{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 4.761904761904762,
  "eval_steps": 50,
  "global_step": 600,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.1984126984126984,
      "grad_norm": 1.0922354459762573,
      "learning_rate": 9.523809523809523e-06,
      "loss": 0.8473,
      "step": 25
    },
    {
      "epoch": 0.3968253968253968,
      "grad_norm": 0.7849488854408264,
      "learning_rate": 1.9444444444444445e-05,
      "loss": 0.3692,
      "step": 50
    },
    {
      "epoch": 0.3968253968253968,
      "eval_loss": 0.3103943467140198,
      "eval_runtime": 10.1545,
      "eval_samples_per_second": 4.136,
      "eval_steps_per_second": 2.068,
      "step": 50
    },
    {
      "epoch": 0.5952380952380952,
      "grad_norm": 0.4042750298976898,
      "learning_rate": 2.9365079365079366e-05,
      "loss": 0.249,
      "step": 75
    },
    {
      "epoch": 0.7936507936507936,
      "grad_norm": 0.505721926689148,
      "learning_rate": 3.928571428571429e-05,
      "loss": 0.2248,
      "step": 100
    },
    {
      "epoch": 0.7936507936507936,
      "eval_loss": 0.23757970333099365,
      "eval_runtime": 10.1307,
      "eval_samples_per_second": 4.146,
      "eval_steps_per_second": 2.073,
      "step": 100
    },
    {
      "epoch": 0.9920634920634921,
      "grad_norm": 0.39118653535842896,
      "learning_rate": 4.9206349206349204e-05,
      "loss": 0.1997,
      "step": 125
    },
    {
      "epoch": 1.1904761904761905,
      "grad_norm": 0.4834806025028229,
      "learning_rate": 5.912698412698413e-05,
      "loss": 0.1717,
      "step": 150
    },
    {
      "epoch": 1.1904761904761905,
      "eval_loss": 0.20604592561721802,
      "eval_runtime": 10.1545,
      "eval_samples_per_second": 4.136,
      "eval_steps_per_second": 2.068,
      "step": 150
    },
    {
      "epoch": 1.3888888888888888,
      "grad_norm": 0.38224297761917114,
      "learning_rate": 6.904761904761905e-05,
      "loss": 0.1684,
      "step": 175
    },
    {
      "epoch": 1.5873015873015874,
      "grad_norm": 0.35443320870399475,
      "learning_rate": 7.896825396825397e-05,
      "loss": 0.1502,
      "step": 200
    },
    {
      "epoch": 1.5873015873015874,
      "eval_loss": 0.19332902133464813,
      "eval_runtime": 10.0977,
      "eval_samples_per_second": 4.159,
      "eval_steps_per_second": 2.08,
      "step": 200
    },
    {
      "epoch": 1.7857142857142856,
      "grad_norm": 0.3137013614177704,
      "learning_rate": 8.888888888888889e-05,
      "loss": 0.1549,
      "step": 225
    },
    {
      "epoch": 1.9841269841269842,
      "grad_norm": 0.30676764249801636,
      "learning_rate": 9.880952380952381e-05,
      "loss": 0.1449,
      "step": 250
    },
    {
      "epoch": 1.9841269841269842,
      "eval_loss": 0.17996199429035187,
      "eval_runtime": 10.0168,
      "eval_samples_per_second": 4.193,
      "eval_steps_per_second": 2.096,
      "step": 250
    },
    {
      "epoch": 2.1825396825396823,
      "grad_norm": 0.31375375390052795,
      "learning_rate": 9.997678517546382e-05,
      "loss": 0.1111,
      "step": 275
    },
    {
      "epoch": 2.380952380952381,
      "grad_norm": 0.2725541293621063,
      "learning_rate": 9.989407561073525e-05,
      "loss": 0.1016,
      "step": 300
    },
    {
      "epoch": 2.380952380952381,
      "eval_loss": 0.17762605845928192,
      "eval_runtime": 10.0289,
      "eval_samples_per_second": 4.188,
      "eval_steps_per_second": 2.094,
      "step": 300
    },
    {
      "epoch": 2.5793650793650795,
      "grad_norm": 0.31162211298942566,
      "learning_rate": 9.975153876827008e-05,
      "loss": 0.1044,
      "step": 325
    },
    {
      "epoch": 2.7777777777777777,
      "grad_norm": 0.2855686843395233,
      "learning_rate": 9.954934556197257e-05,
      "loss": 0.1087,
      "step": 350
    },
    {
      "epoch": 2.7777777777777777,
      "eval_loss": 0.1633986085653305,
      "eval_runtime": 10.056,
      "eval_samples_per_second": 4.177,
      "eval_steps_per_second": 2.088,
      "step": 350
    },
    {
      "epoch": 2.9761904761904763,
      "grad_norm": 0.33260759711265564,
      "learning_rate": 9.928773843884593e-05,
      "loss": 0.1028,
      "step": 375
    },
    {
      "epoch": 3.1746031746031744,
      "grad_norm": 0.3426005244255066,
      "learning_rate": 9.896703108827759e-05,
      "loss": 0.0725,
      "step": 400
    },
    {
      "epoch": 3.1746031746031744,
      "eval_loss": 0.17406687140464783,
      "eval_runtime": 10.0302,
      "eval_samples_per_second": 4.187,
      "eval_steps_per_second": 2.094,
      "step": 400
    },
    {
      "epoch": 3.373015873015873,
      "grad_norm": 0.3507950007915497,
      "learning_rate": 9.85876080658986e-05,
      "loss": 0.0698,
      "step": 425
    },
    {
      "epoch": 3.571428571428571,
      "grad_norm": 0.3167800009250641,
      "learning_rate": 9.814992433246858e-05,
      "loss": 0.0662,
      "step": 450
    },
    {
      "epoch": 3.571428571428571,
      "eval_loss": 0.17003558576107025,
      "eval_runtime": 10.035,
      "eval_samples_per_second": 4.185,
      "eval_steps_per_second": 2.093,
      "step": 450
    },
    {
      "epoch": 3.7698412698412698,
      "grad_norm": 0.31847792863845825,
      "learning_rate": 9.765450470833865e-05,
      "loss": 0.0724,
      "step": 475
    },
    {
      "epoch": 3.9682539682539684,
      "grad_norm": 0.3440268635749817,
      "learning_rate": 9.710194324414683e-05,
      "loss": 0.0736,
      "step": 500
    },
    {
      "epoch": 3.9682539682539684,
      "eval_loss": 0.1597495824098587,
      "eval_runtime": 10.0596,
      "eval_samples_per_second": 4.175,
      "eval_steps_per_second": 2.088,
      "step": 500
    },
    {
      "epoch": 4.166666666666667,
      "grad_norm": 0.20968282222747803,
      "learning_rate": 9.649290250850029e-05,
      "loss": 0.0507,
      "step": 525
    },
    {
      "epoch": 4.365079365079365,
      "grad_norm": 0.6211559772491455,
      "learning_rate": 9.582811279349882e-05,
      "loss": 0.0488,
      "step": 550
    },
    {
      "epoch": 4.365079365079365,
      "eval_loss": 0.1775166541337967,
      "eval_runtime": 10.0328,
      "eval_samples_per_second": 4.186,
      "eval_steps_per_second": 2.093,
      "step": 550
    },
    {
      "epoch": 4.563492063492063,
      "grad_norm": 0.2983364760875702,
      "learning_rate": 9.51083712390519e-05,
      "loss": 0.0519,
      "step": 575
    },
    {
      "epoch": 4.761904761904762,
      "grad_norm": 0.2705860733985901,
      "learning_rate": 9.433454087703954e-05,
      "loss": 0.0522,
      "step": 600
    },
    {
      "epoch": 4.761904761904762,
      "eval_loss": 0.1731569916009903,
      "eval_runtime": 10.0695,
      "eval_samples_per_second": 4.171,
      "eval_steps_per_second": 2.085,
      "step": 600
    }
  ],
  "logging_steps": 25,
  "max_steps": 2520,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 20,
  "save_steps": 50,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 3.627868424091448e+17,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}