{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 10.0,
  "eval_steps": 28,
  "global_step": 560,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.11,
      "learning_rate": 3e-05,
      "loss": 3.5167,
      "step": 6
    },
    {
      "epoch": 0.21,
      "learning_rate": 3e-05,
      "loss": 2.7034,
      "step": 12
    },
    {
      "epoch": 0.32,
      "learning_rate": 3e-05,
      "loss": 2.5409,
      "step": 18
    },
    {
      "epoch": 0.43,
      "learning_rate": 3e-05,
      "loss": 2.4931,
      "step": 24
    },
    {
      "epoch": 0.5,
      "eval_accuracy": 0.5748751060189339,
      "eval_loss": 2.390355348587036,
      "eval_runtime": 25.7729,
      "eval_samples_per_second": 34.261,
      "eval_steps_per_second": 2.173,
      "step": 28
    },
    {
      "epoch": 0.5,
      "eval_bleu": 2.6290615444017718,
      "eval_exact_match": 0.0,
      "eval_prefix_exact_match": 0.0,
      "step": 28
    },
    {
      "epoch": 0.54,
      "learning_rate": 3e-05,
      "loss": 2.4944,
      "step": 30
    },
    {
      "epoch": 0.64,
      "learning_rate": 3e-05,
      "loss": 2.4832,
      "step": 36
    },
    {
      "epoch": 0.75,
      "learning_rate": 3e-05,
      "loss": 2.4336,
      "step": 42
    },
    {
      "epoch": 0.86,
      "learning_rate": 3e-05,
      "loss": 2.5101,
      "step": 48
    },
    {
      "epoch": 0.96,
      "learning_rate": 3e-05,
      "loss": 2.4699,
      "step": 54
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6049526162776446,
      "eval_loss": 2.101423740386963,
      "eval_runtime": 23.9781,
      "eval_samples_per_second": 36.825,
      "eval_steps_per_second": 2.335,
      "step": 56
    },
    {
      "epoch": 1.0,
      "eval_bleu": 2.731416805620766,
      "eval_exact_match": 0.0,
      "eval_prefix_exact_match": 0.0,
      "step": 56
    },
    {
      "epoch": 1.07,
      "learning_rate": 3e-05,
      "loss": 2.2172,
      "step": 60
    },
    {
      "epoch": 1.18,
      "learning_rate": 3e-05,
      "loss": 1.9462,
      "step": 66
    },
    {
      "epoch": 1.29,
      "learning_rate": 3e-05,
      "loss": 1.8014,
      "step": 72
    },
    {
      "epoch": 1.39,
      "learning_rate": 3e-05,
      "loss": 1.8689,
      "step": 78
    },
    {
      "epoch": 1.5,
      "learning_rate": 3e-05,
      "loss": 1.9006,
      "step": 84
    },
    {
      "epoch": 1.5,
      "eval_accuracy": 0.6392953969397911,
      "eval_loss": 1.8079485893249512,
      "eval_runtime": 24.4889,
      "eval_samples_per_second": 36.057,
      "eval_steps_per_second": 2.287,
      "step": 84
    },
    {
      "epoch": 1.5,
      "eval_bleu": 3.957789284907739,
      "eval_exact_match": 0.0,
      "eval_prefix_exact_match": 0.0,
      "step": 84
    },
    {
      "epoch": 1.61,
      "learning_rate": 3e-05,
      "loss": 1.8975,
      "step": 90
    },
    {
      "epoch": 1.71,
      "learning_rate": 3e-05,
      "loss": 1.8936,
      "step": 96
    },
    {
      "epoch": 1.82,
      "learning_rate": 3e-05,
      "loss": 1.8677,
      "step": 102
    },
    {
      "epoch": 1.93,
      "learning_rate": 3e-05,
      "loss": 1.9317,
      "step": 108
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6722017129718149,
      "eval_loss": 1.5510306358337402,
      "eval_runtime": 24.2083,
      "eval_samples_per_second": 36.475,
      "eval_steps_per_second": 2.313,
      "step": 112
    },
    {
      "epoch": 2.0,
      "eval_bleu": 4.403977979732999,
      "eval_exact_match": 0.0,
      "eval_prefix_exact_match": 0.0,
      "step": 112
    },
    {
      "epoch": 2.04,
      "learning_rate": 3e-05,
      "loss": 1.7801,
      "step": 114
    },
    {
      "epoch": 2.14,
      "learning_rate": 3e-05,
      "loss": 1.3602,
      "step": 120
    },
    {
      "epoch": 2.25,
      "learning_rate": 3e-05,
      "loss": 1.3667,
      "step": 126
    },
    {
      "epoch": 2.36,
      "learning_rate": 3e-05,
      "loss": 1.3497,
      "step": 132
    },
    {
      "epoch": 2.46,
      "learning_rate": 3e-05,
      "loss": 1.3984,
      "step": 138
    },
    {
      "epoch": 2.5,
      "eval_accuracy": 0.707495599907831,
      "eval_loss": 1.284958839416504,
      "eval_runtime": 24.875,
      "eval_samples_per_second": 35.498,
      "eval_steps_per_second": 2.251,
      "step": 140
    },
    {
      "epoch": 2.5,
      "eval_bleu": 7.1028643857654,
      "eval_exact_match": 0.0,
      "eval_prefix_exact_match": 0.0,
      "step": 140
    },
    {
      "epoch": 2.57,
      "learning_rate": 3e-05,
      "loss": 1.3143,
      "step": 144
    },
    {
      "epoch": 2.68,
      "learning_rate": 3e-05,
      "loss": 1.3887,
      "step": 150
    },
    {
      "epoch": 2.79,
      "learning_rate": 3e-05,
      "loss": 1.3563,
      "step": 156
    },
    {
      "epoch": 2.89,
      "learning_rate": 3e-05,
      "loss": 1.3749,
      "step": 162
    },
    {
      "epoch": 3.0,
      "learning_rate": 3e-05,
      "loss": 1.3662,
      "step": 168
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.737386810999495,
      "eval_loss": 1.0899983644485474,
      "eval_runtime": 24.1441,
      "eval_samples_per_second": 36.572,
      "eval_steps_per_second": 2.319,
      "step": 168
    },
    {
      "epoch": 3.0,
      "eval_bleu": 8.181972004065782,
      "eval_exact_match": 0.0,
      "eval_prefix_exact_match": 0.0,
      "step": 168
    },
    {
      "epoch": 3.11,
      "learning_rate": 3e-05,
      "loss": 0.9654,
      "step": 174
    },
    {
      "epoch": 3.21,
      "learning_rate": 3e-05,
      "loss": 0.9704,
      "step": 180
    },
    {
      "epoch": 3.32,
      "learning_rate": 3e-05,
      "loss": 0.9269,
      "step": 186
    },
    {
      "epoch": 3.43,
      "learning_rate": 3e-05,
      "loss": 0.9041,
      "step": 192
    },
    {
      "epoch": 3.5,
      "eval_accuracy": 0.7670475994371804,
      "eval_loss": 0.890863835811615,
      "eval_runtime": 24.7388,
      "eval_samples_per_second": 35.693,
      "eval_steps_per_second": 2.264,
      "step": 196
    },
    {
      "epoch": 3.5,
      "eval_bleu": 12.34433922960259,
      "eval_exact_match": 0.0022650056625141564,
      "eval_prefix_exact_match": 0.0022650056625141564,
      "step": 196
    },
    {
      "epoch": 3.54,
      "learning_rate": 3e-05,
      "loss": 0.9277,
      "step": 198
    },
    {
      "epoch": 3.64,
      "learning_rate": 3e-05,
      "loss": 0.8827,
      "step": 204
    },
    {
      "epoch": 3.75,
      "learning_rate": 3e-05,
      "loss": 0.99,
      "step": 210
    },
    {
      "epoch": 3.86,
      "learning_rate": 3e-05,
      "loss": 0.9083,
      "step": 216
    },
    {
      "epoch": 3.96,
      "learning_rate": 3e-05,
      "loss": 0.9056,
      "step": 222
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7902908718310757,
      "eval_loss": 0.7502285242080688,
      "eval_runtime": 24.8951,
      "eval_samples_per_second": 35.469,
      "eval_steps_per_second": 2.249,
      "step": 224
    },
    {
      "epoch": 4.0,
      "eval_bleu": 14.357948484733997,
      "eval_exact_match": 0.0033975084937712344,
      "eval_prefix_exact_match": 0.0033975084937712344,
      "step": 224
    },
    {
      "epoch": 4.07,
      "learning_rate": 3e-05,
      "loss": 0.7462,
      "step": 228
    },
    {
      "epoch": 4.18,
      "learning_rate": 3e-05,
      "loss": 0.583,
      "step": 234
    },
    {
      "epoch": 4.29,
      "learning_rate": 3e-05,
      "loss": 0.5995,
      "step": 240
    },
    {
      "epoch": 4.39,
      "learning_rate": 3e-05,
      "loss": 0.6094,
      "step": 246
    },
    {
      "epoch": 4.5,
      "learning_rate": 3e-05,
      "loss": 0.6131,
      "step": 252
    },
    {
      "epoch": 4.5,
      "eval_accuracy": 0.8066557828732234,
      "eval_loss": 0.6304216980934143,
      "eval_runtime": 25.7927,
      "eval_samples_per_second": 34.234,
      "eval_steps_per_second": 2.171,
      "step": 252
    },
    {
      "epoch": 4.5,
      "eval_bleu": 21.42036295397234,
      "eval_exact_match": 0.010192525481313703,
      "eval_prefix_exact_match": 0.010192525481313703,
      "step": 252
    },
    {
      "epoch": 4.61,
      "learning_rate": 3e-05,
      "loss": 0.6447,
      "step": 258
    },
    {
      "epoch": 4.71,
      "learning_rate": 3e-05,
      "loss": 0.6134,
      "step": 264
    },
    {
      "epoch": 4.82,
      "learning_rate": 3e-05,
      "loss": 0.611,
      "step": 270
    },
    {
      "epoch": 4.93,
      "learning_rate": 3e-05,
      "loss": 0.6101,
      "step": 276
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.82152049535968,
      "eval_loss": 0.5429415106773376,
      "eval_runtime": 24.6601,
      "eval_samples_per_second": 35.807,
      "eval_steps_per_second": 2.271,
      "step": 280
    },
    {
      "epoch": 5.0,
      "eval_bleu": 26.261017838942095,
      "eval_exact_match": 0.02944507361268403,
      "eval_prefix_exact_match": 0.02944507361268403,
      "step": 280
    },
    {
      "epoch": 5.04,
      "learning_rate": 3e-05,
      "loss": 0.5416,
      "step": 282
    },
    {
      "epoch": 5.14,
      "learning_rate": 3e-05,
      "loss": 0.3793,
      "step": 288
    },
    {
      "epoch": 5.25,
      "learning_rate": 3e-05,
      "loss": 0.3858,
      "step": 294
    },
    {
      "epoch": 5.36,
      "learning_rate": 3e-05,
      "loss": 0.3977,
      "step": 300
    },
    {
      "epoch": 5.46,
      "learning_rate": 3e-05,
      "loss": 0.3772,
      "step": 306
    },
    {
      "epoch": 5.5,
      "eval_accuracy": 0.8287371367779068,
      "eval_loss": 0.48723191022872925,
      "eval_runtime": 25.4066,
      "eval_samples_per_second": 34.755,
      "eval_steps_per_second": 2.204,
      "step": 308
    },
    {
      "epoch": 5.5,
      "eval_bleu": 38.82587580024761,
      "eval_exact_match": 0.10419026047565119,
      "eval_prefix_exact_match": 0.10419026047565119,
      "step": 308
    },
    {
      "epoch": 5.57,
      "learning_rate": 3e-05,
      "loss": 0.3898,
      "step": 312
    },
    {
      "epoch": 5.68,
      "learning_rate": 3e-05,
      "loss": 0.3916,
      "step": 318
    },
    {
      "epoch": 5.79,
      "learning_rate": 3e-05,
      "loss": 0.3948,
      "step": 324
    },
    {
      "epoch": 5.89,
      "learning_rate": 3e-05,
      "loss": 0.4125,
      "step": 330
    },
    {
      "epoch": 6.0,
      "learning_rate": 3e-05,
      "loss": 0.4265,
      "step": 336
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8356841346648821,
      "eval_loss": 0.44368451833724976,
      "eval_runtime": 25.151,
      "eval_samples_per_second": 35.108,
      "eval_steps_per_second": 2.227,
      "step": 336
    },
    {
      "epoch": 6.0,
      "eval_bleu": 51.490761143615906,
      "eval_exact_match": 0.16987542468856173,
      "eval_prefix_exact_match": 0.16987542468856173,
      "step": 336
    },
    {
      "epoch": 6.11,
      "learning_rate": 3e-05,
      "loss": 0.2554,
      "step": 342
    },
    {
      "epoch": 6.21,
      "learning_rate": 3e-05,
      "loss": 0.2562,
      "step": 348
    },
    {
      "epoch": 6.32,
      "learning_rate": 3e-05,
      "loss": 0.2531,
      "step": 354
    },
    {
      "epoch": 6.43,
      "learning_rate": 3e-05,
      "loss": 0.2552,
      "step": 360
    },
    {
      "epoch": 6.5,
      "eval_accuracy": 0.8388953439916067,
      "eval_loss": 0.4225637912750244,
      "eval_runtime": 25.2976,
      "eval_samples_per_second": 34.905,
      "eval_steps_per_second": 2.214,
      "step": 364
    },
    {
      "epoch": 6.5,
      "eval_bleu": 66.56501737677819,
      "eval_exact_match": 0.3012457531143828,
      "eval_prefix_exact_match": 0.3012457531143828,
      "step": 364
    },
    {
      "epoch": 6.54,
      "learning_rate": 3e-05,
      "loss": 0.2536,
      "step": 366
    },
    {
      "epoch": 6.64,
      "learning_rate": 3e-05,
      "loss": 0.2617,
      "step": 372
    },
    {
      "epoch": 6.75,
      "learning_rate": 3e-05,
      "loss": 0.2763,
      "step": 378
    },
    {
      "epoch": 6.86,
      "learning_rate": 3e-05,
      "loss": 0.2614,
      "step": 384
    },
    {
      "epoch": 6.96,
      "learning_rate": 3e-05,
      "loss": 0.2875,
      "step": 390
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8417535654228746,
      "eval_loss": 0.4019472599029541,
      "eval_runtime": 24.4914,
      "eval_samples_per_second": 36.053,
      "eval_steps_per_second": 2.287,
      "step": 392
    },
    {
      "epoch": 7.0,
      "eval_bleu": 73.8359090816191,
      "eval_exact_match": 0.3941109852774632,
      "eval_prefix_exact_match": 0.3941109852774632,
      "step": 392
    },
    {
      "epoch": 7.07,
      "learning_rate": 3e-05,
      "loss": 0.2191,
      "step": 396
    },
    {
      "epoch": 7.18,
      "learning_rate": 3e-05,
      "loss": 0.178,
      "step": 402
    },
    {
      "epoch": 7.29,
      "learning_rate": 3e-05,
      "loss": 0.184,
      "step": 408
    },
    {
      "epoch": 7.39,
      "learning_rate": 3e-05,
      "loss": 0.1753,
      "step": 414
    },
    {
      "epoch": 7.5,
      "learning_rate": 3e-05,
      "loss": 0.1874,
      "step": 420
    },
    {
      "epoch": 7.5,
      "eval_accuracy": 0.8429596073990185,
      "eval_loss": 0.39652687311172485,
      "eval_runtime": 25.3956,
      "eval_samples_per_second": 34.77,
      "eval_steps_per_second": 2.205,
      "step": 420
    },
    {
      "epoch": 7.5,
      "eval_bleu": 82.2514512966405,
      "eval_exact_match": 0.5209513023782559,
      "eval_prefix_exact_match": 0.5209513023782559,
      "step": 420
    },
    {
      "epoch": 7.61,
      "learning_rate": 3e-05,
      "loss": 0.1977,
      "step": 426
    },
    {
      "epoch": 7.71,
      "learning_rate": 3e-05,
      "loss": 0.1922,
      "step": 432
    },
    {
      "epoch": 7.82,
      "learning_rate": 3e-05,
      "loss": 0.1977,
      "step": 438
    },
    {
      "epoch": 7.93,
      "learning_rate": 3e-05,
      "loss": 0.1958,
      "step": 444
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.844146038936526,
      "eval_loss": 0.3811873495578766,
      "eval_runtime": 25.8995,
      "eval_samples_per_second": 34.093,
      "eval_steps_per_second": 2.162,
      "step": 448
    },
    {
      "epoch": 8.0,
      "eval_bleu": 86.30069285909471,
      "eval_exact_match": 0.6002265005662514,
      "eval_prefix_exact_match": 0.6002265005662514,
      "step": 448
    },
    {
      "epoch": 8.04,
      "learning_rate": 3e-05,
      "loss": 0.1761,
      "step": 450
    },
    {
      "epoch": 8.14,
      "learning_rate": 3e-05,
      "loss": 0.1388,
      "step": 456
    },
    {
      "epoch": 8.25,
      "learning_rate": 3e-05,
      "loss": 0.1438,
      "step": 462
    },
    {
      "epoch": 8.36,
      "learning_rate": 3e-05,
      "loss": 0.143,
      "step": 468
    },
    {
      "epoch": 8.46,
      "learning_rate": 3e-05,
      "loss": 0.1443,
      "step": 474
    },
    {
      "epoch": 8.5,
      "eval_accuracy": 0.8449598721399401,
      "eval_loss": 0.38515910506248474,
      "eval_runtime": 25.4148,
      "eval_samples_per_second": 34.744,
      "eval_steps_per_second": 2.203,
      "step": 476
    },
    {
      "epoch": 8.5,
      "eval_bleu": 91.19128906028183,
      "eval_exact_match": 0.6772366930917327,
      "eval_prefix_exact_match": 0.6772366930917327,
      "step": 476
    },
    {
      "epoch": 8.57,
      "learning_rate": 3e-05,
      "loss": 0.1387,
      "step": 480
    },
    {
      "epoch": 8.68,
      "learning_rate": 3e-05,
      "loss": 0.146,
      "step": 486
    },
    {
      "epoch": 8.79,
      "learning_rate": 3e-05,
      "loss": 0.1487,
      "step": 492
    },
    {
      "epoch": 8.89,
      "learning_rate": 3e-05,
      "loss": 0.1477,
      "step": 498
    },
    {
      "epoch": 9.0,
      "learning_rate": 3e-05,
      "loss": 0.1535,
      "step": 504
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.8455825035666485,
      "eval_loss": 0.3791412115097046,
      "eval_runtime": 25.7361,
      "eval_samples_per_second": 34.31,
      "eval_steps_per_second": 2.176,
      "step": 504
    },
    {
      "epoch": 9.0,
      "eval_bleu": 93.80439056513916,
      "eval_exact_match": 0.7327293318233296,
      "eval_prefix_exact_match": 0.7327293318233296,
      "step": 504
    },
    {
      "epoch": 9.11,
      "learning_rate": 3e-05,
      "loss": 0.1186,
      "step": 510
    },
    {
      "epoch": 9.21,
      "learning_rate": 3e-05,
      "loss": 0.1124,
      "step": 516
    },
    {
      "epoch": 9.32,
      "learning_rate": 3e-05,
      "loss": 0.1165,
      "step": 522
    },
    {
      "epoch": 9.43,
      "learning_rate": 3e-05,
      "loss": 0.1236,
      "step": 528
    },
    {
      "epoch": 9.5,
      "eval_accuracy": 0.8456070166149441,
      "eval_loss": 0.38488179445266724,
      "eval_runtime": 25.6458,
      "eval_samples_per_second": 34.431,
      "eval_steps_per_second": 2.184,
      "step": 532
    },
    {
      "epoch": 9.5,
      "eval_bleu": 95.44029675299471,
      "eval_exact_match": 0.753114382785957,
      "eval_prefix_exact_match": 0.753114382785957,
      "step": 532
    },
    {
      "epoch": 9.54,
      "learning_rate": 3e-05,
      "loss": 0.1191,
      "step": 534
    },
    {
      "epoch": 9.64,
      "learning_rate": 3e-05,
      "loss": 0.124,
      "step": 540
    },
    {
      "epoch": 9.75,
      "learning_rate": 3e-05,
      "loss": 0.1172,
      "step": 546
    },
    {
      "epoch": 9.86,
      "learning_rate": 3e-05,
      "loss": 0.1247,
      "step": 552
    },
    {
      "epoch": 9.96,
      "learning_rate": 3e-05,
      "loss": 0.1221,
      "step": 558
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.8462492584802891,
      "eval_loss": 0.3696680963039398,
      "eval_runtime": 26.3353,
      "eval_samples_per_second": 33.529,
      "eval_steps_per_second": 2.126,
      "step": 560
    },
    {
      "epoch": 10.0,
      "eval_bleu": 95.90755211395843,
      "eval_exact_match": 0.8131370328425821,
      "eval_prefix_exact_match": 0.8131370328425821,
      "step": 560
    },
    {
      "epoch": 10.0,
      "step": 560,
      "total_flos": 1.36907799224832e+16,
      "train_loss": 0.8552157643118075,
      "train_runtime": 5673.7906,
      "train_samples_per_second": 1.556,
      "train_steps_per_second": 0.099
    }
  ],
  "logging_steps": 6,
  "max_steps": 560,
  "num_train_epochs": 10,
  "save_steps": 500,
  "total_flos": 1.36907799224832e+16,
  "trial_name": null,
  "trial_params": null
}