{
  "best_metric": 0.5713418126106262,
  "best_model_checkpoint": "./output/checkpoint-2200",
  "epoch": 2.5143755134111934,
  "eval_steps": 200,
  "global_step": 2200,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 3.282,
      "step": 10
    },
    {
      "epoch": 0.02,
      "learning_rate": 4.000000000000001e-06,
      "loss": 3.2806,
      "step": 20
    },
    {
      "epoch": 0.03,
      "learning_rate": 6e-06,
      "loss": 3.2842,
      "step": 30
    },
    {
      "epoch": 0.05,
      "learning_rate": 8.000000000000001e-06,
      "loss": 3.2378,
      "step": 40
    },
    {
      "epoch": 0.06,
      "learning_rate": 1e-05,
      "loss": 3.202,
      "step": 50
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.2e-05,
      "loss": 3.1307,
      "step": 60
    },
    {
      "epoch": 0.08,
      "learning_rate": 1.4e-05,
      "loss": 3.0392,
      "step": 70
    },
    {
      "epoch": 0.09,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 2.9102,
      "step": 80
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.8e-05,
      "loss": 2.7595,
      "step": 90
    },
    {
      "epoch": 0.11,
      "learning_rate": 2e-05,
      "loss": 2.5737,
      "step": 100
    },
    {
      "epoch": 0.13,
      "learning_rate": 1.9999224156779658e-05,
      "loss": 2.347,
      "step": 110
    },
    {
      "epoch": 0.14,
      "learning_rate": 1.999689674750517e-05,
      "loss": 2.0934,
      "step": 120
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.9993018133317476e-05,
      "loss": 1.8305,
      "step": 130
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.998758891605588e-05,
      "loss": 1.5447,
      "step": 140
    },
    {
      "epoch": 0.17,
      "learning_rate": 1.9980609938164664e-05,
      "loss": 1.3394,
      "step": 150
    },
    {
      "epoch": 0.18,
      "learning_rate": 1.9972082282562367e-05,
      "loss": 1.2032,
      "step": 160
    },
    {
      "epoch": 0.19,
      "learning_rate": 1.9962007272473744e-05,
      "loss": 1.1096,
      "step": 170
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.9950386471224445e-05,
      "loss": 1.0456,
      "step": 180
    },
    {
      "epoch": 0.22,
      "learning_rate": 1.9937221681998447e-05,
      "loss": 0.9977,
      "step": 190
    },
    {
      "epoch": 0.23,
      "learning_rate": 1.9922514947558248e-05,
      "loss": 0.9786,
      "step": 200
    },
    {
      "epoch": 0.23,
      "eval_loss": 0.982032299041748,
      "eval_runtime": 759.4598,
      "eval_samples_per_second": 16.387,
      "eval_steps_per_second": 2.049,
      "step": 200
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.9906268549927886e-05,
      "loss": 0.948,
      "step": 210
    },
    {
      "epoch": 0.25,
      "learning_rate": 1.9888485010038848e-05,
      "loss": 0.9164,
      "step": 220
    },
    {
      "epoch": 0.26,
      "learning_rate": 1.9869167087338908e-05,
      "loss": 0.8977,
      "step": 230
    },
    {
      "epoch": 0.27,
      "learning_rate": 1.984831777936394e-05,
      "loss": 0.876,
      "step": 240
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.9825940321272792e-05,
      "loss": 0.8604,
      "step": 250
    },
    {
      "epoch": 0.3,
      "learning_rate": 1.980203818534529e-05,
      "loss": 0.8424,
      "step": 260
    },
    {
      "epoch": 0.31,
      "learning_rate": 1.9776615080443462e-05,
      "loss": 0.8434,
      "step": 270
    },
    {
      "epoch": 0.32,
      "learning_rate": 1.9749674951436016e-05,
      "loss": 0.8357,
      "step": 280
    },
    {
      "epoch": 0.33,
      "learning_rate": 1.972122197858625e-05,
      "loss": 0.8239,
      "step": 290
    },
    {
      "epoch": 0.34,
      "learning_rate": 1.9691260576903372e-05,
      "loss": 0.8177,
      "step": 300
    },
    {
      "epoch": 0.35,
      "learning_rate": 1.965979539545746e-05,
      "loss": 0.8151,
      "step": 310
    },
    {
      "epoch": 0.37,
      "learning_rate": 1.9626831316658058e-05,
      "loss": 0.8074,
      "step": 320
    },
    {
      "epoch": 0.38,
      "learning_rate": 1.959237345549657e-05,
      "loss": 0.7768,
      "step": 330
    },
    {
      "epoch": 0.39,
      "learning_rate": 1.955642715875259e-05,
      "loss": 0.7673,
      "step": 340
    },
    {
      "epoch": 0.4,
      "learning_rate": 1.951899800416425e-05,
      "loss": 0.7448,
      "step": 350
    },
    {
      "epoch": 0.41,
      "learning_rate": 1.9480091799562706e-05,
      "loss": 0.7334,
      "step": 360
    },
    {
      "epoch": 0.42,
      "learning_rate": 1.9439714581970983e-05,
      "loss": 0.734,
      "step": 370
    },
    {
      "epoch": 0.43,
      "learning_rate": 1.9397872616667175e-05,
      "loss": 0.7301,
      "step": 380
    },
    {
      "epoch": 0.45,
      "learning_rate": 1.935457239621231e-05,
      "loss": 0.734,
      "step": 390
    },
    {
      "epoch": 0.46,
      "learning_rate": 1.9309820639442886e-05,
      "loss": 0.7062,
      "step": 400
    },
    {
      "epoch": 0.46,
      "eval_loss": 0.7326281070709229,
      "eval_runtime": 757.8879,
      "eval_samples_per_second": 16.421,
      "eval_steps_per_second": 2.053,
      "step": 400
    },
    {
      "epoch": 0.47,
      "learning_rate": 1.9263624290428314e-05,
      "loss": 0.7074,
      "step": 410
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.9215990517393436e-05,
      "loss": 0.7001,
      "step": 420
    },
    {
      "epoch": 0.49,
      "learning_rate": 1.9166926711606222e-05,
      "loss": 0.6991,
      "step": 430
    },
    {
      "epoch": 0.5,
      "learning_rate": 1.9116440486230893e-05,
      "loss": 0.6935,
      "step": 440
    },
    {
      "epoch": 0.51,
      "learning_rate": 1.9064539675146584e-05,
      "loss": 0.6922,
      "step": 450
    },
    {
      "epoch": 0.53,
      "learning_rate": 1.9011232331731773e-05,
      "loss": 0.6861,
      "step": 460
    },
    {
      "epoch": 0.54,
      "learning_rate": 1.8956526727614663e-05,
      "loss": 0.6978,
      "step": 470
    },
    {
      "epoch": 0.55,
      "learning_rate": 1.8900431351389657e-05,
      "loss": 0.6699,
      "step": 480
    },
    {
      "epoch": 0.56,
      "learning_rate": 1.8842954907300236e-05,
      "loss": 0.6846,
      "step": 490
    },
    {
      "epoch": 0.57,
      "learning_rate": 1.878410631388829e-05,
      "loss": 0.6772,
      "step": 500
    },
    {
      "epoch": 0.58,
      "learning_rate": 1.8723894702610258e-05,
      "loss": 0.6808,
      "step": 510
    },
    {
      "epoch": 0.59,
      "learning_rate": 1.866232941642023e-05,
      "loss": 0.6869,
      "step": 520
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.8599420008320176e-05,
      "loss": 0.6595,
      "step": 530
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.8535176239877653e-05,
      "loss": 0.6669,
      "step": 540
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.8469608079711105e-05,
      "loss": 0.6648,
      "step": 550
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.8402725701943028e-05,
      "loss": 0.6486,
      "step": 560
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.8334539484621304e-05,
      "loss": 0.6735,
      "step": 570
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.8265060008108815e-05,
      "loss": 0.6498,
      "step": 580
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.8194298053441725e-05,
      "loss": 0.6578,
      "step": 590
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.812226460065658e-05,
      "loss": 0.6392,
      "step": 600
    },
    {
      "epoch": 0.69,
      "eval_loss": 0.6629914045333862,
      "eval_runtime": 765.6279,
      "eval_samples_per_second": 16.255,
      "eval_steps_per_second": 2.032,
      "step": 600
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.804897082708659e-05,
      "loss": 0.6524,
      "step": 610
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.797442810562721e-05,
      "loss": 0.6513,
      "step": 620
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.7898648002971456e-05,
      "loss": 0.6466,
      "step": 630
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.7821642277815108e-05,
      "loss": 0.6305,
      "step": 640
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.774342287903212e-05,
      "loss": 0.6418,
      "step": 650
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.766400194382054e-05,
      "loss": 0.636,
      "step": 660
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.75833917958192e-05,
      "loss": 0.6279,
      "step": 670
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.7501604943195466e-05,
      "loss": 0.6369,
      "step": 680
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.741865407670435e-05,
      "loss": 0.6308,
      "step": 690
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.733455206771934e-05,
      "loss": 0.616,
      "step": 700
    },
    {
      "epoch": 0.81,
      "learning_rate": 1.724931196623513e-05,
      "loss": 0.6334,
      "step": 710
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.716294699884269e-05,
      "loss": 0.6289,
      "step": 720
    },
    {
      "epoch": 0.83,
      "learning_rate": 1.70754705666769e-05,
      "loss": 0.6251,
      "step": 730
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.6986896243337137e-05,
      "loss": 0.619,
      "step": 740
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.6897237772781046e-05,
      "loss": 0.6226,
      "step": 750
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.6806509067191926e-05,
      "loss": 0.6229,
      "step": 760
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.6714724204820017e-05,
      "loss": 0.6016,
      "step": 770
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.6621897427797947e-05,
      "loss": 0.6061,
      "step": 780
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.6528043139930845e-05,
      "loss": 0.6027,
      "step": 790
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.64331759044613e-05,
      "loss": 0.6018,
      "step": 800
    },
    {
      "epoch": 0.91,
      "eval_loss": 0.6285067796707153,
      "eval_runtime": 765.1757,
      "eval_samples_per_second": 16.264,
      "eval_steps_per_second": 2.034,
      "step": 800
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.63373104418096e-05,
      "loss": 0.6094,
      "step": 810
    },
    {
      "epoch": 0.94,
      "learning_rate": 1.6240461627289607e-05,
      "loss": 0.6224,
      "step": 820
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.614264448880054e-05,
      "loss": 0.6032,
      "step": 830
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.6043874204495152e-05,
      "loss": 0.6048,
      "step": 840
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.5944166100424534e-05,
      "loss": 0.6036,
      "step": 850
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.584353564815999e-05,
      "loss": 0.5932,
      "step": 860
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.5741998462392352e-05,
      "loss": 0.6003,
      "step": 870
    },
    {
      "epoch": 1.01,
      "learning_rate": 1.5639570298509067e-05,
      "loss": 0.6128,
      "step": 880
    },
    {
      "epoch": 1.02,
      "learning_rate": 1.5536267050149428e-05,
      "loss": 0.5968,
      "step": 890
    },
    {
      "epoch": 1.03,
      "learning_rate": 1.5432104746738417e-05,
      "loss": 0.6027,
      "step": 900
    },
    {
      "epoch": 1.04,
      "learning_rate": 1.5327099550999416e-05,
      "loss": 0.6122,
      "step": 910
    },
    {
      "epoch": 1.05,
      "learning_rate": 1.5221267756446266e-05,
      "loss": 0.5984,
      "step": 920
    },
    {
      "epoch": 1.06,
      "learning_rate": 1.5114625784855028e-05,
      "loss": 0.5965,
      "step": 930
    },
    {
      "epoch": 1.07,
      "learning_rate": 1.5007190183715836e-05,
      "loss": 0.5942,
      "step": 940
    },
    {
      "epoch": 1.09,
      "learning_rate": 1.4898977623665243e-05,
      "loss": 0.6,
      "step": 950
    },
    {
      "epoch": 1.1,
      "learning_rate": 1.4790004895899465e-05,
      "loss": 0.6051,
      "step": 960
    },
    {
      "epoch": 1.11,
      "learning_rate": 1.4680288909568905e-05,
      "loss": 0.5903,
      "step": 970
    },
    {
      "epoch": 1.12,
      "learning_rate": 1.4569846689154401e-05,
      "loss": 0.6049,
      "step": 980
    },
    {
      "epoch": 1.13,
      "learning_rate": 1.4458695371825535e-05,
      "loss": 0.5852,
      "step": 990
    },
    {
      "epoch": 1.14,
      "learning_rate": 1.4346852204781511e-05,
      "loss": 0.5869,
      "step": 1000
    },
    {
      "epoch": 1.14,
      "eval_loss": 0.6076732277870178,
      "eval_runtime": 765.3772,
      "eval_samples_per_second": 16.26,
      "eval_steps_per_second": 2.033,
      "step": 1000
    },
    {
      "epoch": 1.15,
      "learning_rate": 1.4234334542574906e-05,
      "loss": 0.5893,
      "step": 1010
    },
    {
      "epoch": 1.17,
      "learning_rate": 1.4121159844418793e-05,
      "loss": 0.5874,
      "step": 1020
    },
    {
      "epoch": 1.18,
      "learning_rate": 1.400734567147763e-05,
      "loss": 0.595,
      "step": 1030
    },
    {
      "epoch": 1.19,
      "learning_rate": 1.3892909684142311e-05,
      "loss": 0.5961,
      "step": 1040
    },
    {
      "epoch": 1.2,
      "learning_rate": 1.3777869639289822e-05,
      "loss": 0.5826,
      "step": 1050
    },
    {
      "epoch": 1.21,
      "learning_rate": 1.3662243387527936e-05,
      "loss": 0.5835,
      "step": 1060
    },
    {
      "epoch": 1.22,
      "learning_rate": 1.3546048870425356e-05,
      "loss": 0.5858,
      "step": 1070
    },
    {
      "epoch": 1.23,
      "learning_rate": 1.3429304117727755e-05,
      "loss": 0.5931,
      "step": 1080
    },
    {
      "epoch": 1.25,
      "learning_rate": 1.3312027244560104e-05,
      "loss": 0.5858,
      "step": 1090
    },
    {
      "epoch": 1.26,
      "learning_rate": 1.3194236448615798e-05,
      "loss": 0.5834,
      "step": 1100
    },
    {
      "epoch": 1.27,
      "learning_rate": 1.3075950007332923e-05,
      "loss": 0.5854,
      "step": 1110
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.2957186275058188e-05,
      "loss": 0.5879,
      "step": 1120
    },
    {
      "epoch": 1.29,
      "learning_rate": 1.2837963680198895e-05,
      "loss": 0.5958,
      "step": 1130
    },
    {
      "epoch": 1.3,
      "learning_rate": 1.2718300722363431e-05,
      "loss": 0.5787,
      "step": 1140
    },
    {
      "epoch": 1.31,
      "learning_rate": 1.2598215969490703e-05,
      "loss": 0.5827,
      "step": 1150
    },
    {
      "epoch": 1.33,
      "learning_rate": 1.2477728054968993e-05,
      "loss": 0.589,
      "step": 1160
    },
    {
      "epoch": 1.34,
      "learning_rate": 1.2356855674744625e-05,
      "loss": 0.5806,
      "step": 1170
    },
    {
      "epoch": 1.35,
      "learning_rate": 1.2235617584420938e-05,
      "loss": 0.563,
      "step": 1180
    },
    {
      "epoch": 1.36,
      "learning_rate": 1.2114032596348023e-05,
      "loss": 0.5812,
      "step": 1190
    },
    {
      "epoch": 1.37,
      "learning_rate": 1.1992119576703614e-05,
      "loss": 0.5995,
      "step": 1200
    },
    {
      "epoch": 1.37,
      "eval_loss": 0.5944839119911194,
      "eval_runtime": 764.7162,
      "eval_samples_per_second": 16.274,
      "eval_steps_per_second": 2.035,
      "step": 1200
    },
    {
      "epoch": 1.38,
      "learning_rate": 1.1869897442565665e-05,
      "loss": 0.5852,
      "step": 1210
    },
    {
      "epoch": 1.39,
      "learning_rate": 1.1747385158977008e-05,
      "loss": 0.5798,
      "step": 1220
    },
    {
      "epoch": 1.41,
      "learning_rate": 1.162460173600256e-05,
      "loss": 0.5846,
      "step": 1230
    },
    {
      "epoch": 1.42,
      "learning_rate": 1.1501566225779592e-05,
      "loss": 0.5674,
      "step": 1240
    },
    {
      "epoch": 1.43,
      "learning_rate": 1.1378297719561388e-05,
      "loss": 0.5616,
      "step": 1250
    },
    {
      "epoch": 1.44,
      "learning_rate": 1.1254815344754919e-05,
      "loss": 0.5738,
      "step": 1260
    },
    {
      "epoch": 1.45,
      "learning_rate": 1.1131138261952845e-05,
      "loss": 0.5831,
      "step": 1270
    },
    {
      "epoch": 1.46,
      "learning_rate": 1.100728566196041e-05,
      "loss": 0.5732,
      "step": 1280
    },
    {
      "epoch": 1.47,
      "learning_rate": 1.088327676281762e-05,
      "loss": 0.5702,
      "step": 1290
    },
    {
      "epoch": 1.49,
      "learning_rate": 1.0759130806817204e-05,
      "loss": 0.5748,
      "step": 1300
    },
    {
      "epoch": 1.5,
      "learning_rate": 1.0634867057518824e-05,
      "loss": 0.5764,
      "step": 1310
    },
    {
      "epoch": 1.51,
      "learning_rate": 1.0510504796759966e-05,
      "loss": 0.5716,
      "step": 1320
    },
    {
      "epoch": 1.52,
      "learning_rate": 1.0386063321664007e-05,
      "loss": 0.5685,
      "step": 1330
    },
    {
      "epoch": 1.53,
      "learning_rate": 1.0261561941645894e-05,
      "loss": 0.5713,
      "step": 1340
    },
    {
      "epoch": 1.54,
      "learning_rate": 1.013701997541596e-05,
      "loss": 0.5668,
      "step": 1350
    },
    {
      "epoch": 1.55,
      "learning_rate": 1.0012456747982231e-05,
      "loss": 0.5714,
      "step": 1360
    },
    {
      "epoch": 1.57,
      "learning_rate": 9.887891587651807e-06,
      "loss": 0.5714,
      "step": 1370
    },
    {
      "epoch": 1.58,
      "learning_rate": 9.763343823031718e-06,
      "loss": 0.5702,
      "step": 1380
    },
    {
      "epoch": 1.59,
      "learning_rate": 9.638832780029716e-06,
      "loss": 0.5706,
      "step": 1390
    },
    {
      "epoch": 1.6,
      "learning_rate": 9.514377778855521e-06,
      "loss": 0.57,
      "step": 1400
    },
    {
      "epoch": 1.6,
      "eval_loss": 0.5853541493415833,
      "eval_runtime": 765.003,
      "eval_samples_per_second": 16.268,
      "eval_steps_per_second": 2.034,
      "step": 1400
    },
    {
      "epoch": 1.61,
      "learning_rate": 9.389998131022906e-06,
      "loss": 0.563,
      "step": 1410
    },
    {
      "epoch": 1.62,
      "learning_rate": 9.265713136353184e-06,
      "loss": 0.5559,
      "step": 1420
    },
    {
      "epoch": 1.63,
      "learning_rate": 9.141542079980453e-06,
      "loss": 0.5765,
      "step": 1430
    },
    {
      "epoch": 1.65,
      "learning_rate": 9.017504229359157e-06,
      "loss": 0.5737,
      "step": 1440
    },
    {
      "epoch": 1.66,
      "learning_rate": 8.893618831274399e-06,
      "loss": 0.5588,
      "step": 1450
    },
    {
      "epoch": 1.67,
      "learning_rate": 8.769905108855412e-06,
      "loss": 0.5619,
      "step": 1460
    },
    {
      "epoch": 1.68,
      "learning_rate": 8.646382258592762e-06,
      "loss": 0.5612,
      "step": 1470
    },
    {
      "epoch": 1.69,
      "learning_rate": 8.523069447359631e-06,
      "loss": 0.5647,
      "step": 1480
    },
    {
      "epoch": 1.7,
      "learning_rate": 8.399985809437737e-06,
      "loss": 0.5803,
      "step": 1490
    },
    {
      "epoch": 1.71,
      "learning_rate": 8.277150443548285e-06,
      "loss": 0.554,
      "step": 1500
    },
    {
      "epoch": 1.73,
      "learning_rate": 8.154582409888441e-06,
      "loss": 0.5734,
      "step": 1510
    },
    {
      "epoch": 1.74,
      "learning_rate": 8.032300727173797e-06,
      "loss": 0.5646,
      "step": 1520
    },
    {
      "epoch": 1.75,
      "learning_rate": 7.91032436968725e-06,
      "loss": 0.5665,
      "step": 1530
    },
    {
      "epoch": 1.76,
      "learning_rate": 7.788672264334804e-06,
      "loss": 0.5519,
      "step": 1540
    },
    {
      "epoch": 1.77,
      "learning_rate": 7.667363287708692e-06,
      "loss": 0.5676,
      "step": 1550
    },
    {
      "epoch": 1.78,
      "learning_rate": 7.546416263158326e-06,
      "loss": 0.5516,
      "step": 1560
    },
    {
      "epoch": 1.79,
      "learning_rate": 7.42584995786952e-06,
      "loss": 0.5553,
      "step": 1570
    },
    {
      "epoch": 1.81,
      "learning_rate": 7.305683079952381e-06,
      "loss": 0.56,
      "step": 1580
    },
    {
      "epoch": 1.82,
      "learning_rate": 7.185934275538416e-06,
      "loss": 0.5715,
      "step": 1590
    },
    {
      "epoch": 1.83,
      "learning_rate": 7.066622125887237e-06,
      "loss": 0.5459,
      "step": 1600
    },
    {
      "epoch": 1.83,
      "eval_loss": 0.5793296098709106,
      "eval_runtime": 764.5785,
      "eval_samples_per_second": 16.277,
      "eval_steps_per_second": 2.035,
      "step": 1600
    },
    {
      "epoch": 1.84,
      "learning_rate": 6.947765144503323e-06,
      "loss": 0.5499,
      "step": 1610
    },
    {
      "epoch": 1.85,
      "learning_rate": 6.8293817742633175e-06,
      "loss": 0.5483,
      "step": 1620
    },
    {
      "epoch": 1.86,
      "learning_rate": 6.711490384554257e-06,
      "loss": 0.5565,
      "step": 1630
    },
    {
      "epoch": 1.87,
      "learning_rate": 6.59410926842323e-06,
      "loss": 0.5596,
      "step": 1640
    },
    {
      "epoch": 1.89,
      "learning_rate": 6.477256639738872e-06,
      "loss": 0.5623,
      "step": 1650
    },
    {
      "epoch": 1.9,
      "learning_rate": 6.360950630365126e-06,
      "loss": 0.5509,
      "step": 1660
    },
    {
      "epoch": 1.91,
      "learning_rate": 6.245209287347767e-06,
      "loss": 0.561,
      "step": 1670
    },
    {
      "epoch": 1.92,
      "learning_rate": 6.130050570114049e-06,
      "loss": 0.57,
      "step": 1680
    },
    {
      "epoch": 1.93,
      "learning_rate": 6.015492347685983e-06,
      "loss": 0.5624,
      "step": 1690
    },
    {
      "epoch": 1.94,
      "learning_rate": 5.901552395907606e-06,
      "loss": 0.5574,
      "step": 1700
    },
    {
      "epoch": 1.95,
      "learning_rate": 5.788248394686743e-06,
      "loss": 0.5588,
      "step": 1710
    },
    {
      "epoch": 1.97,
      "learning_rate": 5.67559792525163e-06,
      "loss": 0.5487,
      "step": 1720
    },
    {
      "epoch": 1.98,
      "learning_rate": 5.563618467422864e-06,
      "loss": 0.5579,
      "step": 1730
    },
    {
      "epoch": 1.99,
      "learning_rate": 5.452327396901078e-06,
      "loss": 0.5577,
      "step": 1740
    },
    {
      "epoch": 2.0,
      "learning_rate": 5.341741982570789e-06,
      "loss": 0.5519,
      "step": 1750
    },
    {
      "epoch": 2.01,
      "learning_rate": 5.231879383820782e-06,
      "loss": 0.5664,
      "step": 1760
    },
    {
      "epoch": 2.02,
      "learning_rate": 5.122756647881543e-06,
      "loss": 0.5572,
      "step": 1770
    },
    {
      "epoch": 2.03,
      "learning_rate": 5.0143907071800445e-06,
      "loss": 0.5516,
      "step": 1780
    },
    {
      "epoch": 2.05,
      "learning_rate": 4.9067983767123736e-06,
      "loss": 0.5498,
      "step": 1790
    },
    {
      "epoch": 2.06,
      "learning_rate": 4.79999635143455e-06,
      "loss": 0.571,
      "step": 1800
    },
    {
      "epoch": 2.06,
      "eval_loss": 0.5752249956130981,
      "eval_runtime": 764.7047,
      "eval_samples_per_second": 16.274,
      "eval_steps_per_second": 2.035,
      "step": 1800
    },
    {
      "epoch": 2.07,
      "learning_rate": 4.694001203672028e-06,
      "loss": 0.5661,
      "step": 1810
    },
    {
      "epoch": 2.08,
      "learning_rate": 4.588829380548168e-06,
      "loss": 0.5733,
      "step": 1820
    },
    {
      "epoch": 2.09,
      "learning_rate": 4.484497201432147e-06,
      "loss": 0.5725,
      "step": 1830
    },
    {
      "epoch": 2.1,
      "learning_rate": 4.381020855406738e-06,
      "loss": 0.556,
      "step": 1840
    },
    {
      "epoch": 2.11,
      "learning_rate": 4.278416398756245e-06,
      "loss": 0.5671,
      "step": 1850
    },
    {
      "epoch": 2.13,
      "learning_rate": 4.176699752475087e-06,
      "loss": 0.566,
      "step": 1860
    },
    {
      "epoch": 2.14,
      "learning_rate": 4.075886699797338e-06,
      "loss": 0.5692,
      "step": 1870
    },
    {
      "epoch": 2.15,
      "learning_rate": 3.975992883747688e-06,
      "loss": 0.5582,
      "step": 1880
    },
    {
      "epoch": 2.16,
      "learning_rate": 3.877033804714133e-06,
      "loss": 0.5477,
      "step": 1890
    },
    {
      "epoch": 2.17,
      "learning_rate": 3.7790248180427758e-06,
      "loss": 0.5488,
      "step": 1900
    },
    {
      "epoch": 2.18,
      "learning_rate": 3.6819811316551867e-06,
      "loss": 0.5591,
      "step": 1910
    },
    {
      "epoch": 2.19,
      "learning_rate": 3.585917803688603e-06,
      "loss": 0.5548,
      "step": 1920
    },
    {
      "epoch": 2.21,
      "learning_rate": 3.4908497401593634e-06,
      "loss": 0.5517,
      "step": 1930
    },
    {
      "epoch": 2.22,
      "learning_rate": 3.3967916926499823e-06,
      "loss": 0.556,
      "step": 1940
    },
    {
      "epoch": 2.23,
      "learning_rate": 3.303758256020154e-06,
      "loss": 0.5398,
      "step": 1950
    },
    {
      "epoch": 2.24,
      "learning_rate": 3.211763866142097e-06,
      "loss": 0.5487,
      "step": 1960
    },
    {
      "epoch": 2.25,
      "learning_rate": 3.1208227976605475e-06,
      "loss": 0.5503,
      "step": 1970
    },
    {
      "epoch": 2.26,
      "learning_rate": 3.0309491617777943e-06,
      "loss": 0.5609,
      "step": 1980
    },
    {
      "epoch": 2.27,
      "learning_rate": 2.9421569040640564e-06,
      "loss": 0.5569,
      "step": 1990
    },
    {
      "epoch": 2.29,
      "learning_rate": 2.854459802293561e-06,
      "loss": 0.5558,
      "step": 2000
    },
    {
      "epoch": 2.29,
      "eval_loss": 0.5727296471595764,
      "eval_runtime": 764.8947,
      "eval_samples_per_second": 16.27,
      "eval_steps_per_second": 2.034,
      "step": 2000
    },
    {
      "epoch": 2.3,
      "learning_rate": 2.767871464306682e-06,
      "loss": 0.5596,
      "step": 2010
    },
    {
      "epoch": 2.31,
      "learning_rate": 2.682405325898416e-06,
      "loss": 0.5565,
      "step": 2020
    },
    {
      "epoch": 2.32,
      "learning_rate": 2.5980746487335785e-06,
      "loss": 0.5482,
      "step": 2030
    },
    {
      "epoch": 2.33,
      "learning_rate": 2.514892518288988e-06,
      "loss": 0.559,
      "step": 2040
    },
    {
      "epoch": 2.34,
      "learning_rate": 2.432871841823047e-06,
      "loss": 0.5621,
      "step": 2050
    },
    {
      "epoch": 2.35,
      "learning_rate": 2.3520253463729058e-06,
      "loss": 0.5485,
      "step": 2060
    },
    {
      "epoch": 2.37,
      "learning_rate": 2.272365576779637e-06,
      "loss": 0.5598,
      "step": 2070
    },
    {
      "epoch": 2.38,
      "learning_rate": 2.1939048937416783e-06,
      "loss": 0.5655,
      "step": 2080
    },
    {
      "epoch": 2.39,
      "learning_rate": 2.116655471896828e-06,
      "loss": 0.5522,
      "step": 2090
    },
    {
      "epoch": 2.4,
      "learning_rate": 2.0406292979331333e-06,
      "loss": 0.552,
      "step": 2100
    },
    {
      "epoch": 2.41,
      "learning_rate": 1.965838168728912e-06,
      "loss": 0.5565,
      "step": 2110
    },
    {
      "epoch": 2.42,
      "learning_rate": 1.8922936895222798e-06,
      "loss": 0.5576,
      "step": 2120
    },
    {
      "epoch": 2.43,
      "learning_rate": 1.8200072721103545e-06,
      "loss": 0.5629,
      "step": 2130
    },
    {
      "epoch": 2.45,
      "learning_rate": 1.7489901330785053e-06,
      "loss": 0.5461,
      "step": 2140
    },
    {
      "epoch": 2.46,
      "learning_rate": 1.6792532920599025e-06,
      "loss": 0.5503,
      "step": 2150
    },
    {
      "epoch": 2.47,
      "learning_rate": 1.610807570025611e-06,
      "loss": 0.5422,
      "step": 2160
    },
    {
      "epoch": 2.48,
      "learning_rate": 1.5436635876055107e-06,
      "loss": 0.5537,
      "step": 2170
    },
    {
      "epoch": 2.49,
      "learning_rate": 1.4778317634403082e-06,
      "loss": 0.5676,
      "step": 2180
    },
    {
      "epoch": 2.5,
      "learning_rate": 1.4133223125648966e-06,
      "loss": 0.5664,
      "step": 2190
    },
    {
      "epoch": 2.51,
      "learning_rate": 1.350145244823302e-06,
      "loss": 0.5467,
      "step": 2200
    },
    {
      "epoch": 2.51,
      "eval_loss": 0.5713418126106262,
      "eval_runtime": 764.6146,
      "eval_samples_per_second": 16.276,
      "eval_steps_per_second": 2.035,
      "step": 2200
    }
  ],
  "logging_steps": 10,
  "max_steps": 2622,
  "num_train_epochs": 3,
  "save_steps": 200,
  "total_flos": 2.647958121057878e+18,
  "trial_name": null,
  "trial_params": null
}