{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 100,
  "global_step": 576,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.05215123859191656,
      "grad_norm": 0.8625667095184326,
      "learning_rate": 1.5517241379310346e-05,
      "loss": 1.0726,
      "step": 10
    },
    {
      "epoch": 0.10430247718383312,
      "grad_norm": 0.5288952589035034,
      "learning_rate": 3.275862068965517e-05,
      "loss": 0.7118,
      "step": 20
    },
    {
      "epoch": 0.15645371577574968,
      "grad_norm": 0.33996039628982544,
      "learning_rate": 5e-05,
      "loss": 0.5481,
      "step": 30
    },
    {
      "epoch": 0.20860495436766624,
      "grad_norm": 0.3226863443851471,
      "learning_rate": 6.724137931034483e-05,
      "loss": 0.4628,
      "step": 40
    },
    {
      "epoch": 0.2607561929595828,
      "grad_norm": 0.2945227324962616,
      "learning_rate": 8.448275862068966e-05,
      "loss": 0.4178,
      "step": 50
    },
    {
      "epoch": 0.31290743155149936,
      "grad_norm": 0.29200270771980286,
      "learning_rate": 9.999908044247358e-05,
      "loss": 0.398,
      "step": 60
    },
    {
      "epoch": 0.3650586701434159,
      "grad_norm": 0.28538617491722107,
      "learning_rate": 9.98887744597968e-05,
      "loss": 0.3707,
      "step": 70
    },
    {
      "epoch": 0.4172099087353325,
      "grad_norm": 0.2749938666820526,
      "learning_rate": 9.959502176294383e-05,
      "loss": 0.3726,
      "step": 80
    },
    {
      "epoch": 0.46936114732724904,
      "grad_norm": 0.3119683861732483,
      "learning_rate": 9.911890251408751e-05,
      "loss": 0.3612,
      "step": 90
    },
    {
      "epoch": 0.5215123859191656,
      "grad_norm": 0.28559353947639465,
      "learning_rate": 9.846216745801365e-05,
      "loss": 0.3491,
      "step": 100
    },
    {
      "epoch": 0.5215123859191656,
      "eval_loss": 0.34256601333618164,
      "eval_runtime": 176.5162,
      "eval_samples_per_second": 3.864,
      "eval_steps_per_second": 0.969,
      "step": 100
    },
    {
      "epoch": 0.5736636245110821,
      "grad_norm": 0.28374797105789185,
      "learning_rate": 9.762723148443296e-05,
      "loss": 0.3511,
      "step": 110
    },
    {
      "epoch": 0.6258148631029987,
      "grad_norm": 0.2709859013557434,
      "learning_rate": 9.661716474815597e-05,
      "loss": 0.3381,
      "step": 120
    },
    {
      "epoch": 0.6779661016949152,
      "grad_norm": 0.2764575183391571,
      "learning_rate": 9.543568137978372e-05,
      "loss": 0.3416,
      "step": 130
    },
    {
      "epoch": 0.7301173402868318,
      "grad_norm": 0.2733752727508545,
      "learning_rate": 9.408712582842583e-05,
      "loss": 0.3346,
      "step": 140
    },
    {
      "epoch": 0.7822685788787483,
      "grad_norm": 0.263992041349411,
      "learning_rate": 9.257645688666556e-05,
      "loss": 0.3294,
      "step": 150
    },
    {
      "epoch": 0.834419817470665,
      "grad_norm": 0.26761433482170105,
      "learning_rate": 9.090922945651399e-05,
      "loss": 0.3244,
      "step": 160
    },
    {
      "epoch": 0.8865710560625815,
      "grad_norm": 0.28852951526641846,
      "learning_rate": 8.90915741234015e-05,
      "loss": 0.3245,
      "step": 170
    },
    {
      "epoch": 0.9387222946544981,
      "grad_norm": 0.2724473178386688,
      "learning_rate": 8.713017461331608e-05,
      "loss": 0.323,
      "step": 180
    },
    {
      "epoch": 0.9908735332464146,
      "grad_norm": 0.2559918463230133,
      "learning_rate": 8.503224321598035e-05,
      "loss": 0.3142,
      "step": 190
    },
    {
      "epoch": 1.0417209908735332,
      "grad_norm": 0.25570759177207947,
      "learning_rate": 8.28054942644397e-05,
      "loss": 0.277,
      "step": 200
    },
    {
      "epoch": 1.0417209908735332,
      "eval_loss": 0.3157625198364258,
      "eval_runtime": 200.113,
      "eval_samples_per_second": 3.408,
      "eval_steps_per_second": 0.855,
      "step": 200
    },
    {
      "epoch": 1.0938722294654497,
      "grad_norm": 0.25102606415748596,
      "learning_rate": 8.04581157685797e-05,
      "loss": 0.2566,
      "step": 210
    },
    {
      "epoch": 1.1460234680573664,
      "grad_norm": 0.2610400915145874,
      "learning_rate": 7.799873930687978e-05,
      "loss": 0.2514,
      "step": 220
    },
    {
      "epoch": 1.198174706649283,
      "grad_norm": 0.26567086577415466,
      "learning_rate": 7.543640828711466e-05,
      "loss": 0.2559,
      "step": 230
    },
    {
      "epoch": 1.2503259452411994,
      "grad_norm": 0.2972986400127411,
      "learning_rate": 7.278054469271245e-05,
      "loss": 0.2606,
      "step": 240
    },
    {
      "epoch": 1.302477183833116,
      "grad_norm": 0.2643340229988098,
      "learning_rate": 7.004091443704681e-05,
      "loss": 0.2535,
      "step": 250
    },
    {
      "epoch": 1.3546284224250327,
      "grad_norm": 0.2536676228046417,
      "learning_rate": 6.72275914530596e-05,
      "loss": 0.2584,
      "step": 260
    },
    {
      "epoch": 1.4067796610169492,
      "grad_norm": 0.26991742849349976,
      "learning_rate": 6.435092065026035e-05,
      "loss": 0.2515,
      "step": 270
    },
    {
      "epoch": 1.4589308996088657,
      "grad_norm": 0.2521091401576996,
      "learning_rate": 6.142147987531407e-05,
      "loss": 0.2516,
      "step": 280
    },
    {
      "epoch": 1.5110821382007824,
      "grad_norm": 0.2592126727104187,
      "learning_rate": 5.8450041016092464e-05,
      "loss": 0.2496,
      "step": 290
    },
    {
      "epoch": 1.5632333767926987,
      "grad_norm": 0.28268447518348694,
      "learning_rate": 5.544753039221354e-05,
      "loss": 0.246,
      "step": 300
    },
    {
      "epoch": 1.5632333767926987,
      "eval_loss": 0.29918840527534485,
      "eval_runtime": 175.5784,
      "eval_samples_per_second": 3.884,
      "eval_steps_per_second": 0.974,
      "step": 300
    },
    {
      "epoch": 1.6153846153846154,
      "grad_norm": 0.24876336753368378,
      "learning_rate": 5.242498857771816e-05,
      "loss": 0.2531,
      "step": 310
    },
    {
      "epoch": 1.667535853976532,
      "grad_norm": 0.27175015211105347,
      "learning_rate": 4.939352980361985e-05,
      "loss": 0.2471,
      "step": 320
    },
    {
      "epoch": 1.7196870925684484,
      "grad_norm": 0.2706572115421295,
      "learning_rate": 4.636430108960975e-05,
      "loss": 0.2484,
      "step": 330
    },
    {
      "epoch": 1.7718383311603652,
      "grad_norm": 0.25022950768470764,
      "learning_rate": 4.3348441255193624e-05,
      "loss": 0.2455,
      "step": 340
    },
    {
      "epoch": 1.8239895697522817,
      "grad_norm": 0.28196290135383606,
      "learning_rate": 4.035703996098214e-05,
      "loss": 0.2399,
      "step": 350
    },
    {
      "epoch": 1.8761408083441982,
      "grad_norm": 0.24447277188301086,
      "learning_rate": 3.740109693074375e-05,
      "loss": 0.2447,
      "step": 360
    },
    {
      "epoch": 1.928292046936115,
      "grad_norm": 0.26620620489120483,
      "learning_rate": 3.44914815041658e-05,
      "loss": 0.2382,
      "step": 370
    },
    {
      "epoch": 1.9804432855280312,
      "grad_norm": 0.27352651953697205,
      "learning_rate": 3.163889266905272e-05,
      "loss": 0.2346,
      "step": 380
    },
    {
      "epoch": 2.03129074315515,
      "grad_norm": 0.22145593166351318,
      "learning_rate": 2.8853819719927434e-05,
      "loss": 0.2004,
      "step": 390
    },
    {
      "epoch": 2.0834419817470664,
      "grad_norm": 0.2577626407146454,
      "learning_rate": 2.6146503687699002e-05,
      "loss": 0.1794,
      "step": 400
    },
    {
      "epoch": 2.0834419817470664,
      "eval_loss": 0.3007373809814453,
      "eval_runtime": 175.4648,
      "eval_samples_per_second": 3.887,
      "eval_steps_per_second": 0.975,
      "step": 400
    },
    {
      "epoch": 2.135593220338983,
      "grad_norm": 0.2457975298166275,
      "learning_rate": 2.3526899682223813e-05,
      "loss": 0.1763,
      "step": 410
    },
    {
      "epoch": 2.1877444589308994,
      "grad_norm": 0.22658194601535797,
      "learning_rate": 2.1004640286231198e-05,
      "loss": 0.1775,
      "step": 420
    },
    {
      "epoch": 2.239895697522816,
      "grad_norm": 0.25427886843681335,
      "learning_rate": 1.858900013521788e-05,
      "loss": 0.1755,
      "step": 430
    },
    {
      "epoch": 2.292046936114733,
      "grad_norm": 0.2568660378456116,
      "learning_rate": 1.628886181355551e-05,
      "loss": 0.1734,
      "step": 440
    },
    {
      "epoch": 2.344198174706649,
      "grad_norm": 0.2692377269268036,
      "learning_rate": 1.4112683192214598e-05,
      "loss": 0.175,
      "step": 450
    },
    {
      "epoch": 2.396349413298566,
      "grad_norm": 0.24048754572868347,
      "learning_rate": 1.2068466328208367e-05,
      "loss": 0.1764,
      "step": 460
    },
    {
      "epoch": 2.448500651890482,
      "grad_norm": 0.26973435282707214,
      "learning_rate": 1.016372804011625e-05,
      "loss": 0.1742,
      "step": 470
    },
    {
      "epoch": 2.500651890482399,
      "grad_norm": 0.25898903608322144,
      "learning_rate": 8.40547226788464e-06,
      "loss": 0.1769,
      "step": 480
    },
    {
      "epoch": 2.5528031290743156,
      "grad_norm": 0.2492300420999527,
      "learning_rate": 6.800164318540836e-06,
      "loss": 0.1758,
      "step": 490
    },
    {
      "epoch": 2.604954367666232,
      "grad_norm": 0.2548163831233978,
      "learning_rate": 5.353707092521582e-06,
      "loss": 0.1734,
      "step": 500
    },
    {
      "epoch": 2.604954367666232,
      "eval_loss": 0.2975265383720398,
      "eval_runtime": 174.2803,
      "eval_samples_per_second": 3.913,
      "eval_steps_per_second": 0.981,
      "step": 500
    },
    {
      "epoch": 2.6571056062581486,
      "grad_norm": 0.27377912402153015,
      "learning_rate": 4.071419378034896e-06,
      "loss": 0.1723,
      "step": 510
    },
    {
      "epoch": 2.7092568448500653,
      "grad_norm": 0.24549677968025208,
      "learning_rate": 2.9580162932691723e-06,
      "loss": 0.1701,
      "step": 520
    },
    {
      "epoch": 2.7614080834419816,
      "grad_norm": 0.2616007328033447,
      "learning_rate": 2.0175919483657213e-06,
      "loss": 0.1685,
      "step": 530
    },
    {
      "epoch": 2.8135593220338984,
      "grad_norm": 0.2631126344203949,
      "learning_rate": 1.2536043909088191e-06,
      "loss": 0.1757,
      "step": 540
    },
    {
      "epoch": 2.865710560625815,
      "grad_norm": 0.25270023941993713,
      "learning_rate": 6.688628902903393e-07,
      "loss": 0.1758,
      "step": 550
    },
    {
      "epoch": 2.9178617992177314,
      "grad_norm": 0.2603110373020172,
      "learning_rate": 2.6551760770569535e-07,
      "loss": 0.1712,
      "step": 560
    },
    {
      "epoch": 2.970013037809648,
      "grad_norm": 0.252510666847229,
      "learning_rate": 4.5051689765929214e-08,
      "loss": 0.1763,
      "step": 570
    },
    {
      "epoch": 3.0,
      "step": 576,
      "total_flos": 6.181870607374746e+17,
      "train_loss": 0.282241683660282,
      "train_runtime": 6736.3042,
      "train_samples_per_second": 2.732,
      "train_steps_per_second": 0.086
    }
  ],
  "logging_steps": 10,
  "max_steps": 576,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 6.181870607374746e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}