{
  "best_metric": 4.538071632385254,
  "best_model_checkpoint": "/content/drive/MyDrive/Colab Notebooks/results_experiment_14/checkpoint-2790",
  "epoch": 30.0,
  "eval_steps": 500,
  "global_step": 2790,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 3060.651611328125,
      "learning_rate": 3.7200000000000004e-06,
      "loss": 49.2182,
      "step": 93
    },
    {
      "epoch": 1.0,
      "eval_loss": 51.180477142333984,
      "eval_runtime": 0.5704,
      "eval_samples_per_second": 122.714,
      "eval_steps_per_second": 21.037,
      "step": 93
    },
    {
      "epoch": 2.0,
      "grad_norm": 2896.443359375,
      "learning_rate": 7.440000000000001e-06,
      "loss": 47.2654,
      "step": 186
    },
    {
      "epoch": 2.0,
      "eval_loss": 45.42986297607422,
      "eval_runtime": 0.6949,
      "eval_samples_per_second": 100.731,
      "eval_steps_per_second": 17.268,
      "step": 186
    },
    {
      "epoch": 3.0,
      "grad_norm": 6535.3466796875,
      "learning_rate": 1.1160000000000002e-05,
      "loss": 43.654,
      "step": 279
    },
    {
      "epoch": 3.0,
      "eval_loss": 38.6143913269043,
      "eval_runtime": 0.665,
      "eval_samples_per_second": 105.257,
      "eval_steps_per_second": 18.044,
      "step": 279
    },
    {
      "epoch": 4.0,
      "grad_norm": 1967.9771728515625,
      "learning_rate": 1.4880000000000002e-05,
      "loss": 39.44,
      "step": 372
    },
    {
      "epoch": 4.0,
      "eval_loss": 33.054176330566406,
      "eval_runtime": 0.9286,
      "eval_samples_per_second": 75.379,
      "eval_steps_per_second": 12.922,
      "step": 372
    },
    {
      "epoch": 5.0,
      "grad_norm": 3827.974609375,
      "learning_rate": 1.86e-05,
      "loss": 35.2868,
      "step": 465
    },
    {
      "epoch": 5.0,
      "eval_loss": 26.869667053222656,
      "eval_runtime": 1.0053,
      "eval_samples_per_second": 69.631,
      "eval_steps_per_second": 11.937,
      "step": 465
    },
    {
      "epoch": 6.0,
      "grad_norm": 7205.50439453125,
      "learning_rate": 1.949344978165939e-05,
      "loss": 31.2026,
      "step": 558
    },
    {
      "epoch": 6.0,
      "eval_loss": 22.986032485961914,
      "eval_runtime": 0.6472,
      "eval_samples_per_second": 108.153,
      "eval_steps_per_second": 18.541,
      "step": 558
    },
    {
      "epoch": 7.0,
      "grad_norm": 956.697998046875,
      "learning_rate": 1.8681222707423585e-05,
      "loss": 27.7061,
      "step": 651
    },
    {
      "epoch": 7.0,
      "eval_loss": 20.32902717590332,
      "eval_runtime": 0.7108,
      "eval_samples_per_second": 98.487,
      "eval_steps_per_second": 16.883,
      "step": 651
    },
    {
      "epoch": 8.0,
      "grad_norm": 1620.7672119140625,
      "learning_rate": 1.7868995633187775e-05,
      "loss": 24.664,
      "step": 744
    },
    {
      "epoch": 8.0,
      "eval_loss": 18.136869430541992,
      "eval_runtime": 1.0986,
      "eval_samples_per_second": 63.72,
      "eval_steps_per_second": 10.923,
      "step": 744
    },
    {
      "epoch": 9.0,
      "grad_norm": 770.911376953125,
      "learning_rate": 1.7056768558951965e-05,
      "loss": 22.1086,
      "step": 837
    },
    {
      "epoch": 9.0,
      "eval_loss": 15.712793350219727,
      "eval_runtime": 0.6588,
      "eval_samples_per_second": 106.254,
      "eval_steps_per_second": 18.215,
      "step": 837
    },
    {
      "epoch": 10.0,
      "grad_norm": 2048.3359375,
      "learning_rate": 1.6244541484716158e-05,
      "loss": 19.4593,
      "step": 930
    },
    {
      "epoch": 10.0,
      "eval_loss": 14.192471504211426,
      "eval_runtime": 0.6664,
      "eval_samples_per_second": 105.044,
      "eval_steps_per_second": 18.007,
      "step": 930
    },
    {
      "epoch": 11.0,
      "grad_norm": 1505.369140625,
      "learning_rate": 1.543231441048035e-05,
      "loss": 17.9093,
      "step": 1023
    },
    {
      "epoch": 11.0,
      "eval_loss": 12.832782745361328,
      "eval_runtime": 0.9591,
      "eval_samples_per_second": 72.986,
      "eval_steps_per_second": 12.512,
      "step": 1023
    },
    {
      "epoch": 12.0,
      "grad_norm": 1106.5869140625,
      "learning_rate": 1.4620087336244542e-05,
      "loss": 16.2349,
      "step": 1116
    },
    {
      "epoch": 12.0,
      "eval_loss": 11.32009220123291,
      "eval_runtime": 1.0717,
      "eval_samples_per_second": 65.315,
      "eval_steps_per_second": 11.197,
      "step": 1116
    },
    {
      "epoch": 13.0,
      "grad_norm": 533.654296875,
      "learning_rate": 1.3807860262008735e-05,
      "loss": 14.9197,
      "step": 1209
    },
    {
      "epoch": 13.0,
      "eval_loss": 10.323843955993652,
      "eval_runtime": 0.6732,
      "eval_samples_per_second": 103.979,
      "eval_steps_per_second": 17.825,
      "step": 1209
    },
    {
      "epoch": 14.0,
      "grad_norm": 1263.10888671875,
      "learning_rate": 1.2995633187772927e-05,
      "loss": 13.5632,
      "step": 1302
    },
    {
      "epoch": 14.0,
      "eval_loss": 9.137203216552734,
      "eval_runtime": 0.647,
      "eval_samples_per_second": 108.192,
      "eval_steps_per_second": 18.547,
      "step": 1302
    },
    {
      "epoch": 15.0,
      "grad_norm": 925.8096313476562,
      "learning_rate": 1.218340611353712e-05,
      "loss": 12.4748,
      "step": 1395
    },
    {
      "epoch": 15.0,
      "eval_loss": 8.293523788452148,
      "eval_runtime": 0.6578,
      "eval_samples_per_second": 106.409,
      "eval_steps_per_second": 18.241,
      "step": 1395
    },
    {
      "epoch": 16.0,
      "grad_norm": 556.30517578125,
      "learning_rate": 1.137117903930131e-05,
      "loss": 11.4513,
      "step": 1488
    },
    {
      "epoch": 16.0,
      "eval_loss": 7.520947456359863,
      "eval_runtime": 0.9608,
      "eval_samples_per_second": 72.855,
      "eval_steps_per_second": 12.489,
      "step": 1488
    },
    {
      "epoch": 17.0,
      "grad_norm": 508.1446228027344,
      "learning_rate": 1.0558951965065502e-05,
      "loss": 10.4789,
      "step": 1581
    },
    {
      "epoch": 17.0,
      "eval_loss": 6.989750862121582,
      "eval_runtime": 0.6317,
      "eval_samples_per_second": 110.811,
      "eval_steps_per_second": 18.996,
      "step": 1581
    },
    {
      "epoch": 18.0,
      "grad_norm": 271.6485900878906,
      "learning_rate": 9.746724890829696e-06,
      "loss": 9.6762,
      "step": 1674
    },
    {
      "epoch": 18.0,
      "eval_loss": 6.519704341888428,
      "eval_runtime": 0.6795,
      "eval_samples_per_second": 103.018,
      "eval_steps_per_second": 17.66,
      "step": 1674
    },
    {
      "epoch": 19.0,
      "grad_norm": 672.8179321289062,
      "learning_rate": 8.934497816593887e-06,
      "loss": 9.1446,
      "step": 1767
    },
    {
      "epoch": 19.0,
      "eval_loss": 6.078932285308838,
      "eval_runtime": 0.6607,
      "eval_samples_per_second": 105.942,
      "eval_steps_per_second": 18.161,
      "step": 1767
    },
    {
      "epoch": 20.0,
      "grad_norm": 4875.142578125,
      "learning_rate": 8.122270742358079e-06,
      "loss": 8.6872,
      "step": 1860
    },
    {
      "epoch": 20.0,
      "eval_loss": 5.780891418457031,
      "eval_runtime": 1.0571,
      "eval_samples_per_second": 66.219,
      "eval_steps_per_second": 11.352,
      "step": 1860
    },
    {
      "epoch": 21.0,
      "grad_norm": 509.7615051269531,
      "learning_rate": 7.310043668122271e-06,
      "loss": 8.3194,
      "step": 1953
    },
    {
      "epoch": 21.0,
      "eval_loss": 5.519812107086182,
      "eval_runtime": 0.7323,
      "eval_samples_per_second": 95.586,
      "eval_steps_per_second": 16.386,
      "step": 1953
    },
    {
      "epoch": 22.0,
      "grad_norm": 804.4761352539062,
      "learning_rate": 6.497816593886463e-06,
      "loss": 7.9528,
      "step": 2046
    },
    {
      "epoch": 22.0,
      "eval_loss": 5.236818313598633,
      "eval_runtime": 0.7166,
      "eval_samples_per_second": 97.685,
      "eval_steps_per_second": 16.746,
      "step": 2046
    },
    {
      "epoch": 23.0,
      "grad_norm": 288.7013244628906,
      "learning_rate": 5.685589519650655e-06,
      "loss": 7.6692,
      "step": 2139
    },
    {
      "epoch": 23.0,
      "eval_loss": 5.120602130889893,
      "eval_runtime": 1.1643,
      "eval_samples_per_second": 60.121,
      "eval_steps_per_second": 10.307,
      "step": 2139
    },
    {
      "epoch": 24.0,
      "grad_norm": 359.6069030761719,
      "learning_rate": 4.873362445414848e-06,
      "loss": 7.2358,
      "step": 2232
    },
    {
      "epoch": 24.0,
      "eval_loss": 4.959656715393066,
      "eval_runtime": 0.7004,
      "eval_samples_per_second": 99.944,
      "eval_steps_per_second": 17.133,
      "step": 2232
    },
    {
      "epoch": 25.0,
      "grad_norm": 3070.00927734375,
      "learning_rate": 4.0611353711790395e-06,
      "loss": 7.1014,
      "step": 2325
    },
    {
      "epoch": 25.0,
      "eval_loss": 4.828472137451172,
      "eval_runtime": 0.6772,
      "eval_samples_per_second": 103.366,
      "eval_steps_per_second": 17.72,
      "step": 2325
    },
    {
      "epoch": 26.0,
      "grad_norm": 2893.34130859375,
      "learning_rate": 3.2489082969432317e-06,
      "loss": 6.8961,
      "step": 2418
    },
    {
      "epoch": 26.0,
      "eval_loss": 4.713611602783203,
      "eval_runtime": 0.6835,
      "eval_samples_per_second": 102.42,
      "eval_steps_per_second": 17.558,
      "step": 2418
    },
    {
      "epoch": 27.0,
      "grad_norm": 473.70867919921875,
      "learning_rate": 2.436681222707424e-06,
      "loss": 6.7949,
      "step": 2511
    },
    {
      "epoch": 27.0,
      "eval_loss": 4.646451950073242,
      "eval_runtime": 0.9181,
      "eval_samples_per_second": 76.241,
      "eval_steps_per_second": 13.07,
      "step": 2511
    },
    {
      "epoch": 28.0,
      "grad_norm": 9106.1611328125,
      "learning_rate": 1.6244541484716159e-06,
      "loss": 6.6925,
      "step": 2604
    },
    {
      "epoch": 28.0,
      "eval_loss": 4.57891845703125,
      "eval_runtime": 0.7287,
      "eval_samples_per_second": 96.065,
      "eval_steps_per_second": 16.468,
      "step": 2604
    },
    {
      "epoch": 29.0,
      "grad_norm": 362.14013671875,
      "learning_rate": 8.122270742358079e-07,
      "loss": 6.6963,
      "step": 2697
    },
    {
      "epoch": 29.0,
      "eval_loss": 4.548434257507324,
      "eval_runtime": 1.1489,
      "eval_samples_per_second": 60.928,
      "eval_steps_per_second": 10.445,
      "step": 2697
    },
    {
      "epoch": 30.0,
      "grad_norm": 415.389892578125,
      "learning_rate": 0.0,
      "loss": 6.6662,
      "step": 2790
    },
    {
      "epoch": 30.0,
      "eval_loss": 4.538071632385254,
      "eval_runtime": 1.0891,
      "eval_samples_per_second": 64.27,
      "eval_steps_per_second": 11.018,
      "step": 2790
    }
  ],
  "logging_steps": 500,
  "max_steps": 2790,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 30,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1052655951974400.0,
  "train_batch_size": 6,
  "trial_name": null,
  "trial_params": null
}