diff --git "a/test-en-ka.comet" "b/test-en-ka.comet" new file mode 100644--- /dev/null +++ "b/test-en-ka.comet" @@ -0,0 +1,1013 @@ +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 0 score: 0.2279 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 1 score: 0.7640 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 2 score: 0.5580 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 3 score: 0.7709 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 4 score: 0.6167 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 5 score: 0.5361 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 6 score: 0.5361 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 7 score: 0.8430 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 8 score: 0.8093 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 9 score: 0.2448 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 10 score: 0.7196 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 11 score: 0.5616 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 12 score: 0.5487 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 13 score: 0.8753 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 14 score: 0.7523 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 15 score: 0.4731 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 16 score: 0.7850 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 17 score: 0.8790 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 18 score: 0.7548 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 19 score: 0.6949 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 20 score: 0.8997 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 21 score: 0.6743 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 22 score: 0.5968 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 23 score: 0.7686 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 24 score: 0.9044 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 25 score: 0.8582 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 26 score: 0.5125 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 27 score: 0.4615 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 28 score: 0.7215 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 29 score: 0.6223 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 30 score: 0.5683 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 31 score: 0.6238 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 32 score: 0.5281 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 33 score: 0.9048 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 34 score: 0.6396 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 35 score: 0.6526 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 36 score: 0.7500 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 37 score: 0.7827 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 38 score: 0.7851 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 39 score: 0.8591 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 40 score: 0.5579 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 41 score: 0.8873 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 42 score: 0.6506 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 43 score: 0.8629 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 44 score: 0.4459 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 45 score: 0.5962 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 46 score: 0.8287 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 47 score: 0.7153 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 48 score: 0.7757 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 49 score: 0.7518 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 50 score: 0.8610 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 51 score: 0.6110 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 52 score: 0.7232 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 53 score: 0.2489 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 54 score: 0.4092 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 55 score: 0.8466 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 56 score: 0.7520 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 57 score: 0.7322 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 58 score: 0.8258 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 59 score: 0.2283 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 60 score: 0.4592 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 61 score: 0.7304 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 62 score: 0.8904 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 63 score: 0.5426 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 64 score: 0.7102 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 65 score: 0.2822 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 66 score: 0.7202 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 67 score: 0.6310 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 68 score: 0.6855 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 69 score: 0.8379 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 70 score: 0.1882 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 71 score: 0.5514 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 72 score: 0.7287 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 73 score: 0.8354 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 74 score: 0.6893 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 75 score: 0.5995 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 76 score: 0.6021 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 77 score: 0.7896 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 78 score: 0.6969 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 79 score: 0.8603 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 80 score: 0.8234 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 81 score: 0.6399 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 82 score: 0.5320 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 83 score: 0.2561 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 84 score: 0.5474 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 85 score: 0.6365 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 86 score: 0.4166 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 87 score: 0.7569 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 88 score: 0.6707 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 89 score: 0.5008 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 90 score: 0.2395 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 91 score: 0.5283 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 92 score: 0.5600 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 93 score: 0.1957 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 94 score: 0.4490 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 95 score: 0.8486 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 96 score: 0.2549 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 97 score: 0.6839 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 98 score: 0.8574 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 99 score: 0.5599 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 100 score: 0.8576 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 101 score: 0.7619 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 102 score: 0.5991 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 103 score: 0.5607 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 104 score: 0.5133 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 105 score: 0.6262 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 106 score: 0.7234 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 107 score: 0.3457 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 108 score: 0.3085 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 109 score: 0.4619 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 110 score: 0.7102 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 111 score: 0.9226 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 112 score: 0.6470 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 113 score: 0.2971 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 114 score: 0.4953 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 115 score: 0.7665 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 116 score: 0.7618 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 117 score: 0.8071 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 118 score: 0.9065 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 119 score: 0.5015 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 120 score: 0.5857 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 121 score: 0.8588 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 122 score: 0.6472 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 123 score: 0.8084 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 124 score: 0.5353 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 125 score: 0.5280 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 126 score: 0.5961 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 127 score: 0.6709 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 128 score: 0.5446 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 129 score: 0.4318 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 130 score: 0.8535 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 131 score: 0.6884 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 132 score: 0.5954 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 133 score: 0.7114 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 134 score: 0.1843 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 135 score: 0.4893 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 136 score: 0.6956 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 137 score: 0.6486 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 138 score: 0.8446 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 139 score: 0.5378 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 140 score: 0.5311 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 141 score: 0.7190 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 142 score: 0.5969 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 143 score: 0.6493 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 144 score: 0.6627 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 145 score: 0.7742 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 146 score: 0.6284 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 147 score: 0.5720 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 148 score: 0.8121 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 149 score: 0.5730 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 150 score: 0.7538 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 151 score: 0.8387 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 152 score: 0.5991 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 153 score: 0.6945 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 154 score: 0.6998 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 155 score: 0.5015 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 156 score: 0.2739 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 157 score: 0.4788 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 158 score: 0.4854 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 159 score: 0.6833 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 160 score: 0.6607 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 161 score: 0.7558 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 162 score: 0.3217 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 163 score: 0.4465 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 164 score: 0.8966 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 165 score: 0.2736 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 166 score: 0.5575 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 167 score: 0.7852 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 168 score: 0.6301 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 169 score: 0.7414 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 170 score: 0.6624 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 171 score: 0.7371 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 172 score: 0.5342 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 173 score: 0.4310 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 174 score: 0.6708 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 175 score: 0.7308 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 176 score: 0.8231 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 177 score: 0.8625 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 178 score: 0.8734 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 179 score: 0.8945 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 180 score: 0.8371 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 181 score: 0.7626 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 182 score: 0.8812 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 183 score: 0.8707 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 184 score: 0.3496 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 185 score: 0.7986 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 186 score: 0.6868 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 187 score: 0.3549 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 188 score: 0.6771 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 189 score: 0.7456 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 190 score: 0.5248 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 191 score: 0.8048 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 192 score: 0.6745 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 193 score: 0.7210 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 194 score: 0.8430 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 195 score: 0.9181 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 196 score: 0.4337 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 197 score: 0.3828 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 198 score: 0.4632 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 199 score: 0.6471 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 200 score: 0.5894 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 201 score: 0.7633 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 202 score: 0.7709 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 203 score: 0.5817 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 204 score: 0.2615 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 205 score: 0.5368 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 206 score: 0.2345 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 207 score: 0.6796 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 208 score: 0.6658 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 209 score: 0.8559 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 210 score: 0.9479 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 211 score: 0.5887 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 212 score: 0.9331 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 213 score: 0.8095 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 214 score: 0.7709 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 215 score: 0.5328 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 216 score: 0.8750 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 217 score: 0.6470 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 218 score: 0.6630 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 219 score: 0.7870 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 220 score: 0.9491 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 221 score: 0.5291 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 222 score: 0.6969 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 223 score: 0.4766 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 224 score: 0.8679 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 225 score: 0.6537 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 226 score: 0.8631 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 227 score: 0.5216 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 228 score: 0.5455 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 229 score: 0.7213 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 230 score: 0.7673 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 231 score: 0.5634 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 232 score: 0.6723 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 233 score: 0.6730 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 234 score: 0.5720 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 235 score: 0.5527 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 236 score: 0.7034 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 237 score: 0.4447 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 238 score: 0.5742 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 239 score: 0.6776 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 240 score: 0.3788 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 241 score: 0.4847 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 242 score: 0.7478 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 243 score: 0.1699 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 244 score: 0.3392 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 245 score: 0.4980 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 246 score: 0.7106 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 247 score: 0.4353 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 248 score: 0.3397 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 249 score: 0.9005 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 250 score: 0.7468 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 251 score: 0.8986 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 252 score: 0.6208 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 253 score: 0.7939 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 254 score: 0.6621 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 255 score: 0.7067 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 256 score: 0.4965 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 257 score: 0.7346 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 258 score: 0.5876 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 259 score: 0.6246 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 260 score: 0.3487 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 261 score: 0.2767 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 262 score: 0.3902 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 263 score: 0.3897 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 264 score: 0.8273 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 265 score: 0.1865 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 266 score: 0.6827 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 267 score: 0.4858 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 268 score: 0.8753 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 269 score: 0.8733 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 270 score: 0.5320 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 271 score: 0.5028 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 272 score: 0.6705 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 273 score: 0.7825 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 274 score: 0.6442 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 275 score: 0.2186 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 276 score: 0.3954 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 277 score: 0.6143 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 278 score: 0.5937 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 279 score: 0.9328 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 280 score: 0.6420 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 281 score: 0.7842 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 282 score: 0.7805 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 283 score: 0.2435 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 284 score: 0.8340 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 285 score: 0.6312 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 286 score: 0.8862 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 287 score: 0.5142 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 288 score: 0.6740 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 289 score: 0.8117 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 290 score: 0.6112 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 291 score: 0.6313 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 292 score: 0.6167 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 293 score: 0.3412 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 294 score: 0.2067 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 295 score: 0.2351 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 296 score: 0.2860 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 297 score: 0.8111 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 298 score: 0.8638 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 299 score: 0.5323 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 300 score: 0.8570 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 301 score: 0.1651 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 302 score: 0.6559 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 303 score: 0.8742 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 304 score: 0.7112 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 305 score: 0.5741 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 306 score: 0.8912 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 307 score: 0.7961 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 308 score: 0.6872 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 309 score: 0.5326 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 310 score: 0.8651 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 311 score: 0.4430 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 312 score: 0.6094 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 313 score: 0.5870 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 314 score: 0.8433 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 315 score: 0.7312 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 316 score: 0.2795 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 317 score: 0.7966 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 318 score: 0.2772 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 319 score: 0.6441 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 320 score: 0.3311 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 321 score: 0.5549 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 322 score: 0.7974 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 323 score: 0.8884 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 324 score: 0.2330 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 325 score: 0.6518 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 326 score: 0.4743 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 327 score: 0.7156 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 328 score: 0.7915 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 329 score: 0.6760 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 330 score: 0.2181 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 331 score: 0.4693 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 332 score: 0.5355 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 333 score: 0.5214 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 334 score: 0.8009 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 335 score: 0.7158 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 336 score: 0.5250 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 337 score: 0.6836 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 338 score: 0.5746 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 339 score: 0.8297 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 340 score: 0.6395 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 341 score: 0.7166 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 342 score: 0.3864 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 343 score: 0.5499 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 344 score: 0.7321 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 345 score: 0.1644 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 346 score: 0.7420 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 347 score: 0.7769 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 348 score: 0.4672 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 349 score: 0.4831 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 350 score: 0.7894 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 351 score: 0.7288 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 352 score: 0.7346 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 353 score: 0.5314 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 354 score: 0.5260 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 355 score: 0.6256 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 356 score: 0.5963 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 357 score: 0.2820 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 358 score: 0.2472 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 359 score: 0.2712 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 360 score: 0.3023 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 361 score: 0.6333 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 362 score: 0.8964 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 363 score: 0.2319 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 364 score: 0.6563 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 365 score: 0.5481 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 366 score: 0.2588 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 367 score: 0.5648 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 368 score: 0.5035 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 369 score: 0.5494 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 370 score: 0.2782 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 371 score: 0.3898 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 372 score: 0.5093 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 373 score: 0.8190 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 374 score: 0.8741 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 375 score: 0.8052 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 376 score: 0.6240 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 377 score: 0.7872 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 378 score: 0.5850 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 379 score: 0.3199 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 380 score: 0.2109 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 381 score: 0.5481 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 382 score: 0.3530 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 383 score: 0.5186 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 384 score: 0.4677 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 385 score: 0.3150 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 386 score: 0.3778 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 387 score: 0.5584 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 388 score: 0.4416 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 389 score: 0.6654 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 390 score: 0.2568 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 391 score: 0.5563 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 392 score: 0.8750 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 393 score: 0.8067 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 394 score: 0.2675 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 395 score: 0.4695 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 396 score: 0.6786 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 397 score: 0.8418 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 398 score: 0.6583 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 399 score: 0.6766 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 400 score: 0.9172 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 401 score: 0.3426 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 402 score: 0.8304 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 403 score: 0.6297 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 404 score: 0.5433 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 405 score: 0.4458 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 406 score: 0.8439 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 407 score: 0.4977 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 408 score: 0.8933 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 409 score: 0.7199 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 410 score: 0.7863 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 411 score: 0.3321 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 412 score: 0.6082 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 413 score: 0.5177 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 414 score: 0.7951 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 415 score: 0.5569 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 416 score: 0.6546 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 417 score: 0.3278 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 418 score: 0.9335 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 419 score: 0.8007 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 420 score: 0.9606 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 421 score: 0.8754 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 422 score: 0.7357 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 423 score: 0.7319 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 424 score: 0.5842 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 425 score: 0.8211 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 426 score: 0.6240 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 427 score: 0.7093 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 428 score: 0.7615 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 429 score: 0.8861 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 430 score: 0.6697 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 431 score: 0.5567 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 432 score: 0.4987 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 433 score: 0.5319 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 434 score: 0.5967 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 435 score: 0.5278 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 436 score: 0.4408 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 437 score: 0.6034 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 438 score: 0.8587 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 439 score: 0.7021 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 440 score: 0.5970 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 441 score: 0.5825 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 442 score: 0.8338 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 443 score: 0.6261 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 444 score: 0.3542 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 445 score: 0.6539 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 446 score: 0.6923 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 447 score: 0.7516 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 448 score: 0.9180 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 449 score: 0.7183 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 450 score: 0.8297 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 451 score: 0.8325 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 452 score: 0.6411 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 453 score: 0.8835 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 454 score: 0.7204 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 455 score: 0.6470 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 456 score: 0.5707 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 457 score: 0.8070 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 458 score: 0.5086 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 459 score: 0.4983 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 460 score: 0.6826 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 461 score: 0.4608 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 462 score: 0.3893 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 463 score: 0.7546 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 464 score: 0.6278 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 465 score: 0.8145 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 466 score: 0.8336 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 467 score: 0.4492 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 468 score: 0.4940 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 469 score: 0.3371 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 470 score: 0.5817 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 471 score: 0.7565 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 472 score: 0.5816 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 473 score: 0.3905 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 474 score: 0.2112 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 475 score: 0.6854 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 476 score: 0.4132 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 477 score: 0.6213 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 478 score: 0.6728 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 479 score: 0.2054 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 480 score: 0.2096 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 481 score: 0.7012 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 482 score: 0.7726 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 483 score: 0.6437 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 484 score: 0.7970 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 485 score: 0.2934 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 486 score: 0.6187 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 487 score: 0.6802 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 488 score: 0.8414 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 489 score: 0.5252 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 490 score: 0.5089 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 491 score: 0.7324 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 492 score: 0.9275 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 493 score: 0.6478 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 494 score: 0.2544 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 495 score: 0.6097 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 496 score: 0.6015 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 497 score: 0.5737 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 498 score: 0.5038 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 499 score: 0.7733 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 500 score: 0.8020 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 501 score: 0.9058 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 502 score: 0.9019 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 503 score: 0.6578 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 504 score: 0.6090 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 505 score: 0.5473 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 506 score: 0.5536 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 507 score: 0.4570 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 508 score: 0.6035 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 509 score: 0.9102 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 510 score: 0.5701 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 511 score: 0.6406 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 512 score: 0.4629 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 513 score: 0.4129 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 514 score: 0.2430 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 515 score: 0.6569 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 516 score: 0.4409 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 517 score: 0.8606 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 518 score: 0.3675 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 519 score: 0.9377 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 520 score: 0.8594 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 521 score: 0.7889 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 522 score: 0.3411 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 523 score: 0.4148 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 524 score: 0.7892 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 525 score: 0.6372 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 526 score: 0.5976 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 527 score: 0.6597 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 528 score: 0.5179 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 529 score: 0.6652 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 530 score: 0.3682 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 531 score: 0.5648 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 532 score: 0.4400 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 533 score: 0.6145 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 534 score: 0.4716 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 535 score: 0.4091 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 536 score: 0.3064 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 537 score: 0.5356 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 538 score: 0.6160 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 539 score: 0.6968 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 540 score: 0.6630 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 541 score: 0.8827 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 542 score: 0.8164 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 543 score: 0.6519 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 544 score: 0.6443 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 545 score: 0.5504 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 546 score: 0.5938 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 547 score: 0.8379 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 548 score: 0.6371 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 549 score: 0.6464 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 550 score: 0.5942 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 551 score: 0.8737 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 552 score: 0.6623 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 553 score: 0.2711 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 554 score: 0.8347 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 555 score: 0.4068 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 556 score: 0.4794 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 557 score: 0.5377 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 558 score: 0.5750 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 559 score: 0.4733 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 560 score: 0.4042 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 561 score: 0.1440 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 562 score: 0.5719 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 563 score: 0.4510 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 564 score: 0.6476 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 565 score: 0.7163 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 566 score: 0.5750 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 567 score: 0.8095 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 568 score: 0.4752 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 569 score: 0.8596 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 570 score: 0.7256 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 571 score: 0.5903 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 572 score: 0.6941 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 573 score: 0.6417 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 574 score: 0.5790 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 575 score: 0.7401 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 576 score: 0.3916 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 577 score: 0.7650 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 578 score: 0.5938 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 579 score: 0.4842 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 580 score: 0.5355 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 581 score: 0.6535 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 582 score: 0.7696 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 583 score: 0.5072 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 584 score: 0.4617 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 585 score: 0.7529 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 586 score: 0.6805 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 587 score: 0.6421 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 588 score: 0.6436 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 589 score: 0.4060 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 590 score: 0.5145 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 591 score: 0.7160 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 592 score: 0.8016 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 593 score: 0.6354 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 594 score: 0.6588 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 595 score: 0.4735 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 596 score: 0.7191 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 597 score: 0.7826 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 598 score: 0.2703 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 599 score: 0.8575 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 600 score: 0.8278 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 601 score: 0.6260 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 602 score: 0.6282 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 603 score: 0.6032 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 604 score: 0.2482 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 605 score: 0.2770 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 606 score: 0.6941 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 607 score: 0.3641 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 608 score: 0.6897 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 609 score: 0.6894 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 610 score: 0.7386 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 611 score: 0.2518 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 612 score: 0.5376 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 613 score: 0.6724 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 614 score: 0.5584 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 615 score: 0.5413 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 616 score: 0.7251 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 617 score: 0.7725 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 618 score: 0.2966 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 619 score: 0.5308 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 620 score: 0.3643 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 621 score: 0.4596 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 622 score: 0.8717 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 623 score: 0.2922 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 624 score: 0.3836 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 625 score: 0.6712 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 626 score: 0.8841 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 627 score: 0.8310 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 628 score: 0.7661 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 629 score: 0.5733 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 630 score: 0.7284 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 631 score: 0.9007 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 632 score: 0.9367 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 633 score: 0.7476 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 634 score: 0.9006 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 635 score: 0.6111 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 636 score: 0.6995 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 637 score: 0.3148 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 638 score: 0.6865 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 639 score: 0.7858 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 640 score: 0.6835 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 641 score: 0.9184 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 642 score: 0.6407 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 643 score: 0.6250 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 644 score: 0.3949 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 645 score: 0.4448 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 646 score: 0.5386 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 647 score: 0.7845 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 648 score: 0.2252 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 649 score: 0.4822 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 650 score: 0.6238 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 651 score: 0.2163 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 652 score: 0.7911 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 653 score: 0.4843 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 654 score: 0.6002 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 655 score: 0.7305 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 656 score: 0.3941 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 657 score: 0.8564 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 658 score: 0.5609 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 659 score: 0.6501 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 660 score: 0.5921 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 661 score: 0.5263 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 662 score: 0.7989 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 663 score: 0.7238 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 664 score: 0.6527 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 665 score: 0.6783 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 666 score: 0.3461 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 667 score: 0.6782 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 668 score: 0.7552 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 669 score: 0.8465 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 670 score: 0.7696 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 671 score: 0.8835 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 672 score: 0.8192 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 673 score: 0.7468 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 674 score: 0.7606 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 675 score: 0.6374 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 676 score: 0.6677 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 677 score: 0.5067 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 678 score: 0.6620 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 679 score: 0.7793 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 680 score: 0.5615 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 681 score: 0.7515 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 682 score: 0.9082 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 683 score: 0.6492 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 684 score: 0.4053 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 685 score: 0.6591 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 686 score: 0.6563 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 687 score: 0.7090 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 688 score: 0.7754 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 689 score: 0.5809 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 690 score: 0.8988 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 691 score: 0.7656 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 692 score: 0.8687 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 693 score: 0.5344 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 694 score: 0.6267 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 695 score: 0.5873 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 696 score: 0.6193 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 697 score: 0.6785 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 698 score: 0.5843 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 699 score: 0.7718 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 700 score: 0.5357 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 701 score: 0.8812 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 702 score: 0.8013 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 703 score: 0.7176 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 704 score: 0.3748 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 705 score: 0.2920 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 706 score: 0.3654 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 707 score: 0.7628 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 708 score: 0.5758 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 709 score: 0.8033 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 710 score: 0.7387 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 711 score: 0.6653 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 712 score: 0.8642 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 713 score: 0.6268 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 714 score: 0.7715 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 715 score: 0.7660 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 716 score: 0.5964 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 717 score: 0.7634 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 718 score: 0.7203 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 719 score: 0.5326 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 720 score: 0.8861 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 721 score: 0.9389 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 722 score: 0.7418 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 723 score: 0.7224 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 724 score: 0.7405 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 725 score: 0.7919 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 726 score: 0.7310 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 727 score: 0.7539 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 728 score: 0.5482 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 729 score: 0.5367 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 730 score: 0.6654 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 731 score: 0.6240 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 732 score: 0.8050 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 733 score: 0.7592 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 734 score: 0.5970 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 735 score: 0.8116 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 736 score: 0.8339 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 737 score: 0.3138 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 738 score: 0.5625 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 739 score: 0.5312 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 740 score: 0.8090 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 741 score: 0.8126 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 742 score: 0.6171 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 743 score: 0.5007 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 744 score: 0.7996 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 745 score: 0.3929 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 746 score: 0.5546 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 747 score: 0.8229 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 748 score: 0.9543 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 749 score: 0.6317 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 750 score: 0.7783 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 751 score: 0.9001 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 752 score: 0.2909 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 753 score: 0.7109 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 754 score: 0.9187 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 755 score: 0.2900 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 756 score: 0.7569 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 757 score: 0.6347 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 758 score: 0.4781 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 759 score: 0.6521 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 760 score: 0.9513 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 761 score: 0.2541 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 762 score: 0.7998 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 763 score: 0.2524 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 764 score: 0.6162 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 765 score: 0.4939 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 766 score: 0.5419 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 767 score: 0.2407 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 768 score: 0.9558 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 769 score: 0.9277 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 770 score: 0.7547 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 771 score: 0.9078 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 772 score: 0.8221 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 773 score: 0.8611 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 774 score: 0.8952 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 775 score: 0.5051 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 776 score: 0.7452 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 777 score: 0.6740 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 778 score: 0.6138 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 779 score: 0.7928 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 780 score: 0.5845 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 781 score: 0.6009 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 782 score: 0.5559 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 783 score: 0.5223 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 784 score: 0.7937 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 785 score: 0.2713 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 786 score: 0.5796 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 787 score: 0.7441 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 788 score: 0.2556 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 789 score: 0.3628 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 790 score: 0.7807 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 791 score: 0.7920 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 792 score: 0.7477 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 793 score: 0.5686 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 794 score: 0.6293 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 795 score: 0.6377 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 796 score: 0.8010 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 797 score: 0.5510 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 798 score: 0.6121 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 799 score: 0.7475 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 800 score: 0.6039 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 801 score: 0.3774 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 802 score: 0.2065 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 803 score: 0.6541 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 804 score: 0.2327 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 805 score: 0.3579 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 806 score: 0.5915 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 807 score: 0.8129 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 808 score: 0.5708 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 809 score: 0.5503 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 810 score: 0.6170 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 811 score: 0.2378 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 812 score: 0.9219 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 813 score: 0.7229 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 814 score: 0.2827 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 815 score: 0.5705 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 816 score: 0.7759 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 817 score: 0.4899 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 818 score: 0.4103 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 819 score: 0.3019 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 820 score: 0.8200 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 821 score: 0.7019 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 822 score: 0.4921 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 823 score: 0.2319 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 824 score: 0.7635 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 825 score: 0.7056 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 826 score: 0.4529 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 827 score: 0.6465 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 828 score: 0.4814 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 829 score: 0.4000 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 830 score: 0.3216 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 831 score: 0.7871 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 832 score: 0.6524 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 833 score: 0.1665 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 834 score: 0.5467 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 835 score: 0.5015 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 836 score: 0.5736 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 837 score: 0.4848 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 838 score: 0.7877 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 839 score: 0.6841 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 840 score: 0.2754 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 841 score: 0.6035 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 842 score: 0.3586 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 843 score: 0.6437 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 844 score: 0.4241 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 845 score: 0.2440 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 846 score: 0.3608 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 847 score: 0.5693 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 848 score: 0.2443 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 849 score: 0.5359 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 850 score: 0.5554 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 851 score: 0.4092 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 852 score: 0.8415 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 853 score: 0.4472 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 854 score: 0.3838 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 855 score: 0.4139 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 856 score: 0.3027 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 857 score: 0.8356 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 858 score: 0.8622 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 859 score: 0.7030 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 860 score: 0.5658 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 861 score: 0.5587 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 862 score: 0.5888 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 863 score: 0.6773 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 864 score: 0.2530 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 865 score: 0.5359 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 866 score: 0.6404 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 867 score: 0.6208 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 868 score: 0.6226 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 869 score: 0.7740 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 870 score: 0.7560 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 871 score: 0.8366 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 872 score: 0.4762 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 873 score: 0.6583 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 874 score: 0.6873 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 875 score: 0.3744 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 876 score: 0.6590 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 877 score: 0.5996 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 878 score: 0.2286 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 879 score: 0.3741 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 880 score: 0.7633 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 881 score: 0.6108 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 882 score: 0.6278 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 883 score: 0.6474 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 884 score: 0.5323 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 885 score: 0.7786 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 886 score: 0.7509 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 887 score: 0.5917 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 888 score: 0.6360 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 889 score: 0.7120 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 890 score: 0.7289 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 891 score: 0.6189 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 892 score: 0.4768 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 893 score: 0.6434 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 894 score: 0.5392 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 895 score: 0.6554 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 896 score: 0.6494 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 897 score: 0.6839 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 898 score: 0.8245 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 899 score: 0.7744 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 900 score: 0.7637 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 901 score: 0.3052 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 902 score: 0.6107 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 903 score: 0.6325 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 904 score: 0.8158 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 905 score: 0.8262 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 906 score: 0.7197 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 907 score: 0.4637 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 908 score: 0.7939 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 909 score: 0.8233 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 910 score: 0.5445 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 911 score: 0.6742 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 912 score: 0.2417 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 913 score: 0.8084 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 914 score: 0.5485 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 915 score: 0.6205 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 916 score: 0.5781 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 917 score: 0.7535 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 918 score: 0.9193 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 919 score: 0.7411 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 920 score: 0.5270 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 921 score: 0.7838 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 922 score: 0.2682 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 923 score: 0.4799 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 924 score: 0.7422 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 925 score: 0.5804 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 926 score: 0.7006 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 927 score: 0.5684 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 928 score: 0.2059 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 929 score: 0.7035 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 930 score: 0.9017 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 931 score: 0.4524 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 932 score: 0.4198 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 933 score: 0.5187 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 934 score: 0.4243 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 935 score: 0.5919 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 936 score: 0.6319 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 937 score: 0.9526 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 938 score: 0.8954 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 939 score: 0.7327 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 940 score: 0.6818 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 941 score: 0.9613 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 942 score: 0.8625 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 943 score: 0.8574 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 944 score: 0.7076 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 945 score: 0.5692 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 946 score: 0.5418 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 947 score: 0.4444 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 948 score: 0.2344 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 949 score: 0.5663 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 950 score: 0.2662 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 951 score: 0.2528 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 952 score: 0.2509 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 953 score: 0.5793 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 954 score: 0.4677 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 955 score: 0.7375 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 956 score: 0.7743 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 957 score: 0.9129 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 958 score: 0.7514 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 959 score: 0.8767 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 960 score: 0.5881 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 961 score: 0.3012 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 962 score: 0.5638 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 963 score: 0.4420 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 964 score: 0.7901 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 965 score: 0.7182 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 966 score: 0.4359 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 967 score: 0.2512 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 968 score: 0.2273 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 969 score: 0.4098 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 970 score: 0.4700 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 971 score: 0.5156 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 972 score: 0.4781 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 973 score: 0.6857 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 974 score: 0.4791 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 975 score: 0.7706 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 976 score: 0.5496 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 977 score: 0.6724 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 978 score: 0.8274 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 979 score: 0.2425 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 980 score: 0.7988 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 981 score: 0.6132 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 982 score: 0.6621 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 983 score: 0.5752 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 984 score: 0.6259 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 985 score: 0.5409 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 986 score: 0.5713 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 987 score: 0.4593 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 988 score: 0.4033 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 989 score: 0.5175 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 990 score: 0.3642 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 991 score: 0.5269 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 992 score: 0.8324 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 993 score: 0.7612 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 994 score: 0.4329 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 995 score: 0.6580 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 996 score: 0.8008 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 997 score: 0.5328 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 998 score: 0.6134 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 999 score: 0.5020 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 1000 score: 0.7101 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 1001 score: 0.3408 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 1002 score: 0.4311 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 1003 score: 0.1842 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 1004 score: 0.4608 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 1005 score: 0.6038 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 1006 score: 0.5006 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 1007 score: 0.4267 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 1008 score: 0.8933 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 1009 score: 0.5258 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 1010 score: 0.2629 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka Segment 1011 score: 0.5788 +/beacon-scratch/tongzh24/ALMA-checkpoint/exp_50_languages/alma-13b-sft-group-6-max-tokens-512//test-en-ka score: 0.6163