
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 854	score: 0.7119
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 855	score: 0.7814
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 856	score: 0.7543
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 857	score: 0.8766
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 858	score: 0.4327
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 859	score: 0.5904
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 860	score: 0.7514
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 861	score: 0.8686
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 862	score: 0.6268
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 863	score: 0.7245
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 864	score: 0.5008
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 865	score: 0.5666
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 866	score: 0.7032
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 867	score: 0.8688
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 868	score: 0.7676
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 869	score: 0.8150
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 870	score: 0.7536
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 871	score: 0.7505
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 872	score: 0.7903
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 873	score: 0.8545
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 874	score: 0.8304
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 875	score: 0.8222
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 876	score: 0.7619
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 877	score: 0.8083
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 878	score: 0.8121
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 879	score: 0.8482
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 880	score: 0.8294
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 881	score: 0.9198
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 882	score: 0.7680
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 883	score: 0.7907
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 884	score: 0.8210
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 885	score: 0.7383
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 886	score: 0.8172
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 887	score: 0.8367
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 888	score: 0.4950
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 889	score: 0.7834
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 890	score: 0.6034
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 891	score: 0.6251
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 892	score: 0.6904
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 893	score: 0.6481
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 894	score: 0.8583
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 895	score: 0.8332
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 896	score: 0.8256
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 897	score: 0.6572
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 898	score: 0.7838
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 899	score: 0.5514
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 900	score: 0.8321
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 901	score: 0.5272
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 902	score: 0.8345
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 903	score: 0.7636
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 904	score: 0.8373
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 905	score: 0.7757
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 906	score: 0.7443
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 907	score: 0.6055
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 908	score: 0.7650
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 909	score: 0.7585
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 910	score: 0.8686
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 911	score: 0.7886
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 912	score: 0.8848
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 913	score: 0.8888
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 914	score: 0.7768
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 915	score: 0.8400
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 916	score: 0.9392
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 917	score: 0.8052
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 918	score: 0.7605
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 919	score: 0.6055
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 920	score: 0.6036
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 921	score: 0.8707
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 922	score: 0.5988
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 923	score: 0.8764
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 924	score: 0.8467
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 925	score: 0.6992
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 926	score: 0.8602
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 927	score: 0.7278
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 928	score: 0.7915
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 929	score: 0.6834
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 930	score: 0.8683
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 931	score: 0.8736
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 932	score: 0.7770
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 933	score: 0.7995
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 934	score: 0.8046
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 935	score: 0.7899
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 936	score: 0.8634
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 937	score: 0.9489
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 938	score: 0.9306
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 939	score: 0.8264
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 940	score: 0.8745
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 941	score: 0.9193
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 942	score: 0.8455
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 943	score: 0.8658
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 944	score: 0.6186
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 945	score: 0.7580
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 946	score: 0.7368
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 947	score: 0.7356
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 948	score: 0.8061
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 949	score: 0.8654
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 950	score: 0.8071
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 951	score: 0.6980
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 952	score: 0.8063
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 953	score: 0.6024
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 954	score: 0.5641
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 955	score: 0.9245
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 956	score: 0.7010
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 957	score: 0.8770
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 958	score: 0.8369
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 959	score: 0.8851
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 960	score: 0.5145
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 961	score: 0.6777
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 962	score: 0.9042
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 963	score: 0.8350
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 964	score: 0.9463
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 965	score: 0.6900
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 966	score: 0.6140
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 967	score: 0.7960
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 968	score: 0.7274
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 969	score: 0.6213
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 970	score: 0.9032
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 971	score: 0.7788
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 972	score: 0.5541
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 973	score: 0.8340
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 974	score: 0.7550
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 975	score: 0.8679
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 976	score: 0.8291
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 977	score: 0.8277
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 978	score: 0.7647
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 979	score: 0.8016
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 980	score: 0.7381
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 981	score: 0.8383
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 982	score: 0.8120
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 983	score: 0.7691
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 984	score: 0.7078
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 985	score: 0.8809
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 986	score: 0.9043
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 987	score: 0.8794
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 988	score: 0.5184
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 989	score: 0.6220
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 990	score: 0.9185
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 991	score: 0.7758
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 992	score: 0.8440
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 993	score: 0.4811
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 994	score: 0.5914
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 995	score: 0.8461
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 996	score: 0.8382
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 997	score: 0.8793
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 998	score: 0.8885
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 999	score: 0.8359
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 1000	score: 0.7390
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 1001	score: 0.8219
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 1002	score: 0.4998
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 1003	score: 0.7650
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 1004	score: 0.8447
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 1005	score: 0.8869
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 1006	score: 0.7567
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 1007	score: 0.8575
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 1008	score: 0.8741
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 1009	score: 0.7134
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 1010	score: 0.7961
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	Segment 1011	score: 0.8425
+/beacon-scratch/tongzh24/ALMA-checkpoint/exp_16_languages/alma-13b-sft-16-languages-mr-max-tokens-512//test-mr-en	score: 0.7660