{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 81,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.18867924528301888,
      "grad_norm": NaN,
      "learning_rate": 4.998119881260576e-05,
      "loss": 4.699731063842774,
      "num_input_tokens_seen": 91920,
      "step": 5,
      "train_runtime": 98.4322,
      "train_tokens_per_second": 933.841
    },
    {
      "epoch": 0.37735849056603776,
      "grad_norm": NaN,
      "learning_rate": 4.998119881260576e-05,
      "loss": 10.709168243408204,
      "num_input_tokens_seen": 191440,
      "step": 10,
      "train_runtime": 200.6741,
      "train_tokens_per_second": 953.985
    },
    {
      "epoch": 0.5660377358490566,
      "grad_norm": NaN,
      "learning_rate": 4.998119881260576e-05,
      "loss": 11.792630004882813,
      "num_input_tokens_seen": 290528,
      "step": 15,
      "train_runtime": 303.0055,
      "train_tokens_per_second": 958.821
    },
    {
      "epoch": 0.7547169811320755,
      "grad_norm": NaN,
      "learning_rate": 4.99248235291948e-05,
      "loss": 13.518409729003906,
      "num_input_tokens_seen": 384736,
      "step": 20,
      "train_runtime": 402.2728,
      "train_tokens_per_second": 956.406
    },
    {
      "epoch": 0.9433962264150944,
      "grad_norm": NaN,
      "learning_rate": 4.99248235291948e-05,
      "loss": 20.178372192382813,
      "num_input_tokens_seen": 488112,
      "step": 25,
      "train_runtime": 508.669,
      "train_tokens_per_second": 959.587
    },
    {
      "epoch": 1.1132075471698113,
      "grad_norm": NaN,
      "learning_rate": 4.983095894354858e-05,
      "loss": 7.201190948486328,
      "num_input_tokens_seen": 584008,
      "step": 30,
      "train_runtime": 612.5705,
      "train_tokens_per_second": 953.373
    },
    {
      "epoch": 1.3018867924528301,
      "grad_norm": NaN,
      "learning_rate": 4.983095894354858e-05,
      "loss": 16.000038146972656,
      "num_input_tokens_seen": 687112,
      "step": 35,
      "train_runtime": 721.4375,
      "train_tokens_per_second": 952.421
    },
    {
      "epoch": 1.490566037735849,
      "grad_norm": NaN,
      "learning_rate": 4.983095894354858e-05,
      "loss": 9.682757568359374,
      "num_input_tokens_seen": 793608,
      "step": 40,
      "train_runtime": 833.895,
      "train_tokens_per_second": 951.688
    },
    {
      "epoch": 1.6792452830188678,
      "grad_norm": 0.0,
      "learning_rate": 4.983095894354858e-05,
      "loss": 6.101836395263672,
      "num_input_tokens_seen": 902728,
      "step": 45,
      "train_runtime": 949.9633,
      "train_tokens_per_second": 950.277
    },
    {
      "epoch": 1.8679245283018868,
      "grad_norm": NaN,
      "learning_rate": 4.969974623692023e-05,
      "loss": 10.295503997802735,
      "num_input_tokens_seen": 992696,
      "step": 50,
      "train_runtime": 1041.5791,
      "train_tokens_per_second": 953.068
    },
    {
      "epoch": 2.0377358490566038,
      "grad_norm": NaN,
      "learning_rate": 4.9326121764495596e-05,
      "loss": 4.089427185058594,
      "num_input_tokens_seen": 1091000,
      "step": 55,
      "train_runtime": 1146.5611,
      "train_tokens_per_second": 951.541
    },
    {
      "epoch": 2.2264150943396226,
      "grad_norm": NaN,
      "learning_rate": 4.9326121764495596e-05,
      "loss": 26.63399658203125,
      "num_input_tokens_seen": 1190312,
      "step": 60,
      "train_runtime": 1249.7851,
      "train_tokens_per_second": 952.413
    },
    {
      "epoch": 2.4150943396226414,
      "grad_norm": NaN,
      "learning_rate": 4.9084271965397014e-05,
      "loss": 5.875924301147461,
      "num_input_tokens_seen": 1291288,
      "step": 65,
      "train_runtime": 1358.4058,
      "train_tokens_per_second": 950.591
    },
    {
      "epoch": 2.6037735849056602,
      "grad_norm": NaN,
      "learning_rate": 4.9084271965397014e-05,
      "loss": 4.943584823608399,
      "num_input_tokens_seen": 1398264,
      "step": 70,
      "train_runtime": 1475.5209,
      "train_tokens_per_second": 947.641
    },
    {
      "epoch": 2.7924528301886795,
      "grad_norm": NaN,
      "learning_rate": 4.9084271965397014e-05,
      "loss": 11.34936752319336,
      "num_input_tokens_seen": 1502360,
      "step": 75,
      "train_runtime": 1584.2311,
      "train_tokens_per_second": 948.321
    },
    {
      "epoch": 2.981132075471698,
      "grad_norm": NaN,
      "learning_rate": 4.849231551964771e-05,
      "loss": 3.3955615997314452,
      "num_input_tokens_seen": 1586952,
      "step": 80,
      "train_runtime": 1668.9061,
      "train_tokens_per_second": 950.894
    },
    {
      "epoch": 3.0,
      "num_input_tokens_seen": 1595528,
      "step": 81,
      "total_flos": 3053620026209280.0,
      "train_loss": 10.295250705730767,
      "train_runtime": 1677.9647,
      "train_samples_per_second": 0.756,
      "train_steps_per_second": 0.048
    }
  ],
  "logging_steps": 5,
  "max_steps": 81,
  "num_input_tokens_seen": 1595528,
  "num_train_epochs": 3,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 3053620026209280.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}