{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.3722533390779836,
  "eval_steps": 200,
  "global_step": 200,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.034467901766479965,
      "grad_norm": 14.77929973602295,
      "learning_rate": 1.3333333333333336e-07,
      "loss": 1.1838324546813965,
      "num_input_tokens_seen": 763216,
      "step": 5,
      "train_runtime": 194.2336,
      "train_tokens_per_second": 3929.372
    },
    {
      "epoch": 0.06893580353295993,
      "grad_norm": 12.38934326171875,
      "learning_rate": 3.0000000000000004e-07,
      "loss": 1.1945976257324218,
      "num_input_tokens_seen": 1580304,
      "step": 10,
      "train_runtime": 435.5774,
      "train_tokens_per_second": 3628.067
    },
    {
      "epoch": 0.1034037052994399,
      "grad_norm": 8.67546558380127,
      "learning_rate": 4.666666666666667e-07,
      "loss": 1.135545539855957,
      "num_input_tokens_seen": 2409616,
      "step": 15,
      "train_runtime": 661.463,
      "train_tokens_per_second": 3642.859
    },
    {
      "epoch": 0.13787160706591986,
      "grad_norm": 5.253643989562988,
      "learning_rate": 6.333333333333334e-07,
      "loss": 1.0469856262207031,
      "num_input_tokens_seen": 3198880,
      "step": 20,
      "train_runtime": 886.8402,
      "train_tokens_per_second": 3607.054
    },
    {
      "epoch": 0.17233950883239982,
      "grad_norm": 7.118396759033203,
      "learning_rate": 8.000000000000001e-07,
      "loss": 1.0419632911682128,
      "num_input_tokens_seen": 3947184,
      "step": 25,
      "train_runtime": 1087.6324,
      "train_tokens_per_second": 3629.153
    },
    {
      "epoch": 0.2068074105988798,
      "grad_norm": 3.814739465713501,
      "learning_rate": 9.666666666666668e-07,
      "loss": 0.9899843215942383,
      "num_input_tokens_seen": 4812960,
      "step": 30,
      "train_runtime": 1327.9833,
      "train_tokens_per_second": 3624.263
    },
    {
      "epoch": 0.24127531236535976,
      "grad_norm": 2.879802942276001,
      "learning_rate": 1.1333333333333334e-06,
      "loss": 0.881479549407959,
      "num_input_tokens_seen": 5571696,
      "step": 35,
      "train_runtime": 1533.2898,
      "train_tokens_per_second": 3633.818
    },
    {
      "epoch": 0.2757432141318397,
      "grad_norm": 3.848572015762329,
      "learning_rate": 1.3e-06,
      "loss": 0.9485733032226562,
      "num_input_tokens_seen": 6305024,
      "step": 40,
      "train_runtime": 1724.1275,
      "train_tokens_per_second": 3656.936
    },
    {
      "epoch": 0.3102111158983197,
      "grad_norm": 3.2666518688201904,
      "learning_rate": 1.4666666666666669e-06,
      "loss": 0.9032028198242188,
      "num_input_tokens_seen": 7011248,
      "step": 45,
      "train_runtime": 1918.5938,
      "train_tokens_per_second": 3654.368
    },
    {
      "epoch": 0.34467901766479964,
      "grad_norm": 3.2290165424346924,
      "learning_rate": 1.6333333333333335e-06,
      "loss": 0.8558422088623047,
      "num_input_tokens_seen": 7775856,
      "step": 50,
      "train_runtime": 2122.8803,
      "train_tokens_per_second": 3662.88
    },
    {
      "epoch": 0.3791469194312796,
      "grad_norm": 2.6570162773132324,
      "learning_rate": 1.8000000000000001e-06,
      "loss": 0.8577234268188476,
      "num_input_tokens_seen": 8615568,
      "step": 55,
      "train_runtime": 2358.8841,
      "train_tokens_per_second": 3652.391
    },
    {
      "epoch": 0.4136148211977596,
      "grad_norm": 2.952378511428833,
      "learning_rate": 1.9666666666666668e-06,
      "loss": 0.841306209564209,
      "num_input_tokens_seen": 9449232,
      "step": 60,
      "train_runtime": 2589.9555,
      "train_tokens_per_second": 3648.415
    },
    {
      "epoch": 0.44808272296423957,
      "grad_norm": 2.7049005031585693,
      "learning_rate": 2.133333333333334e-06,
      "loss": 0.8797351837158203,
      "num_input_tokens_seen": 10336816,
      "step": 65,
      "train_runtime": 2834.4719,
      "train_tokens_per_second": 3646.823
    },
    {
      "epoch": 0.4825506247307195,
      "grad_norm": 3.8687472343444824,
      "learning_rate": 2.3000000000000004e-06,
      "loss": 0.8679376602172851,
      "num_input_tokens_seen": 11152672,
      "step": 70,
      "train_runtime": 3068.1857,
      "train_tokens_per_second": 3634.94
    },
    {
      "epoch": 0.5170185264971995,
      "grad_norm": 3.071903705596924,
      "learning_rate": 2.466666666666667e-06,
      "loss": 0.8320528030395508,
      "num_input_tokens_seen": 11986736,
      "step": 75,
      "train_runtime": 3312.496,
      "train_tokens_per_second": 3618.642
    },
    {
      "epoch": 0.5514864282636794,
      "grad_norm": 3.149940252304077,
      "learning_rate": 2.6333333333333332e-06,
      "loss": 0.8066154479980469,
      "num_input_tokens_seen": 12906032,
      "step": 80,
      "train_runtime": 3581.749,
      "train_tokens_per_second": 3603.277
    },
    {
      "epoch": 0.5859543300301594,
      "grad_norm": 2.614154815673828,
      "learning_rate": 2.8000000000000003e-06,
      "loss": 0.9063267707824707,
      "num_input_tokens_seen": 13654768,
      "step": 85,
      "train_runtime": 3783.4247,
      "train_tokens_per_second": 3609.103
    },
    {
      "epoch": 0.6204222317966394,
      "grad_norm": 2.9211513996124268,
      "learning_rate": 2.9666666666666673e-06,
      "loss": 0.8308260917663575,
      "num_input_tokens_seen": 14459312,
      "step": 90,
      "train_runtime": 4017.7253,
      "train_tokens_per_second": 3598.88
    },
    {
      "epoch": 0.6548901335631193,
      "grad_norm": 3.020777702331543,
      "learning_rate": 3.133333333333334e-06,
      "loss": 0.7900361061096192,
      "num_input_tokens_seen": 15329440,
      "step": 95,
      "train_runtime": 4289.7023,
      "train_tokens_per_second": 3573.544
    },
    {
      "epoch": 0.6893580353295993,
      "grad_norm": 2.8474903106689453,
      "learning_rate": 3.3000000000000006e-06,
      "loss": 0.7851523399353028,
      "num_input_tokens_seen": 16187808,
      "step": 100,
      "train_runtime": 4560.5791,
      "train_tokens_per_second": 3549.507
    },
    {
      "epoch": 0.7238259370960792,
      "grad_norm": 3.340893507003784,
      "learning_rate": 3.4666666666666672e-06,
      "loss": 0.8186498641967773,
      "num_input_tokens_seen": 17039232,
      "step": 105,
      "train_runtime": 4811.3877,
      "train_tokens_per_second": 3541.438
    },
    {
      "epoch": 0.7582938388625592,
      "grad_norm": 4.5749640464782715,
      "learning_rate": 3.633333333333334e-06,
      "loss": 0.7917331695556641,
      "num_input_tokens_seen": 17895712,
      "step": 110,
      "train_runtime": 5065.4959,
      "train_tokens_per_second": 3532.865
    },
    {
      "epoch": 0.7927617406290393,
      "grad_norm": 2.5426993370056152,
      "learning_rate": 3.8000000000000005e-06,
      "loss": 0.805422592163086,
      "num_input_tokens_seen": 18782352,
      "step": 115,
      "train_runtime": 5320.4956,
      "train_tokens_per_second": 3530.188
    },
    {
      "epoch": 0.8272296423955192,
      "grad_norm": 2.7823312282562256,
      "learning_rate": 3.966666666666667e-06,
      "loss": 0.8618875503540039,
      "num_input_tokens_seen": 19706272,
      "step": 120,
      "train_runtime": 5600.0798,
      "train_tokens_per_second": 3518.927
    },
    {
      "epoch": 0.8616975441619992,
      "grad_norm": 3.244311809539795,
      "learning_rate": 4.133333333333333e-06,
      "loss": 0.8201093673706055,
      "num_input_tokens_seen": 20649888,
      "step": 125,
      "train_runtime": 5884.6585,
      "train_tokens_per_second": 3509.106
    },
    {
      "epoch": 0.8961654459284791,
      "grad_norm": 2.4104573726654053,
      "learning_rate": 4.3e-06,
      "loss": 0.8672359466552735,
      "num_input_tokens_seen": 21475840,
      "step": 130,
      "train_runtime": 6117.6488,
      "train_tokens_per_second": 3510.473
    },
    {
      "epoch": 0.9306333476949591,
      "grad_norm": 2.78433895111084,
      "learning_rate": 4.4666666666666665e-06,
      "loss": 0.874744987487793,
      "num_input_tokens_seen": 22324272,
      "step": 135,
      "train_runtime": 6379.4873,
      "train_tokens_per_second": 3499.383
    },
    {
      "epoch": 0.965101249461439,
      "grad_norm": 2.9161031246185303,
      "learning_rate": 4.633333333333334e-06,
      "loss": 0.7852656364440918,
      "num_input_tokens_seen": 23122368,
      "step": 140,
      "train_runtime": 6604.2994,
      "train_tokens_per_second": 3501.108
    },
    {
      "epoch": 0.999569151227919,
      "grad_norm": 4.696719169616699,
      "learning_rate": 4.800000000000001e-06,
      "loss": 0.8108936309814453,
      "num_input_tokens_seen": 23932368,
      "step": 145,
      "train_runtime": 6838.5342,
      "train_tokens_per_second": 3499.634
    },
    {
      "epoch": 1.027574321413184,
      "grad_norm": 2.551420211791992,
      "learning_rate": 4.966666666666667e-06,
      "loss": 0.713538122177124,
      "num_input_tokens_seen": 24582112,
      "step": 150,
      "train_runtime": 7009.9112,
      "train_tokens_per_second": 3506.765
    },
    {
      "epoch": 1.062042223179664,
      "grad_norm": 2.4008963108062744,
      "learning_rate": 4.997620553954645e-06,
      "loss": 0.6745035171508789,
      "num_input_tokens_seen": 25397968,
      "step": 155,
      "train_runtime": 7243.1623,
      "train_tokens_per_second": 3506.475
    },
    {
      "epoch": 1.096510124946144,
      "grad_norm": 2.7162740230560303,
      "learning_rate": 4.987961816680493e-06,
      "loss": 0.6937138080596924,
      "num_input_tokens_seen": 26289504,
      "step": 160,
      "train_runtime": 7514.0888,
      "train_tokens_per_second": 3498.695
    },
    {
      "epoch": 1.130978026712624,
      "grad_norm": 3.437391996383667,
      "learning_rate": 4.970903776169403e-06,
      "loss": 0.6533127784729004,
      "num_input_tokens_seen": 27060384,
      "step": 165,
      "train_runtime": 7729.4081,
      "train_tokens_per_second": 3500.965
    },
    {
      "epoch": 1.1654459284791039,
      "grad_norm": 2.6510908603668213,
      "learning_rate": 4.9464971637523465e-06,
      "loss": 0.7480425834655762,
      "num_input_tokens_seen": 27845072,
      "step": 170,
      "train_runtime": 7950.7509,
      "train_tokens_per_second": 3502.194
    },
    {
      "epoch": 1.1999138302455838,
      "grad_norm": 2.822885036468506,
      "learning_rate": 4.914814565722671e-06,
      "loss": 0.7171806335449219,
      "num_input_tokens_seen": 28641024,
      "step": 175,
      "train_runtime": 8174.1371,
      "train_tokens_per_second": 3503.859
    },
    {
      "epoch": 1.2343817320120638,
      "grad_norm": 3.3805811405181885,
      "learning_rate": 4.875950207461403e-06,
      "loss": 0.6780893802642822,
      "num_input_tokens_seen": 29432784,
      "step": 180,
      "train_runtime": 8407.0246,
      "train_tokens_per_second": 3500.975
    },
    {
      "epoch": 1.2688496337785438,
      "grad_norm": 2.527313709259033,
      "learning_rate": 4.830019673206997e-06,
      "loss": 0.6984196662902832,
      "num_input_tokens_seen": 30224352,
      "step": 185,
      "train_runtime": 8632.6235,
      "train_tokens_per_second": 3501.178
    },
    {
      "epoch": 1.3033175355450237,
      "grad_norm": 3.7762036323547363,
      "learning_rate": 4.77715956230294e-06,
      "loss": 0.6465680599212646,
      "num_input_tokens_seen": 31085888,
      "step": 190,
      "train_runtime": 8907.5477,
      "train_tokens_per_second": 3489.837
    },
    {
      "epoch": 1.3377854373115037,
      "grad_norm": 2.801496982574463,
      "learning_rate": 4.717527082945555e-06,
      "loss": 0.6420586585998536,
      "num_input_tokens_seen": 31874240,
      "step": 195,
      "train_runtime": 9141.109,
      "train_tokens_per_second": 3486.912
    },
    {
      "epoch": 1.3722533390779836,
      "grad_norm": 3.313443660736084,
      "learning_rate": 4.651299584640198e-06,
      "loss": 0.7168011665344238,
      "num_input_tokens_seen": 32621712,
      "step": 200,
      "train_runtime": 9343.2559,
      "train_tokens_per_second": 3491.472
    },
    {
      "epoch": 1.3722533390779836,
      "eval_loss": 0.5891635417938232,
      "eval_runtime": 41.6296,
      "eval_samples_per_second": 23.493,
      "eval_steps_per_second": 11.746,
      "num_input_tokens_seen": 32621712,
      "step": 200
    }
  ],
  "logging_steps": 5,
  "max_steps": 438,
  "num_input_tokens_seen": 32621712,
  "num_train_epochs": 3,
  "save_steps": 200,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 2.1058924130661827e+18,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}