{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 942,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.03184713375796178,
      "grad_norm": 5.6323702767298744,
      "learning_rate": 5e-06,
      "loss": 0.942,
      "step": 10
    },
    {
      "epoch": 0.06369426751592357,
      "grad_norm": 1.8772154794170284,
      "learning_rate": 5e-06,
      "loss": 0.8332,
      "step": 20
    },
    {
      "epoch": 0.09554140127388536,
      "grad_norm": 1.15240428378361,
      "learning_rate": 5e-06,
      "loss": 0.7857,
      "step": 30
    },
    {
      "epoch": 0.12738853503184713,
      "grad_norm": 0.9723964053330802,
      "learning_rate": 5e-06,
      "loss": 0.7718,
      "step": 40
    },
    {
      "epoch": 0.1592356687898089,
      "grad_norm": 1.180485085299224,
      "learning_rate": 5e-06,
      "loss": 0.7492,
      "step": 50
    },
    {
      "epoch": 0.1910828025477707,
      "grad_norm": 1.047903477889757,
      "learning_rate": 5e-06,
      "loss": 0.7413,
      "step": 60
    },
    {
      "epoch": 0.2229299363057325,
      "grad_norm": 0.8341131630541828,
      "learning_rate": 5e-06,
      "loss": 0.719,
      "step": 70
    },
    {
      "epoch": 0.25477707006369427,
      "grad_norm": 0.9122278937918772,
      "learning_rate": 5e-06,
      "loss": 0.7184,
      "step": 80
    },
    {
      "epoch": 0.28662420382165604,
      "grad_norm": 0.7073501886373434,
      "learning_rate": 5e-06,
      "loss": 0.7121,
      "step": 90
    },
    {
      "epoch": 0.3184713375796178,
      "grad_norm": 0.8281770199094491,
      "learning_rate": 5e-06,
      "loss": 0.7092,
      "step": 100
    },
    {
      "epoch": 0.3503184713375796,
      "grad_norm": 0.7210347968056159,
      "learning_rate": 5e-06,
      "loss": 0.7013,
      "step": 110
    },
    {
      "epoch": 0.3821656050955414,
      "grad_norm": 0.5738125544201067,
      "learning_rate": 5e-06,
      "loss": 0.7022,
      "step": 120
    },
    {
      "epoch": 0.4140127388535032,
      "grad_norm": 1.2444329803886918,
      "learning_rate": 5e-06,
      "loss": 0.7044,
      "step": 130
    },
    {
      "epoch": 0.445859872611465,
      "grad_norm": 0.75470283252076,
      "learning_rate": 5e-06,
      "loss": 0.692,
      "step": 140
    },
    {
      "epoch": 0.47770700636942676,
      "grad_norm": 0.6449167354505037,
      "learning_rate": 5e-06,
      "loss": 0.7023,
      "step": 150
    },
    {
      "epoch": 0.5095541401273885,
      "grad_norm": 0.5910288139996942,
      "learning_rate": 5e-06,
      "loss": 0.6944,
      "step": 160
    },
    {
      "epoch": 0.5414012738853503,
      "grad_norm": 0.5604434405342383,
      "learning_rate": 5e-06,
      "loss": 0.6966,
      "step": 170
    },
    {
      "epoch": 0.5732484076433121,
      "grad_norm": 0.6654582315660812,
      "learning_rate": 5e-06,
      "loss": 0.6876,
      "step": 180
    },
    {
      "epoch": 0.6050955414012739,
      "grad_norm": 0.9166199094846909,
      "learning_rate": 5e-06,
      "loss": 0.6891,
      "step": 190
    },
    {
      "epoch": 0.6369426751592356,
      "grad_norm": 0.7178723052852772,
      "learning_rate": 5e-06,
      "loss": 0.6832,
      "step": 200
    },
    {
      "epoch": 0.6687898089171974,
      "grad_norm": 0.6541344411840553,
      "learning_rate": 5e-06,
      "loss": 0.6803,
      "step": 210
    },
    {
      "epoch": 0.7006369426751592,
      "grad_norm": 0.637825014720494,
      "learning_rate": 5e-06,
      "loss": 0.6768,
      "step": 220
    },
    {
      "epoch": 0.732484076433121,
      "grad_norm": 0.5549500561748851,
      "learning_rate": 5e-06,
      "loss": 0.6825,
      "step": 230
    },
    {
      "epoch": 0.7643312101910829,
      "grad_norm": 0.654612026735291,
      "learning_rate": 5e-06,
      "loss": 0.6806,
      "step": 240
    },
    {
      "epoch": 0.7961783439490446,
      "grad_norm": 0.5538249059296838,
      "learning_rate": 5e-06,
      "loss": 0.6848,
      "step": 250
    },
    {
      "epoch": 0.8280254777070064,
      "grad_norm": 0.5723698853730642,
      "learning_rate": 5e-06,
      "loss": 0.6842,
      "step": 260
    },
    {
      "epoch": 0.8598726114649682,
      "grad_norm": 0.5768848558596881,
      "learning_rate": 5e-06,
      "loss": 0.6826,
      "step": 270
    },
    {
      "epoch": 0.89171974522293,
      "grad_norm": 0.531302038623778,
      "learning_rate": 5e-06,
      "loss": 0.6812,
      "step": 280
    },
    {
      "epoch": 0.9235668789808917,
      "grad_norm": 0.543420146899659,
      "learning_rate": 5e-06,
      "loss": 0.6774,
      "step": 290
    },
    {
      "epoch": 0.9554140127388535,
      "grad_norm": 0.509668170466935,
      "learning_rate": 5e-06,
      "loss": 0.6757,
      "step": 300
    },
    {
      "epoch": 0.9872611464968153,
      "grad_norm": 0.5426815942872757,
      "learning_rate": 5e-06,
      "loss": 0.6774,
      "step": 310
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.6760347485542297,
      "eval_runtime": 30.8722,
      "eval_samples_per_second": 273.871,
      "eval_steps_per_second": 1.101,
      "step": 314
    },
    {
      "epoch": 1.019108280254777,
      "grad_norm": 0.7477023983107832,
      "learning_rate": 5e-06,
      "loss": 0.6476,
      "step": 320
    },
    {
      "epoch": 1.0509554140127388,
      "grad_norm": 0.6149216256910138,
      "learning_rate": 5e-06,
      "loss": 0.6341,
      "step": 330
    },
    {
      "epoch": 1.0828025477707006,
      "grad_norm": 0.7013272885905804,
      "learning_rate": 5e-06,
      "loss": 0.6285,
      "step": 340
    },
    {
      "epoch": 1.1146496815286624,
      "grad_norm": 0.6645252684100534,
      "learning_rate": 5e-06,
      "loss": 0.6293,
      "step": 350
    },
    {
      "epoch": 1.1464968152866242,
      "grad_norm": 0.6691516868475139,
      "learning_rate": 5e-06,
      "loss": 0.624,
      "step": 360
    },
    {
      "epoch": 1.178343949044586,
      "grad_norm": 0.6277574945118575,
      "learning_rate": 5e-06,
      "loss": 0.628,
      "step": 370
    },
    {
      "epoch": 1.2101910828025477,
      "grad_norm": 0.5594126038913051,
      "learning_rate": 5e-06,
      "loss": 0.6293,
      "step": 380
    },
    {
      "epoch": 1.2420382165605095,
      "grad_norm": 0.6377404973712193,
      "learning_rate": 5e-06,
      "loss": 0.6272,
      "step": 390
    },
    {
      "epoch": 1.2738853503184713,
      "grad_norm": 0.5512534318917951,
      "learning_rate": 5e-06,
      "loss": 0.6314,
      "step": 400
    },
    {
      "epoch": 1.305732484076433,
      "grad_norm": 0.6708898369555326,
      "learning_rate": 5e-06,
      "loss": 0.6309,
      "step": 410
    },
    {
      "epoch": 1.3375796178343948,
      "grad_norm": 0.662681777058966,
      "learning_rate": 5e-06,
      "loss": 0.6322,
      "step": 420
    },
    {
      "epoch": 1.3694267515923566,
      "grad_norm": 0.5855564523983591,
      "learning_rate": 5e-06,
      "loss": 0.6293,
      "step": 430
    },
    {
      "epoch": 1.4012738853503186,
      "grad_norm": 0.7027168380973281,
      "learning_rate": 5e-06,
      "loss": 0.6231,
      "step": 440
    },
    {
      "epoch": 1.4331210191082802,
      "grad_norm": 0.5875190666099424,
      "learning_rate": 5e-06,
      "loss": 0.6308,
      "step": 450
    },
    {
      "epoch": 1.4649681528662422,
      "grad_norm": 0.5695855789396209,
      "learning_rate": 5e-06,
      "loss": 0.6316,
      "step": 460
    },
    {
      "epoch": 1.4968152866242037,
      "grad_norm": 0.5640229543248884,
      "learning_rate": 5e-06,
      "loss": 0.6216,
      "step": 470
    },
    {
      "epoch": 1.5286624203821657,
      "grad_norm": 0.5284708915831005,
      "learning_rate": 5e-06,
      "loss": 0.634,
      "step": 480
    },
    {
      "epoch": 1.5605095541401273,
      "grad_norm": 0.6601370586932329,
      "learning_rate": 5e-06,
      "loss": 0.6211,
      "step": 490
    },
    {
      "epoch": 1.5923566878980893,
      "grad_norm": 0.5141465072879543,
      "learning_rate": 5e-06,
      "loss": 0.6263,
      "step": 500
    },
    {
      "epoch": 1.6242038216560508,
      "grad_norm": 0.7951970752704245,
      "learning_rate": 5e-06,
      "loss": 0.6283,
      "step": 510
    },
    {
      "epoch": 1.6560509554140128,
      "grad_norm": 0.6915600915890423,
      "learning_rate": 5e-06,
      "loss": 0.6296,
      "step": 520
    },
    {
      "epoch": 1.6878980891719744,
      "grad_norm": 0.5352783895102649,
      "learning_rate": 5e-06,
      "loss": 0.6243,
      "step": 530
    },
    {
      "epoch": 1.7197452229299364,
      "grad_norm": 0.6389694940943629,
      "learning_rate": 5e-06,
      "loss": 0.6262,
      "step": 540
    },
    {
      "epoch": 1.7515923566878981,
      "grad_norm": 0.638658826251017,
      "learning_rate": 5e-06,
      "loss": 0.6233,
      "step": 550
    },
    {
      "epoch": 1.78343949044586,
      "grad_norm": 0.5164381003657269,
      "learning_rate": 5e-06,
      "loss": 0.6325,
      "step": 560
    },
    {
      "epoch": 1.8152866242038217,
      "grad_norm": 0.5462044960453822,
      "learning_rate": 5e-06,
      "loss": 0.6262,
      "step": 570
    },
    {
      "epoch": 1.8471337579617835,
      "grad_norm": 0.5320271319641363,
      "learning_rate": 5e-06,
      "loss": 0.6238,
      "step": 580
    },
    {
      "epoch": 1.8789808917197452,
      "grad_norm": 0.6783189674697148,
      "learning_rate": 5e-06,
      "loss": 0.6195,
      "step": 590
    },
    {
      "epoch": 1.910828025477707,
      "grad_norm": 0.6838086842362504,
      "learning_rate": 5e-06,
      "loss": 0.6242,
      "step": 600
    },
    {
      "epoch": 1.9426751592356688,
      "grad_norm": 0.6094514581891239,
      "learning_rate": 5e-06,
      "loss": 0.6282,
      "step": 610
    },
    {
      "epoch": 1.9745222929936306,
      "grad_norm": 0.627522298883166,
      "learning_rate": 5e-06,
      "loss": 0.6192,
      "step": 620
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.6656389832496643,
      "eval_runtime": 31.0662,
      "eval_samples_per_second": 272.161,
      "eval_steps_per_second": 1.094,
      "step": 628
    },
    {
      "epoch": 2.0063694267515926,
      "grad_norm": 0.9573075915273472,
      "learning_rate": 5e-06,
      "loss": 0.6157,
      "step": 630
    },
    {
      "epoch": 2.038216560509554,
      "grad_norm": 0.7301893995079715,
      "learning_rate": 5e-06,
      "loss": 0.5735,
      "step": 640
    },
    {
      "epoch": 2.070063694267516,
      "grad_norm": 0.7084282739053801,
      "learning_rate": 5e-06,
      "loss": 0.5778,
      "step": 650
    },
    {
      "epoch": 2.1019108280254777,
      "grad_norm": 0.6585976344607456,
      "learning_rate": 5e-06,
      "loss": 0.5784,
      "step": 660
    },
    {
      "epoch": 2.1337579617834397,
      "grad_norm": 0.6540062522321651,
      "learning_rate": 5e-06,
      "loss": 0.5751,
      "step": 670
    },
    {
      "epoch": 2.1656050955414012,
      "grad_norm": 0.5639591666878391,
      "learning_rate": 5e-06,
      "loss": 0.5766,
      "step": 680
    },
    {
      "epoch": 2.1974522292993632,
      "grad_norm": 0.8212954305221217,
      "learning_rate": 5e-06,
      "loss": 0.5798,
      "step": 690
    },
    {
      "epoch": 2.229299363057325,
      "grad_norm": 0.8109275454168715,
      "learning_rate": 5e-06,
      "loss": 0.5767,
      "step": 700
    },
    {
      "epoch": 2.261146496815287,
      "grad_norm": 0.6107672621726716,
      "learning_rate": 5e-06,
      "loss": 0.5773,
      "step": 710
    },
    {
      "epoch": 2.2929936305732483,
      "grad_norm": 0.6385570175836712,
      "learning_rate": 5e-06,
      "loss": 0.579,
      "step": 720
    },
    {
      "epoch": 2.3248407643312103,
      "grad_norm": 0.5907370356401401,
      "learning_rate": 5e-06,
      "loss": 0.5796,
      "step": 730
    },
    {
      "epoch": 2.356687898089172,
      "grad_norm": 0.5656820007312235,
      "learning_rate": 5e-06,
      "loss": 0.5748,
      "step": 740
    },
    {
      "epoch": 2.388535031847134,
      "grad_norm": 0.5760709804095724,
      "learning_rate": 5e-06,
      "loss": 0.5834,
      "step": 750
    },
    {
      "epoch": 2.4203821656050954,
      "grad_norm": 0.6305043111751713,
      "learning_rate": 5e-06,
      "loss": 0.5758,
      "step": 760
    },
    {
      "epoch": 2.4522292993630574,
      "grad_norm": 0.5901331102554143,
      "learning_rate": 5e-06,
      "loss": 0.5856,
      "step": 770
    },
    {
      "epoch": 2.484076433121019,
      "grad_norm": 0.6608011253619817,
      "learning_rate": 5e-06,
      "loss": 0.5786,
      "step": 780
    },
    {
      "epoch": 2.515923566878981,
      "grad_norm": 0.6413426727699342,
      "learning_rate": 5e-06,
      "loss": 0.577,
      "step": 790
    },
    {
      "epoch": 2.5477707006369426,
      "grad_norm": 0.5719942317033819,
      "learning_rate": 5e-06,
      "loss": 0.581,
      "step": 800
    },
    {
      "epoch": 2.5796178343949046,
      "grad_norm": 0.5846298214850376,
      "learning_rate": 5e-06,
      "loss": 0.5809,
      "step": 810
    },
    {
      "epoch": 2.611464968152866,
      "grad_norm": 0.6682316534474163,
      "learning_rate": 5e-06,
      "loss": 0.5845,
      "step": 820
    },
    {
      "epoch": 2.643312101910828,
      "grad_norm": 0.5948936776108811,
      "learning_rate": 5e-06,
      "loss": 0.5837,
      "step": 830
    },
    {
      "epoch": 2.6751592356687897,
      "grad_norm": 0.8151554129031496,
      "learning_rate": 5e-06,
      "loss": 0.5871,
      "step": 840
    },
    {
      "epoch": 2.7070063694267517,
      "grad_norm": 0.5916261196390682,
      "learning_rate": 5e-06,
      "loss": 0.5797,
      "step": 850
    },
    {
      "epoch": 2.738853503184713,
      "grad_norm": 0.6442969800604952,
      "learning_rate": 5e-06,
      "loss": 0.5814,
      "step": 860
    },
    {
      "epoch": 2.770700636942675,
      "grad_norm": 0.6420974104025357,
      "learning_rate": 5e-06,
      "loss": 0.585,
      "step": 870
    },
    {
      "epoch": 2.802547770700637,
      "grad_norm": 0.6318903979436119,
      "learning_rate": 5e-06,
      "loss": 0.5813,
      "step": 880
    },
    {
      "epoch": 2.8343949044585988,
      "grad_norm": 0.6105841622731739,
      "learning_rate": 5e-06,
      "loss": 0.5823,
      "step": 890
    },
    {
      "epoch": 2.8662420382165603,
      "grad_norm": 0.5727955806455897,
      "learning_rate": 5e-06,
      "loss": 0.5799,
      "step": 900
    },
    {
      "epoch": 2.8980891719745223,
      "grad_norm": 0.5970248698087671,
      "learning_rate": 5e-06,
      "loss": 0.583,
      "step": 910
    },
    {
      "epoch": 2.9299363057324843,
      "grad_norm": 0.6176747095322584,
      "learning_rate": 5e-06,
      "loss": 0.5816,
      "step": 920
    },
    {
      "epoch": 2.961783439490446,
      "grad_norm": 0.550436105577002,
      "learning_rate": 5e-06,
      "loss": 0.585,
      "step": 930
    },
    {
      "epoch": 2.9936305732484074,
      "grad_norm": 0.6334063259900834,
      "learning_rate": 5e-06,
      "loss": 0.5826,
      "step": 940
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.6691574454307556,
      "eval_runtime": 30.8039,
      "eval_samples_per_second": 274.478,
      "eval_steps_per_second": 1.104,
      "step": 942
    },
    {
      "epoch": 3.0,
      "step": 942,
      "total_flos": 1577885085204480.0,
      "train_loss": 0.6401269947141093,
      "train_runtime": 5912.9311,
      "train_samples_per_second": 81.505,
      "train_steps_per_second": 0.159
    }
  ],
  "logging_steps": 10,
  "max_steps": 942,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "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": 1577885085204480.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}