{
  "best_global_step": 500,
  "best_metric": 0.3842231333255768,
  "best_model_checkpoint": null,
  "epoch": 100.0,
  "eval_steps": 500,
  "global_step": 5400,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.8518518518518519,
      "grad_norm": 7.101864814758301,
      "learning_rate": 9.900000000000002e-06,
      "loss": 0.8974,
      "step": 100
    },
    {
      "epoch": 3.7037037037037037,
      "grad_norm": 10.462117195129395,
      "learning_rate": 1.9900000000000003e-05,
      "loss": 0.4183,
      "step": 200
    },
    {
      "epoch": 5.555555555555555,
      "grad_norm": 0.6395331025123596,
      "learning_rate": 2.9900000000000002e-05,
      "loss": 0.109,
      "step": 300
    },
    {
      "epoch": 7.407407407407407,
      "grad_norm": 11.952120780944824,
      "learning_rate": 3.99e-05,
      "loss": 0.0507,
      "step": 400
    },
    {
      "epoch": 9.25925925925926,
      "grad_norm": 0.16501592099666595,
      "learning_rate": 4.99e-05,
      "loss": 0.0316,
      "step": 500
    },
    {
      "epoch": 9.25925925925926,
      "eval_loss": 0.3842231333255768,
      "eval_runtime": 0.2479,
      "eval_samples_per_second": 44.376,
      "eval_steps_per_second": 12.103,
      "step": 500
    },
    {
      "epoch": 11.11111111111111,
      "grad_norm": 0.6681220531463623,
      "learning_rate": 4.898979591836735e-05,
      "loss": 0.0549,
      "step": 600
    },
    {
      "epoch": 12.962962962962964,
      "grad_norm": 0.18139339983463287,
      "learning_rate": 4.796938775510204e-05,
      "loss": 0.0309,
      "step": 700
    },
    {
      "epoch": 14.814814814814815,
      "grad_norm": 0.1267923265695572,
      "learning_rate": 4.6948979591836736e-05,
      "loss": 0.023,
      "step": 800
    },
    {
      "epoch": 16.666666666666668,
      "grad_norm": 0.012054799124598503,
      "learning_rate": 4.592857142857143e-05,
      "loss": 0.0031,
      "step": 900
    },
    {
      "epoch": 18.51851851851852,
      "grad_norm": 0.014743717387318611,
      "learning_rate": 4.4908163265306125e-05,
      "loss": 0.026,
      "step": 1000
    },
    {
      "epoch": 18.51851851851852,
      "eval_loss": 0.4349963665008545,
      "eval_runtime": 0.1782,
      "eval_samples_per_second": 61.73,
      "eval_steps_per_second": 16.835,
      "step": 1000
    },
    {
      "epoch": 20.37037037037037,
      "grad_norm": 0.014977260492742062,
      "learning_rate": 4.388775510204082e-05,
      "loss": 0.001,
      "step": 1100
    },
    {
      "epoch": 22.22222222222222,
      "grad_norm": 0.008753976784646511,
      "learning_rate": 4.286734693877551e-05,
      "loss": 0.0004,
      "step": 1200
    },
    {
      "epoch": 24.074074074074073,
      "grad_norm": 0.014590133912861347,
      "learning_rate": 4.184693877551021e-05,
      "loss": 0.0005,
      "step": 1300
    },
    {
      "epoch": 25.925925925925927,
      "grad_norm": 0.03366304934024811,
      "learning_rate": 4.08265306122449e-05,
      "loss": 0.0001,
      "step": 1400
    },
    {
      "epoch": 27.77777777777778,
      "grad_norm": 0.015341528691351414,
      "learning_rate": 3.980612244897959e-05,
      "loss": 0.0001,
      "step": 1500
    },
    {
      "epoch": 27.77777777777778,
      "eval_loss": 0.5423622131347656,
      "eval_runtime": 0.1826,
      "eval_samples_per_second": 60.25,
      "eval_steps_per_second": 16.432,
      "step": 1500
    },
    {
      "epoch": 29.62962962962963,
      "grad_norm": 0.004611788783222437,
      "learning_rate": 3.8785714285714285e-05,
      "loss": 0.0001,
      "step": 1600
    },
    {
      "epoch": 31.48148148148148,
      "grad_norm": 0.0017561103450134397,
      "learning_rate": 3.776530612244898e-05,
      "loss": 0.0075,
      "step": 1700
    },
    {
      "epoch": 33.333333333333336,
      "grad_norm": 0.011370708234608173,
      "learning_rate": 3.674489795918368e-05,
      "loss": 0.0046,
      "step": 1800
    },
    {
      "epoch": 35.18518518518518,
      "grad_norm": 0.0015067611820995808,
      "learning_rate": 3.572448979591837e-05,
      "loss": 0.0001,
      "step": 1900
    },
    {
      "epoch": 37.03703703703704,
      "grad_norm": 0.0012736656935885549,
      "learning_rate": 3.470408163265306e-05,
      "loss": 0.0001,
      "step": 2000
    },
    {
      "epoch": 37.03703703703704,
      "eval_loss": 0.5318173766136169,
      "eval_runtime": 0.1851,
      "eval_samples_per_second": 59.427,
      "eval_steps_per_second": 16.207,
      "step": 2000
    },
    {
      "epoch": 38.888888888888886,
      "grad_norm": 0.0020575050730258226,
      "learning_rate": 3.3683673469387757e-05,
      "loss": 0.0001,
      "step": 2100
    },
    {
      "epoch": 40.74074074074074,
      "grad_norm": 0.002286889823153615,
      "learning_rate": 3.266326530612245e-05,
      "loss": 0.0007,
      "step": 2200
    },
    {
      "epoch": 42.592592592592595,
      "grad_norm": 0.011977535672485828,
      "learning_rate": 3.1642857142857145e-05,
      "loss": 0.0029,
      "step": 2300
    },
    {
      "epoch": 44.44444444444444,
      "grad_norm": 0.18020747601985931,
      "learning_rate": 3.062244897959184e-05,
      "loss": 0.0293,
      "step": 2400
    },
    {
      "epoch": 46.2962962962963,
      "grad_norm": 0.00249957456253469,
      "learning_rate": 2.960204081632653e-05,
      "loss": 0.0201,
      "step": 2500
    },
    {
      "epoch": 46.2962962962963,
      "eval_loss": 0.49513018131256104,
      "eval_runtime": 0.1867,
      "eval_samples_per_second": 58.916,
      "eval_steps_per_second": 16.068,
      "step": 2500
    },
    {
      "epoch": 48.148148148148145,
      "grad_norm": 0.0021083306055516005,
      "learning_rate": 2.8581632653061225e-05,
      "loss": 0.0052,
      "step": 2600
    },
    {
      "epoch": 50.0,
      "grad_norm": 0.006301419343799353,
      "learning_rate": 2.7561224489795916e-05,
      "loss": 0.0001,
      "step": 2700
    },
    {
      "epoch": 51.851851851851855,
      "grad_norm": 0.001658000866882503,
      "learning_rate": 2.6540816326530614e-05,
      "loss": 0.0001,
      "step": 2800
    },
    {
      "epoch": 53.7037037037037,
      "grad_norm": 0.0013888307148590684,
      "learning_rate": 2.5520408163265308e-05,
      "loss": 0.0018,
      "step": 2900
    },
    {
      "epoch": 55.55555555555556,
      "grad_norm": 0.059469956904649734,
      "learning_rate": 2.45e-05,
      "loss": 0.0001,
      "step": 3000
    },
    {
      "epoch": 55.55555555555556,
      "eval_loss": 0.5068791508674622,
      "eval_runtime": 0.1872,
      "eval_samples_per_second": 58.764,
      "eval_steps_per_second": 16.026,
      "step": 3000
    },
    {
      "epoch": 57.407407407407405,
      "grad_norm": 0.002175951609387994,
      "learning_rate": 2.3479591836734697e-05,
      "loss": 0.0001,
      "step": 3100
    },
    {
      "epoch": 59.25925925925926,
      "grad_norm": 0.0020811487920582294,
      "learning_rate": 2.2459183673469388e-05,
      "loss": 0.0,
      "step": 3200
    },
    {
      "epoch": 61.111111111111114,
      "grad_norm": 0.002122443402186036,
      "learning_rate": 2.1438775510204082e-05,
      "loss": 0.0114,
      "step": 3300
    },
    {
      "epoch": 62.96296296296296,
      "grad_norm": 0.00158906658180058,
      "learning_rate": 2.0418367346938777e-05,
      "loss": 0.0,
      "step": 3400
    },
    {
      "epoch": 64.81481481481481,
      "grad_norm": 0.0007862930651754141,
      "learning_rate": 1.939795918367347e-05,
      "loss": 0.0,
      "step": 3500
    },
    {
      "epoch": 64.81481481481481,
      "eval_loss": 0.5442617535591125,
      "eval_runtime": 0.1844,
      "eval_samples_per_second": 59.657,
      "eval_steps_per_second": 16.27,
      "step": 3500
    },
    {
      "epoch": 66.66666666666667,
      "grad_norm": 0.008862162008881569,
      "learning_rate": 1.8377551020408165e-05,
      "loss": 0.0005,
      "step": 3600
    },
    {
      "epoch": 68.51851851851852,
      "grad_norm": 0.0025289393961429596,
      "learning_rate": 1.7357142857142856e-05,
      "loss": 0.0078,
      "step": 3700
    },
    {
      "epoch": 70.37037037037037,
      "grad_norm": 0.0008086733869276941,
      "learning_rate": 1.633673469387755e-05,
      "loss": 0.0001,
      "step": 3800
    },
    {
      "epoch": 72.22222222222223,
      "grad_norm": 0.0024846054147928953,
      "learning_rate": 1.5316326530612245e-05,
      "loss": 0.0,
      "step": 3900
    },
    {
      "epoch": 74.07407407407408,
      "grad_norm": 0.001633179490454495,
      "learning_rate": 1.4295918367346938e-05,
      "loss": 0.0,
      "step": 4000
    },
    {
      "epoch": 74.07407407407408,
      "eval_loss": 0.5683199167251587,
      "eval_runtime": 0.1872,
      "eval_samples_per_second": 58.759,
      "eval_steps_per_second": 16.025,
      "step": 4000
    },
    {
      "epoch": 75.92592592592592,
      "grad_norm": 0.002557038329541683,
      "learning_rate": 1.3275510204081634e-05,
      "loss": 0.0,
      "step": 4100
    },
    {
      "epoch": 77.77777777777777,
      "grad_norm": 0.002890221541747451,
      "learning_rate": 1.2255102040816327e-05,
      "loss": 0.0,
      "step": 4200
    },
    {
      "epoch": 79.62962962962963,
      "grad_norm": 0.0013914318988099694,
      "learning_rate": 1.1234693877551021e-05,
      "loss": 0.0,
      "step": 4300
    },
    {
      "epoch": 81.48148148148148,
      "grad_norm": 0.0009156959713436663,
      "learning_rate": 1.0214285714285715e-05,
      "loss": 0.0,
      "step": 4400
    },
    {
      "epoch": 83.33333333333333,
      "grad_norm": 0.0012320117093622684,
      "learning_rate": 9.19387755102041e-06,
      "loss": 0.0,
      "step": 4500
    },
    {
      "epoch": 83.33333333333333,
      "eval_loss": 0.5750726461410522,
      "eval_runtime": 0.1871,
      "eval_samples_per_second": 58.787,
      "eval_steps_per_second": 16.033,
      "step": 4500
    },
    {
      "epoch": 85.18518518518519,
      "grad_norm": 0.000415924092521891,
      "learning_rate": 8.173469387755103e-06,
      "loss": 0.0,
      "step": 4600
    },
    {
      "epoch": 87.03703703703704,
      "grad_norm": 0.0011823369422927499,
      "learning_rate": 7.153061224489796e-06,
      "loss": 0.0,
      "step": 4700
    },
    {
      "epoch": 88.88888888888889,
      "grad_norm": 0.0014568202896043658,
      "learning_rate": 6.1326530612244905e-06,
      "loss": 0.0,
      "step": 4800
    },
    {
      "epoch": 90.74074074074075,
      "grad_norm": 0.0004411433474160731,
      "learning_rate": 5.112244897959184e-06,
      "loss": 0.0,
      "step": 4900
    },
    {
      "epoch": 92.5925925925926,
      "grad_norm": 0.0016104909591376781,
      "learning_rate": 4.0918367346938775e-06,
      "loss": 0.0,
      "step": 5000
    },
    {
      "epoch": 92.5925925925926,
      "eval_loss": 0.5782183408737183,
      "eval_runtime": 0.1887,
      "eval_samples_per_second": 58.296,
      "eval_steps_per_second": 15.899,
      "step": 5000
    },
    {
      "epoch": 94.44444444444444,
      "grad_norm": 0.0005303300567902625,
      "learning_rate": 3.0714285714285715e-06,
      "loss": 0.0,
      "step": 5100
    },
    {
      "epoch": 96.29629629629629,
      "grad_norm": 0.00046598046901635826,
      "learning_rate": 2.0510204081632654e-06,
      "loss": 0.0,
      "step": 5200
    },
    {
      "epoch": 98.14814814814815,
      "grad_norm": 0.0015102053293958306,
      "learning_rate": 1.0306122448979592e-06,
      "loss": 0.0,
      "step": 5300
    },
    {
      "epoch": 100.0,
      "grad_norm": 0.001883647171780467,
      "learning_rate": 1.0204081632653062e-08,
      "loss": 0.0,
      "step": 5400
    }
  ],
  "logging_steps": 100,
  "max_steps": 5400,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 100,
  "save_steps": 1000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 728814845952000.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}