{
  "best_metric": 0.500215470790863,
  "best_model_checkpoint": "./MT-deer-7B/checkpoint-1400",
  "epoch": 2.662546059669559,
  "global_step": 1400,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.02,
      "learning_rate": 2.9999999999999997e-05,
      "loss": 1.4291,
      "step": 10
    },
    {
      "epoch": 0.04,
      "learning_rate": 5.9999999999999995e-05,
      "loss": 1.3933,
      "step": 20
    },
    {
      "epoch": 0.06,
      "learning_rate": 8.999999999999999e-05,
      "loss": 1.3044,
      "step": 30
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.00011999999999999999,
      "loss": 1.1498,
      "step": 40
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.00015,
      "loss": 0.9306,
      "step": 50
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.00017999999999999998,
      "loss": 0.8534,
      "step": 60
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.00020999999999999998,
      "loss": 0.8292,
      "step": 70
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.00023999999999999998,
      "loss": 0.8054,
      "step": 80
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.00027,
      "loss": 0.7794,
      "step": 90
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0003,
      "loss": 0.7457,
      "step": 100
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0002979661016949152,
      "loss": 0.7358,
      "step": 110
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0002959322033898305,
      "loss": 0.729,
      "step": 120
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0002938983050847457,
      "loss": 0.7056,
      "step": 130
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.000291864406779661,
      "loss": 0.6986,
      "step": 140
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00028983050847457623,
      "loss": 0.6859,
      "step": 150
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0002877966101694915,
      "loss": 0.6639,
      "step": 160
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.00028576271186440673,
      "loss": 0.6708,
      "step": 170
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.000283728813559322,
      "loss": 0.6593,
      "step": 180
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00028169491525423723,
      "loss": 0.6497,
      "step": 190
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.0002796610169491525,
      "loss": 0.6432,
      "step": 200
    },
    {
      "epoch": 0.38,
      "eval_loss": 0.642841637134552,
      "eval_runtime": 92.942,
      "eval_samples_per_second": 21.519,
      "eval_steps_per_second": 2.69,
      "step": 200
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00027762711864406773,
      "loss": 0.6384,
      "step": 210
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00027559322033898304,
      "loss": 0.6344,
      "step": 220
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0002735593220338983,
      "loss": 0.6308,
      "step": 230
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.00027152542372881354,
      "loss": 0.6239,
      "step": 240
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.0002694915254237288,
      "loss": 0.6229,
      "step": 250
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00026745762711864404,
      "loss": 0.6149,
      "step": 260
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0002654237288135593,
      "loss": 0.6062,
      "step": 270
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.00026338983050847454,
      "loss": 0.615,
      "step": 280
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.0002613559322033898,
      "loss": 0.6043,
      "step": 290
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0002593220338983051,
      "loss": 0.6004,
      "step": 300
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00025728813559322035,
      "loss": 0.6058,
      "step": 310
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.0002552542372881356,
      "loss": 0.5902,
      "step": 320
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.00025322033898305085,
      "loss": 0.5912,
      "step": 330
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.0002511864406779661,
      "loss": 0.5988,
      "step": 340
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.00024915254237288135,
      "loss": 0.5806,
      "step": 350
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.0002471186440677966,
      "loss": 0.5879,
      "step": 360
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.00024508474576271185,
      "loss": 0.5846,
      "step": 370
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.0002430508474576271,
      "loss": 0.5789,
      "step": 380
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.00024101694915254235,
      "loss": 0.5787,
      "step": 390
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.0002389830508474576,
      "loss": 0.5802,
      "step": 400
    },
    {
      "epoch": 0.76,
      "eval_loss": 0.5802366137504578,
      "eval_runtime": 93.7341,
      "eval_samples_per_second": 21.337,
      "eval_steps_per_second": 2.667,
      "step": 400
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.00023694915254237286,
      "loss": 0.5854,
      "step": 410
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.0002349152542372881,
      "loss": 0.5757,
      "step": 420
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.00023288135593220336,
      "loss": 0.572,
      "step": 430
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.0002308474576271186,
      "loss": 0.5626,
      "step": 440
    },
    {
      "epoch": 0.86,
      "learning_rate": 0.00022881355932203386,
      "loss": 0.5605,
      "step": 450
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.0002267796610169491,
      "loss": 0.5656,
      "step": 460
    },
    {
      "epoch": 0.89,
      "learning_rate": 0.0002247457627118644,
      "loss": 0.5592,
      "step": 470
    },
    {
      "epoch": 0.91,
      "learning_rate": 0.00022271186440677964,
      "loss": 0.5639,
      "step": 480
    },
    {
      "epoch": 0.93,
      "learning_rate": 0.0002206779661016949,
      "loss": 0.5616,
      "step": 490
    },
    {
      "epoch": 0.95,
      "learning_rate": 0.00021864406779661014,
      "loss": 0.5677,
      "step": 500
    },
    {
      "epoch": 0.97,
      "learning_rate": 0.00021661016949152542,
      "loss": 0.5562,
      "step": 510
    },
    {
      "epoch": 0.99,
      "learning_rate": 0.00021457627118644067,
      "loss": 0.5535,
      "step": 520
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.00021254237288135592,
      "loss": 0.5604,
      "step": 530
    },
    {
      "epoch": 1.03,
      "learning_rate": 0.00021050847457627117,
      "loss": 0.5507,
      "step": 540
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.00020847457627118642,
      "loss": 0.5434,
      "step": 550
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.0002064406779661017,
      "loss": 0.5517,
      "step": 560
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.00020440677966101695,
      "loss": 0.5459,
      "step": 570
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.0002023728813559322,
      "loss": 0.548,
      "step": 580
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.00020033898305084745,
      "loss": 0.5463,
      "step": 590
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.0001983050847457627,
      "loss": 0.5402,
      "step": 600
    },
    {
      "epoch": 1.14,
      "eval_loss": 0.5485313534736633,
      "eval_runtime": 93.9991,
      "eval_samples_per_second": 21.277,
      "eval_steps_per_second": 2.66,
      "step": 600
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.00019627118644067795,
      "loss": 0.541,
      "step": 610
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.0001942372881355932,
      "loss": 0.5471,
      "step": 620
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.00019220338983050845,
      "loss": 0.536,
      "step": 630
    },
    {
      "epoch": 1.22,
      "learning_rate": 0.0001901694915254237,
      "loss": 0.5386,
      "step": 640
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.00018813559322033895,
      "loss": 0.5337,
      "step": 650
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.0001861016949152542,
      "loss": 0.5351,
      "step": 660
    },
    {
      "epoch": 1.27,
      "learning_rate": 0.00018406779661016948,
      "loss": 0.5225,
      "step": 670
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.00018203389830508473,
      "loss": 0.5366,
      "step": 680
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.00017999999999999998,
      "loss": 0.5323,
      "step": 690
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.00017796610169491523,
      "loss": 0.5316,
      "step": 700
    },
    {
      "epoch": 1.35,
      "learning_rate": 0.00017593220338983049,
      "loss": 0.5385,
      "step": 710
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.00017389830508474574,
      "loss": 0.5293,
      "step": 720
    },
    {
      "epoch": 1.39,
      "learning_rate": 0.000171864406779661,
      "loss": 0.5233,
      "step": 730
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.00016983050847457624,
      "loss": 0.5135,
      "step": 740
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.0001677966101694915,
      "loss": 0.5266,
      "step": 750
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.00016576271186440674,
      "loss": 0.5266,
      "step": 760
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.000163728813559322,
      "loss": 0.5217,
      "step": 770
    },
    {
      "epoch": 1.48,
      "learning_rate": 0.0001616949152542373,
      "loss": 0.5149,
      "step": 780
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.00015966101694915255,
      "loss": 0.5239,
      "step": 790
    },
    {
      "epoch": 1.52,
      "learning_rate": 0.0001576271186440678,
      "loss": 0.5202,
      "step": 800
    },
    {
      "epoch": 1.52,
      "eval_loss": 0.5286936163902283,
      "eval_runtime": 94.1069,
      "eval_samples_per_second": 21.252,
      "eval_steps_per_second": 2.657,
      "step": 800
    },
    {
      "epoch": 1.54,
      "learning_rate": 0.00015559322033898305,
      "loss": 0.5321,
      "step": 810
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.0001535593220338983,
      "loss": 0.5203,
      "step": 820
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.00015152542372881355,
      "loss": 0.5205,
      "step": 830
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0001494915254237288,
      "loss": 0.5255,
      "step": 840
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.00014745762711864405,
      "loss": 0.5164,
      "step": 850
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.00014542372881355933,
      "loss": 0.5182,
      "step": 860
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.00014338983050847458,
      "loss": 0.5171,
      "step": 870
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.00014135593220338983,
      "loss": 0.5172,
      "step": 880
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.00013932203389830508,
      "loss": 0.5171,
      "step": 890
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.00013728813559322033,
      "loss": 0.5114,
      "step": 900
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.00013525423728813558,
      "loss": 0.5079,
      "step": 910
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.00013322033898305083,
      "loss": 0.5158,
      "step": 920
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.00013118644067796608,
      "loss": 0.5128,
      "step": 930
    },
    {
      "epoch": 1.79,
      "learning_rate": 0.00012915254237288133,
      "loss": 0.5198,
      "step": 940
    },
    {
      "epoch": 1.81,
      "learning_rate": 0.00012711864406779658,
      "loss": 0.5063,
      "step": 950
    },
    {
      "epoch": 1.83,
      "learning_rate": 0.00012508474576271184,
      "loss": 0.5068,
      "step": 960
    },
    {
      "epoch": 1.84,
      "learning_rate": 0.00012305084745762709,
      "loss": 0.5136,
      "step": 970
    },
    {
      "epoch": 1.86,
      "learning_rate": 0.00012101694915254236,
      "loss": 0.5012,
      "step": 980
    },
    {
      "epoch": 1.88,
      "learning_rate": 0.00011898305084745763,
      "loss": 0.5071,
      "step": 990
    },
    {
      "epoch": 1.9,
      "learning_rate": 0.00011694915254237288,
      "loss": 0.5092,
      "step": 1000
    },
    {
      "epoch": 1.9,
      "eval_loss": 0.514962375164032,
      "eval_runtime": 93.8613,
      "eval_samples_per_second": 21.308,
      "eval_steps_per_second": 2.664,
      "step": 1000
    },
    {
      "epoch": 1.92,
      "learning_rate": 0.00011491525423728813,
      "loss": 0.5185,
      "step": 1010
    },
    {
      "epoch": 1.94,
      "learning_rate": 0.00011288135593220338,
      "loss": 0.5055,
      "step": 1020
    },
    {
      "epoch": 1.96,
      "learning_rate": 0.00011084745762711863,
      "loss": 0.5078,
      "step": 1030
    },
    {
      "epoch": 1.98,
      "learning_rate": 0.00010881355932203388,
      "loss": 0.5092,
      "step": 1040
    },
    {
      "epoch": 2.0,
      "learning_rate": 0.00010677966101694915,
      "loss": 0.4933,
      "step": 1050
    },
    {
      "epoch": 2.02,
      "learning_rate": 0.0001047457627118644,
      "loss": 0.5052,
      "step": 1060
    },
    {
      "epoch": 2.03,
      "learning_rate": 0.00010271186440677965,
      "loss": 0.4914,
      "step": 1070
    },
    {
      "epoch": 2.05,
      "learning_rate": 0.0001006779661016949,
      "loss": 0.4962,
      "step": 1080
    },
    {
      "epoch": 2.07,
      "learning_rate": 9.864406779661015e-05,
      "loss": 0.4969,
      "step": 1090
    },
    {
      "epoch": 2.09,
      "learning_rate": 9.661016949152541e-05,
      "loss": 0.5053,
      "step": 1100
    },
    {
      "epoch": 2.11,
      "learning_rate": 9.457627118644068e-05,
      "loss": 0.4934,
      "step": 1110
    },
    {
      "epoch": 2.13,
      "learning_rate": 9.254237288135593e-05,
      "loss": 0.4864,
      "step": 1120
    },
    {
      "epoch": 2.15,
      "learning_rate": 9.050847457627118e-05,
      "loss": 0.4974,
      "step": 1130
    },
    {
      "epoch": 2.17,
      "learning_rate": 8.847457627118644e-05,
      "loss": 0.4985,
      "step": 1140
    },
    {
      "epoch": 2.19,
      "learning_rate": 8.64406779661017e-05,
      "loss": 0.4945,
      "step": 1150
    },
    {
      "epoch": 2.21,
      "learning_rate": 8.440677966101695e-05,
      "loss": 0.4986,
      "step": 1160
    },
    {
      "epoch": 2.23,
      "learning_rate": 8.23728813559322e-05,
      "loss": 0.4887,
      "step": 1170
    },
    {
      "epoch": 2.24,
      "learning_rate": 8.033898305084745e-05,
      "loss": 0.5004,
      "step": 1180
    },
    {
      "epoch": 2.26,
      "learning_rate": 7.83050847457627e-05,
      "loss": 0.486,
      "step": 1190
    },
    {
      "epoch": 2.28,
      "learning_rate": 7.627118644067796e-05,
      "loss": 0.4863,
      "step": 1200
    },
    {
      "epoch": 2.28,
      "eval_loss": 0.5064148902893066,
      "eval_runtime": 93.4587,
      "eval_samples_per_second": 21.4,
      "eval_steps_per_second": 2.675,
      "step": 1200
    },
    {
      "epoch": 2.3,
      "learning_rate": 7.423728813559321e-05,
      "loss": 0.4959,
      "step": 1210
    },
    {
      "epoch": 2.32,
      "learning_rate": 7.220338983050848e-05,
      "loss": 0.4961,
      "step": 1220
    },
    {
      "epoch": 2.34,
      "learning_rate": 7.016949152542373e-05,
      "loss": 0.4902,
      "step": 1230
    },
    {
      "epoch": 2.36,
      "learning_rate": 6.813559322033898e-05,
      "loss": 0.4957,
      "step": 1240
    },
    {
      "epoch": 2.38,
      "learning_rate": 6.610169491525423e-05,
      "loss": 0.4998,
      "step": 1250
    },
    {
      "epoch": 2.4,
      "learning_rate": 6.406779661016948e-05,
      "loss": 0.4945,
      "step": 1260
    },
    {
      "epoch": 2.42,
      "learning_rate": 6.203389830508473e-05,
      "loss": 0.4914,
      "step": 1270
    },
    {
      "epoch": 2.43,
      "learning_rate": 5.9999999999999995e-05,
      "loss": 0.4984,
      "step": 1280
    },
    {
      "epoch": 2.45,
      "learning_rate": 5.796610169491525e-05,
      "loss": 0.4926,
      "step": 1290
    },
    {
      "epoch": 2.47,
      "learning_rate": 5.59322033898305e-05,
      "loss": 0.4891,
      "step": 1300
    },
    {
      "epoch": 2.49,
      "learning_rate": 5.389830508474576e-05,
      "loss": 0.4895,
      "step": 1310
    },
    {
      "epoch": 2.51,
      "learning_rate": 5.186440677966101e-05,
      "loss": 0.4817,
      "step": 1320
    },
    {
      "epoch": 2.53,
      "learning_rate": 4.983050847457626e-05,
      "loss": 0.491,
      "step": 1330
    },
    {
      "epoch": 2.55,
      "learning_rate": 4.779661016949152e-05,
      "loss": 0.4878,
      "step": 1340
    },
    {
      "epoch": 2.57,
      "learning_rate": 4.576271186440678e-05,
      "loss": 0.4869,
      "step": 1350
    },
    {
      "epoch": 2.59,
      "learning_rate": 4.3728813559322035e-05,
      "loss": 0.488,
      "step": 1360
    },
    {
      "epoch": 2.61,
      "learning_rate": 4.1694915254237285e-05,
      "loss": 0.4866,
      "step": 1370
    },
    {
      "epoch": 2.62,
      "learning_rate": 3.9661016949152536e-05,
      "loss": 0.4892,
      "step": 1380
    },
    {
      "epoch": 2.64,
      "learning_rate": 3.7627118644067794e-05,
      "loss": 0.486,
      "step": 1390
    },
    {
      "epoch": 2.66,
      "learning_rate": 3.559322033898305e-05,
      "loss": 0.4879,
      "step": 1400
    },
    {
      "epoch": 2.66,
      "eval_loss": 0.500215470790863,
      "eval_runtime": 93.9201,
      "eval_samples_per_second": 21.295,
      "eval_steps_per_second": 2.662,
      "step": 1400
    }
  ],
  "max_steps": 1575,
  "num_train_epochs": 3,
  "total_flos": 1.8195722467776922e+18,
  "trial_name": null,
  "trial_params": null
}