{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 7.0,
  "eval_steps": 500,
  "global_step": 91,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.40540540540540543,
      "grad_norm": 2.9168423719016774,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 0.2887,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08512411266565323,
      "step": 5,
      "valid_targets_mean": 8782.1,
      "valid_targets_min": 1763
    },
    {
      "epoch": 0.8108108108108109,
      "grad_norm": 0.595660920768862,
      "learning_rate": 3.6e-05,
      "loss": 0.2156,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.06299532949924469,
      "step": 10,
      "valid_targets_mean": 5523.5,
      "valid_targets_min": 718
    },
    {
      "epoch": 1.1621621621621623,
      "grad_norm": 0.37712343506232265,
      "learning_rate": 3.9759796989536185e-05,
      "loss": 0.1737,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.050383515655994415,
      "step": 15,
      "valid_targets_mean": 5867.5,
      "valid_targets_min": 433
    },
    {
      "epoch": 1.5675675675675675,
      "grad_norm": 0.2347575723639623,
      "learning_rate": 3.879385241571817e-05,
      "loss": 0.1435,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.04610014706850052,
      "step": 20,
      "valid_targets_mean": 7495.9,
      "valid_targets_min": 1500
    },
    {
      "epoch": 1.972972972972973,
      "grad_norm": 0.16981735187794836,
      "learning_rate": 3.7123337990605335e-05,
      "loss": 0.1272,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.04025791585445404,
      "step": 25,
      "valid_targets_mean": 7071.2,
      "valid_targets_min": 342
    },
    {
      "epoch": 2.3243243243243246,
      "grad_norm": 0.21607756445297158,
      "learning_rate": 3.48108802621801e-05,
      "loss": 0.1199,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.038428571075201035,
      "step": 30,
      "valid_targets_mean": 6605.4,
      "valid_targets_min": 476
    },
    {
      "epoch": 2.72972972972973,
      "grad_norm": 0.1316432415711279,
      "learning_rate": 3.194317183405573e-05,
      "loss": 0.1151,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.03687749803066254,
      "step": 35,
      "valid_targets_mean": 6514.1,
      "valid_targets_min": 615
    },
    {
      "epoch": 3.081081081081081,
      "grad_norm": 0.1311970853547964,
      "learning_rate": 2.8627721313625073e-05,
      "loss": 0.1049,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.02935028076171875,
      "step": 40,
      "valid_targets_mean": 5418.0,
      "valid_targets_min": 1029
    },
    {
      "epoch": 3.4864864864864864,
      "grad_norm": 0.12539144867417315,
      "learning_rate": 2.4988822881159627e-05,
      "loss": 0.1049,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.03292456641793251,
      "step": 45,
      "valid_targets_mean": 6729.9,
      "valid_targets_min": 1521
    },
    {
      "epoch": 3.891891891891892,
      "grad_norm": 0.13127449105052028,
      "learning_rate": 2.1162896578209517e-05,
      "loss": 0.1008,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.030270356684923172,
      "step": 50,
      "valid_targets_mean": 6693.1,
      "valid_targets_min": 1410
    },
    {
      "epoch": 4.243243243243243,
      "grad_norm": 0.12876211465298387,
      "learning_rate": 1.7293374004997384e-05,
      "loss": 0.0981,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.03473636507987976,
      "step": 55,
      "valid_targets_mean": 7737.6,
      "valid_targets_min": 372
    },
    {
      "epoch": 4.648648648648649,
      "grad_norm": 0.12816877872251659,
      "learning_rate": 1.3525321158833582e-05,
      "loss": 0.0957,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.028843272477388382,
      "step": 60,
      "valid_targets_mean": 5910.2,
      "valid_targets_min": 802
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.23264567962716864,
      "learning_rate": 1.0000000000000006e-05,
      "loss": 0.0926,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08814641833305359,
      "step": 65,
      "valid_targets_mean": 6465.5,
      "valid_targets_min": 802
    },
    {
      "epoch": 5.405405405405405,
      "grad_norm": 0.13528023785947504,
      "learning_rate": 6.84957262861873e-06,
      "loss": 0.0928,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.036027610301971436,
      "step": 70,
      "valid_targets_mean": 7925.5,
      "valid_targets_min": 2353
    },
    {
      "epoch": 5.8108108108108105,
      "grad_norm": 0.13158309042631827,
      "learning_rate": 4.1921466096248164e-06,
      "loss": 0.0914,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.030884653329849243,
      "step": 75,
      "valid_targets_mean": 7174.2,
      "valid_targets_min": 1482
    },
    {
      "epoch": 6.162162162162162,
      "grad_norm": 0.12264132333565113,
      "learning_rate": 2.127347193531757e-06,
      "loss": 0.0936,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.030903231352567673,
      "step": 80,
      "valid_targets_mean": 6811.2,
      "valid_targets_min": 1310
    },
    {
      "epoch": 6.5675675675675675,
      "grad_norm": 0.11272346517044092,
      "learning_rate": 7.325824276823934e-07,
      "loss": 0.0893,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.03379923850297928,
      "step": 85,
      "valid_targets_mean": 7168.5,
      "valid_targets_min": 615
    },
    {
      "epoch": 6.972972972972973,
      "grad_norm": 0.23203422934347503,
      "learning_rate": 6.014117664415953e-08,
      "loss": 0.0919,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.02850460261106491,
      "step": 90,
      "valid_targets_mean": 6414.8,
      "valid_targets_min": 1121
    },
    {
      "epoch": 7.0,
      "step": 91,
      "total_flos": 4.487730344563835e+17,
      "train_loss": 0.0,
      "train_runtime": 0.9071,
      "train_samples_per_second": 8967.387,
      "train_steps_per_second": 100.324
    }
  ],
  "logging_steps": 5,
  "max_steps": 91,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "save_steps": 300,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 4.487730344563835e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}