{
  "best_metric": 0.7214915752410889,
  "best_model_checkpoint": "./output/checkpoint-750",
  "epoch": 16.666666666666668,
  "eval_steps": 150,
  "global_step": 750,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.2222222222222222,
      "grad_norm": 1.9090766906738281,
      "learning_rate": 2.9999999999999984e-06,
      "loss": 0.662,
      "step": 10
    },
    {
      "epoch": 0.4444444444444444,
      "grad_norm": 1.888444185256958,
      "learning_rate": 5.999999999999997e-06,
      "loss": 0.7254,
      "step": 20
    },
    {
      "epoch": 0.6666666666666666,
      "grad_norm": 1.6710277795791626,
      "learning_rate": 8.999999999999993e-06,
      "loss": 0.7334,
      "step": 30
    },
    {
      "epoch": 0.8888888888888888,
      "grad_norm": 1.5766774415969849,
      "learning_rate": 1.1999999999999994e-05,
      "loss": 0.7369,
      "step": 40
    },
    {
      "epoch": 1.1111111111111112,
      "grad_norm": 1.2492249011993408,
      "learning_rate": 1.499999999999999e-05,
      "loss": 0.8941,
      "step": 50
    },
    {
      "epoch": 1.3333333333333333,
      "grad_norm": 1.1361218690872192,
      "learning_rate": 1.7999999999999987e-05,
      "loss": 0.7144,
      "step": 60
    },
    {
      "epoch": 1.5555555555555556,
      "grad_norm": 1.5108751058578491,
      "learning_rate": 2.0999999999999985e-05,
      "loss": 0.7581,
      "step": 70
    },
    {
      "epoch": 1.7777777777777777,
      "grad_norm": 1.873094081878662,
      "learning_rate": 2.3999999999999987e-05,
      "loss": 0.7072,
      "step": 80
    },
    {
      "epoch": 2.0,
      "grad_norm": 2.0551469326019287,
      "learning_rate": 2.6999999999999982e-05,
      "loss": 0.7699,
      "step": 90
    },
    {
      "epoch": 2.2222222222222223,
      "grad_norm": 1.5656124353408813,
      "learning_rate": 2.999999999999998e-05,
      "loss": 0.7272,
      "step": 100
    },
    {
      "epoch": 2.4444444444444446,
      "grad_norm": 1.7015224695205688,
      "learning_rate": 2.999999702723961e-05,
      "loss": 0.7469,
      "step": 110
    },
    {
      "epoch": 2.6666666666666665,
      "grad_norm": 2.7948575019836426,
      "learning_rate": 2.9999988108959667e-05,
      "loss": 0.57,
      "step": 120
    },
    {
      "epoch": 2.888888888888889,
      "grad_norm": 2.1129257678985596,
      "learning_rate": 2.9999973245163695e-05,
      "loss": 0.7045,
      "step": 130
    },
    {
      "epoch": 3.111111111111111,
      "grad_norm": 2.2446517944335938,
      "learning_rate": 2.999995243585758e-05,
      "loss": 0.7167,
      "step": 140
    },
    {
      "epoch": 3.3333333333333335,
      "grad_norm": 1.9295082092285156,
      "learning_rate": 2.9999925681049573e-05,
      "loss": 0.6639,
      "step": 150
    },
    {
      "epoch": 3.3333333333333335,
      "eval_loss": 0.7685600519180298,
      "eval_runtime": 0.4511,
      "eval_samples_per_second": 22.169,
      "eval_steps_per_second": 22.169,
      "step": 150
    },
    {
      "epoch": 3.5555555555555554,
      "grad_norm": 2.560298204421997,
      "learning_rate": 2.9999892980750276e-05,
      "loss": 0.6943,
      "step": 160
    },
    {
      "epoch": 3.7777777777777777,
      "grad_norm": 2.3921022415161133,
      "learning_rate": 2.9999854334972655e-05,
      "loss": 0.6927,
      "step": 170
    },
    {
      "epoch": 4.0,
      "grad_norm": 2.541400909423828,
      "learning_rate": 2.999980974373202e-05,
      "loss": 0.6681,
      "step": 180
    },
    {
      "epoch": 4.222222222222222,
      "grad_norm": 1.6048011779785156,
      "learning_rate": 2.9999759207046055e-05,
      "loss": 0.59,
      "step": 190
    },
    {
      "epoch": 4.444444444444445,
      "grad_norm": 1.6808319091796875,
      "learning_rate": 2.9999702724934783e-05,
      "loss": 0.7109,
      "step": 200
    },
    {
      "epoch": 4.666666666666667,
      "grad_norm": 3.4204533100128174,
      "learning_rate": 2.99996402974206e-05,
      "loss": 0.6277,
      "step": 210
    },
    {
      "epoch": 4.888888888888889,
      "grad_norm": 2.078854560852051,
      "learning_rate": 2.9999571924528243e-05,
      "loss": 0.6732,
      "step": 220
    },
    {
      "epoch": 5.111111111111111,
      "grad_norm": 2.6296238899230957,
      "learning_rate": 2.9999497606284816e-05,
      "loss": 0.6029,
      "step": 230
    },
    {
      "epoch": 5.333333333333333,
      "grad_norm": 1.4844911098480225,
      "learning_rate": 2.9999417342719775e-05,
      "loss": 0.6941,
      "step": 240
    },
    {
      "epoch": 5.555555555555555,
      "grad_norm": 2.2181289196014404,
      "learning_rate": 2.9999331133864935e-05,
      "loss": 0.6476,
      "step": 250
    },
    {
      "epoch": 5.777777777777778,
      "grad_norm": 1.705496907234192,
      "learning_rate": 2.9999238979754465e-05,
      "loss": 0.6095,
      "step": 260
    },
    {
      "epoch": 6.0,
      "grad_norm": 1.9006123542785645,
      "learning_rate": 2.99991408804249e-05,
      "loss": 0.5755,
      "step": 270
    },
    {
      "epoch": 6.222222222222222,
      "grad_norm": 2.1468276977539062,
      "learning_rate": 2.999903683591511e-05,
      "loss": 0.5742,
      "step": 280
    },
    {
      "epoch": 6.444444444444445,
      "grad_norm": 1.3946986198425293,
      "learning_rate": 2.9998926846266345e-05,
      "loss": 0.6137,
      "step": 290
    },
    {
      "epoch": 6.666666666666667,
      "grad_norm": 2.292116641998291,
      "learning_rate": 2.9998810911522193e-05,
      "loss": 0.6226,
      "step": 300
    },
    {
      "epoch": 6.666666666666667,
      "eval_loss": 0.7391407489776611,
      "eval_runtime": 0.5623,
      "eval_samples_per_second": 17.783,
      "eval_steps_per_second": 17.783,
      "step": 300
    },
    {
      "Start_State_loss": 0.7391407489776611,
      "Start_State_runtime": 0.5004,
      "Start_State_samples_per_second": 19.984,
      "Start_State_steps_per_second": 19.984,
      "epoch": 6.666666666666667,
      "step": 300
    },
    {
      "SWA_loss": 0.7391407489776611,
      "SWA_runtime": 0.5139,
      "SWA_samples_per_second": 19.46,
      "SWA_steps_per_second": 19.46,
      "epoch": 6.666666666666667,
      "step": 300
    },
    {
      "EMA_loss": 0.7391407489776611,
      "EMA_runtime": 0.515,
      "EMA_samples_per_second": 19.417,
      "EMA_steps_per_second": 19.417,
      "epoch": 6.666666666666667,
      "step": 300
    },
    {
      "epoch": 6.888888888888889,
      "grad_norm": 2.7107701301574707,
      "learning_rate": 2.9998689031728615e-05,
      "loss": 0.6481,
      "step": 310
    },
    {
      "epoch": 7.111111111111111,
      "grad_norm": 1.764064908027649,
      "learning_rate": 2.9998561206933918e-05,
      "loss": 0.5866,
      "step": 320
    },
    {
      "epoch": 7.333333333333333,
      "grad_norm": 1.7632637023925781,
      "learning_rate": 2.9998427437188766e-05,
      "loss": 0.5798,
      "step": 330
    },
    {
      "epoch": 7.555555555555555,
      "grad_norm": 2.3483335971832275,
      "learning_rate": 2.999828772254618e-05,
      "loss": 0.6034,
      "step": 340
    },
    {
      "epoch": 7.777777777777778,
      "grad_norm": 2.47190260887146,
      "learning_rate": 2.9998142063061544e-05,
      "loss": 0.662,
      "step": 350
    },
    {
      "epoch": 8.0,
      "grad_norm": 1.323142409324646,
      "learning_rate": 2.9997990458792583e-05,
      "loss": 0.6039,
      "step": 360
    },
    {
      "epoch": 8.222222222222221,
      "grad_norm": 1.9020463228225708,
      "learning_rate": 2.9997832909799397e-05,
      "loss": 0.5489,
      "step": 370
    },
    {
      "epoch": 8.444444444444445,
      "grad_norm": 1.9343500137329102,
      "learning_rate": 2.9997669416144432e-05,
      "loss": 0.641,
      "step": 380
    },
    {
      "epoch": 8.666666666666666,
      "grad_norm": 1.0505070686340332,
      "learning_rate": 2.999749997789249e-05,
      "loss": 0.5396,
      "step": 390
    },
    {
      "epoch": 8.88888888888889,
      "grad_norm": 1.5202258825302124,
      "learning_rate": 2.9997324595110723e-05,
      "loss": 0.6543,
      "step": 400
    },
    {
      "epoch": 9.11111111111111,
      "grad_norm": 1.384507656097412,
      "learning_rate": 2.9997143267868663e-05,
      "loss": 0.5946,
      "step": 410
    },
    {
      "epoch": 9.333333333333334,
      "grad_norm": 2.468230962753296,
      "learning_rate": 2.999695599623817e-05,
      "loss": 0.6231,
      "step": 420
    },
    {
      "epoch": 9.555555555555555,
      "grad_norm": 2.60021710395813,
      "learning_rate": 2.9996762780293483e-05,
      "loss": 0.575,
      "step": 430
    },
    {
      "epoch": 9.777777777777779,
      "grad_norm": 1.5323718786239624,
      "learning_rate": 2.9996563620111176e-05,
      "loss": 0.529,
      "step": 440
    },
    {
      "epoch": 10.0,
      "grad_norm": 1.3856033086776733,
      "learning_rate": 2.9996358515770198e-05,
      "loss": 0.5417,
      "step": 450
    },
    {
      "epoch": 10.0,
      "eval_loss": 0.7259252071380615,
      "eval_runtime": 0.4435,
      "eval_samples_per_second": 22.546,
      "eval_steps_per_second": 22.546,
      "step": 450
    },
    {
      "Start_State_loss": 0.7391407489776611,
      "Start_State_runtime": 0.4445,
      "Start_State_samples_per_second": 22.495,
      "Start_State_steps_per_second": 22.495,
      "epoch": 10.0,
      "step": 450
    },
    {
      "Raw_Model_loss": 0.7259252071380615,
      "Raw_Model_runtime": 0.4437,
      "Raw_Model_samples_per_second": 22.539,
      "Raw_Model_steps_per_second": 22.539,
      "epoch": 10.0,
      "step": 450
    },
    {
      "SWA_loss": 0.733832836151123,
      "SWA_runtime": 0.4445,
      "SWA_samples_per_second": 22.496,
      "SWA_steps_per_second": 22.496,
      "epoch": 10.0,
      "step": 450
    },
    {
      "EMA_loss": 0.738980233669281,
      "EMA_runtime": 0.4444,
      "EMA_samples_per_second": 22.503,
      "EMA_steps_per_second": 22.503,
      "epoch": 10.0,
      "step": 450
    },
    {
      "epoch": 10.222222222222221,
      "grad_norm": 2.220747470855713,
      "learning_rate": 2.9996147467351836e-05,
      "loss": 0.5056,
      "step": 460
    },
    {
      "epoch": 10.444444444444445,
      "grad_norm": 1.4205608367919922,
      "learning_rate": 2.9995930474939753e-05,
      "loss": 0.4901,
      "step": 470
    },
    {
      "epoch": 10.666666666666666,
      "grad_norm": 1.9306081533432007,
      "learning_rate": 2.9995707538619954e-05,
      "loss": 0.6361,
      "step": 480
    },
    {
      "epoch": 10.88888888888889,
      "grad_norm": 2.1457133293151855,
      "learning_rate": 2.9995478658480802e-05,
      "loss": 0.5528,
      "step": 490
    },
    {
      "epoch": 11.11111111111111,
      "grad_norm": 1.8677959442138672,
      "learning_rate": 2.9995243834613023e-05,
      "loss": 0.5233,
      "step": 500
    },
    {
      "epoch": 11.333333333333334,
      "grad_norm": 1.6708972454071045,
      "learning_rate": 2.9995003067109687e-05,
      "loss": 0.5387,
      "step": 510
    },
    {
      "epoch": 11.555555555555555,
      "grad_norm": 2.6434991359710693,
      "learning_rate": 2.9994756356066226e-05,
      "loss": 0.5847,
      "step": 520
    },
    {
      "epoch": 11.777777777777779,
      "grad_norm": 2.2601070404052734,
      "learning_rate": 2.999450370158044e-05,
      "loss": 0.5341,
      "step": 530
    },
    {
      "epoch": 12.0,
      "grad_norm": 1.5335863828659058,
      "learning_rate": 2.9994245103752457e-05,
      "loss": 0.5242,
      "step": 540
    },
    {
      "epoch": 12.222222222222221,
      "grad_norm": 1.2394074201583862,
      "learning_rate": 2.999398056268479e-05,
      "loss": 0.5356,
      "step": 550
    },
    {
      "epoch": 12.444444444444445,
      "grad_norm": 1.472650170326233,
      "learning_rate": 2.9993710078482286e-05,
      "loss": 0.415,
      "step": 560
    },
    {
      "epoch": 12.666666666666666,
      "grad_norm": 3.3844995498657227,
      "learning_rate": 2.9993433651252164e-05,
      "loss": 0.6192,
      "step": 570
    },
    {
      "epoch": 12.88888888888889,
      "grad_norm": 1.4811444282531738,
      "learning_rate": 2.9993151281103986e-05,
      "loss": 0.5351,
      "step": 580
    },
    {
      "epoch": 13.11111111111111,
      "grad_norm": 2.4430384635925293,
      "learning_rate": 2.9992862968149675e-05,
      "loss": 0.4177,
      "step": 590
    },
    {
      "epoch": 13.333333333333334,
      "grad_norm": 2.456298351287842,
      "learning_rate": 2.9992568712503513e-05,
      "loss": 0.5317,
      "step": 600
    },
    {
      "epoch": 13.333333333333334,
      "eval_loss": 0.7220126986503601,
      "eval_runtime": 0.5023,
      "eval_samples_per_second": 19.91,
      "eval_steps_per_second": 19.91,
      "step": 600
    },
    {
      "Start_State_loss": 0.7391407489776611,
      "Start_State_runtime": 0.4727,
      "Start_State_samples_per_second": 21.154,
      "Start_State_steps_per_second": 21.154,
      "epoch": 13.333333333333334,
      "step": 600
    },
    {
      "Raw_Model_loss": 0.7220126986503601,
      "Raw_Model_runtime": 0.5347,
      "Raw_Model_samples_per_second": 18.703,
      "Raw_Model_steps_per_second": 18.703,
      "epoch": 13.333333333333334,
      "step": 600
    },
    {
      "SWA_loss": 0.7282296419143677,
      "SWA_runtime": 0.5752,
      "SWA_samples_per_second": 17.384,
      "SWA_steps_per_second": 17.384,
      "epoch": 13.333333333333334,
      "step": 600
    },
    {
      "EMA_loss": 0.7385488748550415,
      "EMA_runtime": 0.5662,
      "EMA_samples_per_second": 17.661,
      "EMA_steps_per_second": 17.661,
      "epoch": 13.333333333333334,
      "step": 600
    },
    {
      "epoch": 13.555555555555555,
      "grad_norm": 2.3377010822296143,
      "learning_rate": 2.9992268514282122e-05,
      "loss": 0.565,
      "step": 610
    },
    {
      "epoch": 13.777777777777779,
      "grad_norm": 2.2196319103240967,
      "learning_rate": 2.99919623736045e-05,
      "loss": 0.441,
      "step": 620
    },
    {
      "epoch": 14.0,
      "grad_norm": 2.2767350673675537,
      "learning_rate": 2.9991650290591996e-05,
      "loss": 0.6033,
      "step": 630
    },
    {
      "epoch": 14.222222222222221,
      "grad_norm": 2.253643035888672,
      "learning_rate": 2.99913322653683e-05,
      "loss": 0.4925,
      "step": 640
    },
    {
      "epoch": 14.444444444444445,
      "grad_norm": 1.8424692153930664,
      "learning_rate": 2.9991008298059473e-05,
      "loss": 0.5007,
      "step": 650
    },
    {
      "epoch": 14.666666666666666,
      "grad_norm": 1.5401960611343384,
      "learning_rate": 2.9990678388793924e-05,
      "loss": 0.5318,
      "step": 660
    },
    {
      "epoch": 14.88888888888889,
      "grad_norm": 1.2824598550796509,
      "learning_rate": 2.999034253770242e-05,
      "loss": 0.4575,
      "step": 670
    },
    {
      "epoch": 15.11111111111111,
      "grad_norm": 2.5211098194122314,
      "learning_rate": 2.9990000744918076e-05,
      "loss": 0.449,
      "step": 680
    },
    {
      "epoch": 15.333333333333334,
      "grad_norm": 1.6035919189453125,
      "learning_rate": 2.9989653010576372e-05,
      "loss": 0.4529,
      "step": 690
    },
    {
      "epoch": 15.555555555555555,
      "grad_norm": 1.4720438718795776,
      "learning_rate": 2.9989299334815138e-05,
      "loss": 0.4804,
      "step": 700
    },
    {
      "epoch": 15.777777777777779,
      "grad_norm": 2.241570472717285,
      "learning_rate": 2.9988939717774558e-05,
      "loss": 0.524,
      "step": 710
    },
    {
      "epoch": 16.0,
      "grad_norm": 1.3463960886001587,
      "learning_rate": 2.9988574159597174e-05,
      "loss": 0.5105,
      "step": 720
    },
    {
      "epoch": 16.22222222222222,
      "grad_norm": 2.1436588764190674,
      "learning_rate": 2.9988202660427887e-05,
      "loss": 0.4644,
      "step": 730
    },
    {
      "epoch": 16.444444444444443,
      "grad_norm": 3.0679702758789062,
      "learning_rate": 2.9987825220413937e-05,
      "loss": 0.5349,
      "step": 740
    },
    {
      "epoch": 16.666666666666668,
      "grad_norm": 1.908082127571106,
      "learning_rate": 2.998744183970494e-05,
      "loss": 0.4263,
      "step": 750
    },
    {
      "epoch": 16.666666666666668,
      "eval_loss": 0.7214915752410889,
      "eval_runtime": 0.4155,
      "eval_samples_per_second": 24.069,
      "eval_steps_per_second": 24.069,
      "step": 750
    },
    {
      "Start_State_loss": 0.7391407489776611,
      "Start_State_runtime": 0.4325,
      "Start_State_samples_per_second": 23.123,
      "Start_State_steps_per_second": 23.123,
      "epoch": 16.666666666666668,
      "step": 750
    },
    {
      "Raw_Model_loss": 0.7214915752410889,
      "Raw_Model_runtime": 0.4148,
      "Raw_Model_samples_per_second": 24.108,
      "Raw_Model_steps_per_second": 24.108,
      "epoch": 16.666666666666668,
      "step": 750
    },
    {
      "SWA_loss": 0.7257974147796631,
      "SWA_runtime": 0.4246,
      "SWA_samples_per_second": 23.553,
      "SWA_steps_per_second": 23.553,
      "epoch": 16.666666666666668,
      "step": 750
    },
    {
      "EMA_loss": 0.7391572594642639,
      "EMA_runtime": 0.4233,
      "EMA_samples_per_second": 23.622,
      "EMA_steps_per_second": 23.622,
      "epoch": 16.666666666666668,
      "step": 750
    }
  ],
  "logging_steps": 10,
  "max_steps": 50000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1112,
  "save_steps": 150,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 7981049240027136.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}