{
  "best_global_step": 150,
  "best_metric": 0.34326156973838806,
  "best_model_checkpoint": "/content/output/checkpoint-150",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 375,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.13333333333333333,
      "grad_norm": 0.9156391620635986,
      "learning_rate": 1.7543859649122806e-05,
      "loss": 1.2171,
      "mean_token_accuracy": 0.7375583708286285,
      "num_tokens": 32413.0,
      "step": 10
    },
    {
      "epoch": 0.26666666666666666,
      "grad_norm": 0.815642774105072,
      "learning_rate": 3.508771929824561e-05,
      "loss": 0.9632,
      "mean_token_accuracy": 0.7655525088310242,
      "num_tokens": 64714.0,
      "step": 20
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.39404281973838806,
      "learning_rate": 5.2631578947368424e-05,
      "loss": 0.5951,
      "mean_token_accuracy": 0.8334747165441513,
      "num_tokens": 96352.0,
      "step": 30
    },
    {
      "epoch": 0.5333333333333333,
      "grad_norm": 0.41063451766967773,
      "learning_rate": 7.017543859649122e-05,
      "loss": 0.4305,
      "mean_token_accuracy": 0.8707121074199676,
      "num_tokens": 125717.0,
      "step": 40
    },
    {
      "epoch": 0.6666666666666666,
      "grad_norm": 0.3393928110599518,
      "learning_rate": 8.771929824561403e-05,
      "loss": 0.3874,
      "mean_token_accuracy": 0.8828857451677322,
      "num_tokens": 157446.0,
      "step": 50
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.2939395606517792,
      "learning_rate": 9.905660377358492e-05,
      "loss": 0.3613,
      "mean_token_accuracy": 0.8902347713708878,
      "num_tokens": 187947.0,
      "step": 60
    },
    {
      "epoch": 0.9333333333333333,
      "grad_norm": 0.30780646204948425,
      "learning_rate": 9.59119496855346e-05,
      "loss": 0.3545,
      "mean_token_accuracy": 0.8908359378576278,
      "num_tokens": 219528.0,
      "step": 70
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.35836032032966614,
      "eval_mean_token_accuracy": 0.8900894594192504,
      "eval_num_tokens": 234547.0,
      "eval_runtime": 47.5564,
      "eval_samples_per_second": 2.103,
      "eval_steps_per_second": 0.526,
      "step": 75
    },
    {
      "epoch": 1.0666666666666667,
      "grad_norm": 0.29364290833473206,
      "learning_rate": 9.276729559748428e-05,
      "loss": 0.3282,
      "mean_token_accuracy": 0.8956525355577469,
      "num_tokens": 249942.0,
      "step": 80
    },
    {
      "epoch": 1.2,
      "grad_norm": 0.3256814479827881,
      "learning_rate": 8.962264150943397e-05,
      "loss": 0.3033,
      "mean_token_accuracy": 0.9027387470006942,
      "num_tokens": 280660.0,
      "step": 90
    },
    {
      "epoch": 1.3333333333333333,
      "grad_norm": 0.3860395848751068,
      "learning_rate": 8.647798742138365e-05,
      "loss": 0.3139,
      "mean_token_accuracy": 0.9029639422893524,
      "num_tokens": 312269.0,
      "step": 100
    },
    {
      "epoch": 1.4666666666666668,
      "grad_norm": 0.35767263174057007,
      "learning_rate": 8.333333333333334e-05,
      "loss": 0.3188,
      "mean_token_accuracy": 0.9010519236326218,
      "num_tokens": 343451.0,
      "step": 110
    },
    {
      "epoch": 1.6,
      "grad_norm": 0.37099793553352356,
      "learning_rate": 8.018867924528302e-05,
      "loss": 0.3053,
      "mean_token_accuracy": 0.9043950855731964,
      "num_tokens": 375605.0,
      "step": 120
    },
    {
      "epoch": 1.7333333333333334,
      "grad_norm": 0.42942672967910767,
      "learning_rate": 7.704402515723272e-05,
      "loss": 0.2819,
      "mean_token_accuracy": 0.9102458357810974,
      "num_tokens": 407072.0,
      "step": 130
    },
    {
      "epoch": 1.8666666666666667,
      "grad_norm": 0.43511271476745605,
      "learning_rate": 7.389937106918238e-05,
      "loss": 0.2928,
      "mean_token_accuracy": 0.9080147057771683,
      "num_tokens": 439028.0,
      "step": 140
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.4053526818752289,
      "learning_rate": 7.075471698113208e-05,
      "loss": 0.2847,
      "mean_token_accuracy": 0.9118992000818252,
      "num_tokens": 469094.0,
      "step": 150
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.34326156973838806,
      "eval_mean_token_accuracy": 0.8937170529365539,
      "eval_num_tokens": 469094.0,
      "eval_runtime": 47.6074,
      "eval_samples_per_second": 2.101,
      "eval_steps_per_second": 0.525,
      "step": 150
    },
    {
      "epoch": 2.1333333333333333,
      "grad_norm": 0.47702592611312866,
      "learning_rate": 6.761006289308176e-05,
      "loss": 0.2118,
      "mean_token_accuracy": 0.9336899906396866,
      "num_tokens": 500956.0,
      "step": 160
    },
    {
      "epoch": 2.2666666666666666,
      "grad_norm": 0.5411467552185059,
      "learning_rate": 6.446540880503145e-05,
      "loss": 0.2049,
      "mean_token_accuracy": 0.9344588816165924,
      "num_tokens": 531097.0,
      "step": 170
    },
    {
      "epoch": 2.4,
      "grad_norm": 0.5084139704704285,
      "learning_rate": 6.132075471698113e-05,
      "loss": 0.218,
      "mean_token_accuracy": 0.930207297205925,
      "num_tokens": 562566.0,
      "step": 180
    },
    {
      "epoch": 2.533333333333333,
      "grad_norm": 0.5238709449768066,
      "learning_rate": 5.817610062893082e-05,
      "loss": 0.2164,
      "mean_token_accuracy": 0.930533480644226,
      "num_tokens": 594889.0,
      "step": 190
    },
    {
      "epoch": 2.6666666666666665,
      "grad_norm": 0.5833751559257507,
      "learning_rate": 5.503144654088051e-05,
      "loss": 0.212,
      "mean_token_accuracy": 0.9320303469896316,
      "num_tokens": 625913.0,
      "step": 200
    },
    {
      "epoch": 2.8,
      "grad_norm": 0.6267488598823547,
      "learning_rate": 5.188679245283019e-05,
      "loss": 0.1922,
      "mean_token_accuracy": 0.9386632025241852,
      "num_tokens": 658003.0,
      "step": 210
    },
    {
      "epoch": 2.9333333333333336,
      "grad_norm": 0.5038567781448364,
      "learning_rate": 4.8742138364779875e-05,
      "loss": 0.2032,
      "mean_token_accuracy": 0.9358770161867142,
      "num_tokens": 688971.0,
      "step": 220
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.35934150218963623,
      "eval_mean_token_accuracy": 0.8932444596290589,
      "eval_num_tokens": 703641.0,
      "eval_runtime": 47.592,
      "eval_samples_per_second": 2.101,
      "eval_steps_per_second": 0.525,
      "step": 225
    },
    {
      "epoch": 3.066666666666667,
      "grad_norm": 0.5663590431213379,
      "learning_rate": 4.559748427672956e-05,
      "loss": 0.1781,
      "mean_token_accuracy": 0.9423291176557541,
      "num_tokens": 720008.0,
      "step": 230
    },
    {
      "epoch": 3.2,
      "grad_norm": 0.6952011585235596,
      "learning_rate": 4.245283018867925e-05,
      "loss": 0.1373,
      "mean_token_accuracy": 0.9570749044418335,
      "num_tokens": 750286.0,
      "step": 240
    },
    {
      "epoch": 3.3333333333333335,
      "grad_norm": 0.6387498378753662,
      "learning_rate": 3.9308176100628936e-05,
      "loss": 0.1362,
      "mean_token_accuracy": 0.956162053346634,
      "num_tokens": 782161.0,
      "step": 250
    },
    {
      "epoch": 3.466666666666667,
      "grad_norm": 0.7637629508972168,
      "learning_rate": 3.6163522012578616e-05,
      "loss": 0.1207,
      "mean_token_accuracy": 0.9621568471193314,
      "num_tokens": 813190.0,
      "step": 260
    },
    {
      "epoch": 3.6,
      "grad_norm": 0.7092714309692383,
      "learning_rate": 3.30188679245283e-05,
      "loss": 0.1273,
      "mean_token_accuracy": 0.9595555752515793,
      "num_tokens": 844684.0,
      "step": 270
    },
    {
      "epoch": 3.7333333333333334,
      "grad_norm": 0.7261990904808044,
      "learning_rate": 2.9874213836477987e-05,
      "loss": 0.1218,
      "mean_token_accuracy": 0.9604292154312134,
      "num_tokens": 876896.0,
      "step": 280
    },
    {
      "epoch": 3.8666666666666667,
      "grad_norm": 0.7577289938926697,
      "learning_rate": 2.6729559748427674e-05,
      "loss": 0.1296,
      "mean_token_accuracy": 0.9597998350858689,
      "num_tokens": 907432.0,
      "step": 290
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.6601764559745789,
      "learning_rate": 2.358490566037736e-05,
      "loss": 0.1278,
      "mean_token_accuracy": 0.961115637421608,
      "num_tokens": 938188.0,
      "step": 300
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.4141911566257477,
      "eval_mean_token_accuracy": 0.8913846659660339,
      "eval_num_tokens": 938188.0,
      "eval_runtime": 47.5859,
      "eval_samples_per_second": 2.101,
      "eval_steps_per_second": 0.525,
      "step": 300
    },
    {
      "epoch": 4.133333333333334,
      "grad_norm": 0.8669018745422363,
      "learning_rate": 2.0440251572327044e-05,
      "loss": 0.081,
      "mean_token_accuracy": 0.9762738227844239,
      "num_tokens": 969846.0,
      "step": 310
    },
    {
      "epoch": 4.266666666666667,
      "grad_norm": 0.8257538080215454,
      "learning_rate": 1.7295597484276728e-05,
      "loss": 0.0816,
      "mean_token_accuracy": 0.9748408764600753,
      "num_tokens": 1001376.0,
      "step": 320
    },
    {
      "epoch": 4.4,
      "grad_norm": 0.7576424479484558,
      "learning_rate": 1.4150943396226415e-05,
      "loss": 0.0779,
      "mean_token_accuracy": 0.9772728830575943,
      "num_tokens": 1033904.0,
      "step": 330
    },
    {
      "epoch": 4.533333333333333,
      "grad_norm": 0.7902347445487976,
      "learning_rate": 1.1006289308176102e-05,
      "loss": 0.082,
      "mean_token_accuracy": 0.9764015346765518,
      "num_tokens": 1064634.0,
      "step": 340
    },
    {
      "epoch": 4.666666666666667,
      "grad_norm": 0.8522145748138428,
      "learning_rate": 7.861635220125786e-06,
      "loss": 0.0761,
      "mean_token_accuracy": 0.9762265920639038,
      "num_tokens": 1094465.0,
      "step": 350
    },
    {
      "epoch": 4.8,
      "grad_norm": 0.5950550436973572,
      "learning_rate": 4.716981132075472e-06,
      "loss": 0.0826,
      "mean_token_accuracy": 0.974476671218872,
      "num_tokens": 1125816.0,
      "step": 360
    },
    {
      "epoch": 4.933333333333334,
      "grad_norm": 0.8709597587585449,
      "learning_rate": 1.5723270440251573e-06,
      "loss": 0.0836,
      "mean_token_accuracy": 0.975111848115921,
      "num_tokens": 1157854.0,
      "step": 370
    },
    {
      "epoch": 5.0,
      "eval_loss": 0.47538331151008606,
      "eval_mean_token_accuracy": 0.8894158458709717,
      "eval_num_tokens": 1172735.0,
      "eval_runtime": 47.5654,
      "eval_samples_per_second": 2.102,
      "eval_steps_per_second": 0.526,
      "step": 375
    }
  ],
  "logging_steps": 10,
  "max_steps": 375,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "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": 6.547412120671027e+16,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}