{
  "best_metric": 2.700421094894409,
  "best_model_checkpoint": "cdetr-mist1-brain-gt-tumors-8ah-6l/checkpoint-9890",
  "epoch": 100.0,
  "eval_steps": 500,
  "global_step": 11500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "learning_rate": 9.624347826086956e-06,
      "loss": 5.4149,
      "step": 115
    },
    {
      "epoch": 1.0,
      "eval_loss": 4.397365093231201,
      "eval_runtime": 5.8932,
      "eval_samples_per_second": 6.787,
      "eval_steps_per_second": 0.848,
      "step": 115
    },
    {
      "epoch": 2.0,
      "learning_rate": 9.231304347826087e-06,
      "loss": 3.9453,
      "step": 230
    },
    {
      "epoch": 2.0,
      "eval_loss": 3.651952028274536,
      "eval_runtime": 5.9289,
      "eval_samples_per_second": 6.747,
      "eval_steps_per_second": 0.843,
      "step": 230
    },
    {
      "epoch": 3.0,
      "learning_rate": 8.831304347826088e-06,
      "loss": 3.7269,
      "step": 345
    },
    {
      "epoch": 3.0,
      "eval_loss": 3.7602438926696777,
      "eval_runtime": 5.8799,
      "eval_samples_per_second": 6.803,
      "eval_steps_per_second": 0.85,
      "step": 345
    },
    {
      "epoch": 4.0,
      "learning_rate": 8.434782608695653e-06,
      "loss": 3.5898,
      "step": 460
    },
    {
      "epoch": 4.0,
      "eval_loss": 3.5670828819274902,
      "eval_runtime": 5.9645,
      "eval_samples_per_second": 6.706,
      "eval_steps_per_second": 0.838,
      "step": 460
    },
    {
      "epoch": 5.0,
      "learning_rate": 8.034782608695652e-06,
      "loss": 3.486,
      "step": 575
    },
    {
      "epoch": 5.0,
      "eval_loss": 3.4911675453186035,
      "eval_runtime": 5.9241,
      "eval_samples_per_second": 6.752,
      "eval_steps_per_second": 0.844,
      "step": 575
    },
    {
      "epoch": 6.0,
      "learning_rate": 7.634782608695653e-06,
      "loss": 3.4073,
      "step": 690
    },
    {
      "epoch": 6.0,
      "eval_loss": 3.409463405609131,
      "eval_runtime": 5.8302,
      "eval_samples_per_second": 6.861,
      "eval_steps_per_second": 0.858,
      "step": 690
    },
    {
      "epoch": 7.0,
      "learning_rate": 7.234782608695653e-06,
      "loss": 3.4181,
      "step": 805
    },
    {
      "epoch": 7.0,
      "eval_loss": 3.3183083534240723,
      "eval_runtime": 5.8894,
      "eval_samples_per_second": 6.792,
      "eval_steps_per_second": 0.849,
      "step": 805
    },
    {
      "epoch": 8.0,
      "learning_rate": 6.834782608695653e-06,
      "loss": 3.3603,
      "step": 920
    },
    {
      "epoch": 8.0,
      "eval_loss": 3.1111273765563965,
      "eval_runtime": 5.9175,
      "eval_samples_per_second": 6.76,
      "eval_steps_per_second": 0.845,
      "step": 920
    },
    {
      "epoch": 9.0,
      "learning_rate": 6.434782608695652e-06,
      "loss": 3.2777,
      "step": 1035
    },
    {
      "epoch": 9.0,
      "eval_loss": 3.1992499828338623,
      "eval_runtime": 5.9275,
      "eval_samples_per_second": 6.748,
      "eval_steps_per_second": 0.844,
      "step": 1035
    },
    {
      "epoch": 10.0,
      "learning_rate": 6.034782608695653e-06,
      "loss": 3.2851,
      "step": 1150
    },
    {
      "epoch": 10.0,
      "eval_loss": 3.3997130393981934,
      "eval_runtime": 5.8939,
      "eval_samples_per_second": 6.787,
      "eval_steps_per_second": 0.848,
      "step": 1150
    },
    {
      "epoch": 11.0,
      "learning_rate": 5.634782608695652e-06,
      "loss": 3.266,
      "step": 1265
    },
    {
      "epoch": 11.0,
      "eval_loss": 3.286081314086914,
      "eval_runtime": 6.1287,
      "eval_samples_per_second": 6.527,
      "eval_steps_per_second": 0.816,
      "step": 1265
    },
    {
      "epoch": 12.0,
      "learning_rate": 5.234782608695652e-06,
      "loss": 3.2803,
      "step": 1380
    },
    {
      "epoch": 12.0,
      "eval_loss": 3.18131947517395,
      "eval_runtime": 6.1123,
      "eval_samples_per_second": 6.544,
      "eval_steps_per_second": 0.818,
      "step": 1380
    },
    {
      "epoch": 13.0,
      "learning_rate": 4.834782608695652e-06,
      "loss": 3.1733,
      "step": 1495
    },
    {
      "epoch": 13.0,
      "eval_loss": 2.9837594032287598,
      "eval_runtime": 5.925,
      "eval_samples_per_second": 6.751,
      "eval_steps_per_second": 0.844,
      "step": 1495
    },
    {
      "epoch": 14.0,
      "learning_rate": 4.434782608695653e-06,
      "loss": 3.2094,
      "step": 1610
    },
    {
      "epoch": 14.0,
      "eval_loss": 3.1175174713134766,
      "eval_runtime": 5.9137,
      "eval_samples_per_second": 6.764,
      "eval_steps_per_second": 0.845,
      "step": 1610
    },
    {
      "epoch": 15.0,
      "learning_rate": 4.034782608695652e-06,
      "loss": 3.1718,
      "step": 1725
    },
    {
      "epoch": 15.0,
      "eval_loss": 3.0063586235046387,
      "eval_runtime": 5.9077,
      "eval_samples_per_second": 6.771,
      "eval_steps_per_second": 0.846,
      "step": 1725
    },
    {
      "epoch": 16.0,
      "learning_rate": 3.6347826086956524e-06,
      "loss": 3.1303,
      "step": 1840
    },
    {
      "epoch": 16.0,
      "eval_loss": 3.086920738220215,
      "eval_runtime": 5.9305,
      "eval_samples_per_second": 6.745,
      "eval_steps_per_second": 0.843,
      "step": 1840
    },
    {
      "epoch": 17.0,
      "learning_rate": 3.2347826086956524e-06,
      "loss": 3.0897,
      "step": 1955
    },
    {
      "epoch": 17.0,
      "eval_loss": 3.0305533409118652,
      "eval_runtime": 6.0975,
      "eval_samples_per_second": 6.56,
      "eval_steps_per_second": 0.82,
      "step": 1955
    },
    {
      "epoch": 18.0,
      "learning_rate": 2.8347826086956525e-06,
      "loss": 3.0233,
      "step": 2070
    },
    {
      "epoch": 18.0,
      "eval_loss": 2.9478588104248047,
      "eval_runtime": 5.9801,
      "eval_samples_per_second": 6.689,
      "eval_steps_per_second": 0.836,
      "step": 2070
    },
    {
      "epoch": 19.0,
      "learning_rate": 2.4347826086956525e-06,
      "loss": 3.0156,
      "step": 2185
    },
    {
      "epoch": 19.0,
      "eval_loss": 2.914536952972412,
      "eval_runtime": 6.0155,
      "eval_samples_per_second": 6.649,
      "eval_steps_per_second": 0.831,
      "step": 2185
    },
    {
      "epoch": 20.0,
      "learning_rate": 2.0347826086956525e-06,
      "loss": 3.0277,
      "step": 2300
    },
    {
      "epoch": 20.0,
      "eval_loss": 2.891925811767578,
      "eval_runtime": 6.058,
      "eval_samples_per_second": 6.603,
      "eval_steps_per_second": 0.825,
      "step": 2300
    },
    {
      "epoch": 21.0,
      "learning_rate": 1.6382608695652176e-06,
      "loss": 3.0847,
      "step": 2415
    },
    {
      "epoch": 21.0,
      "eval_loss": 2.9321396350860596,
      "eval_runtime": 6.1412,
      "eval_samples_per_second": 6.513,
      "eval_steps_per_second": 0.814,
      "step": 2415
    },
    {
      "epoch": 22.0,
      "learning_rate": 1.2382608695652176e-06,
      "loss": 3.0333,
      "step": 2530
    },
    {
      "epoch": 22.0,
      "eval_loss": 2.91283917427063,
      "eval_runtime": 6.139,
      "eval_samples_per_second": 6.516,
      "eval_steps_per_second": 0.814,
      "step": 2530
    },
    {
      "epoch": 23.0,
      "learning_rate": 8.382608695652175e-07,
      "loss": 3.0126,
      "step": 2645
    },
    {
      "epoch": 23.0,
      "eval_loss": 2.8627383708953857,
      "eval_runtime": 6.0565,
      "eval_samples_per_second": 6.604,
      "eval_steps_per_second": 0.826,
      "step": 2645
    },
    {
      "epoch": 24.0,
      "learning_rate": 5.234782608695652e-06,
      "loss": 2.9968,
      "step": 2760
    },
    {
      "epoch": 24.0,
      "eval_loss": 3.0186123847961426,
      "eval_runtime": 5.7749,
      "eval_samples_per_second": 6.926,
      "eval_steps_per_second": 0.866,
      "step": 2760
    },
    {
      "epoch": 25.0,
      "learning_rate": 5.036521739130436e-06,
      "loss": 3.0295,
      "step": 2875
    },
    {
      "epoch": 25.0,
      "eval_loss": 3.0147616863250732,
      "eval_runtime": 5.799,
      "eval_samples_per_second": 6.898,
      "eval_steps_per_second": 0.862,
      "step": 2875
    },
    {
      "epoch": 26.0,
      "learning_rate": 4.836521739130435e-06,
      "loss": 3.0294,
      "step": 2990
    },
    {
      "epoch": 26.0,
      "eval_loss": 3.034066915512085,
      "eval_runtime": 5.8474,
      "eval_samples_per_second": 6.841,
      "eval_steps_per_second": 0.855,
      "step": 2990
    },
    {
      "epoch": 27.0,
      "learning_rate": 4.636521739130435e-06,
      "loss": 3.0395,
      "step": 3105
    },
    {
      "epoch": 27.0,
      "eval_loss": 2.99969220161438,
      "eval_runtime": 5.7579,
      "eval_samples_per_second": 6.947,
      "eval_steps_per_second": 0.868,
      "step": 3105
    },
    {
      "epoch": 28.0,
      "learning_rate": 4.436521739130435e-06,
      "loss": 3.0445,
      "step": 3220
    },
    {
      "epoch": 28.0,
      "eval_loss": 3.0574753284454346,
      "eval_runtime": 5.6888,
      "eval_samples_per_second": 7.031,
      "eval_steps_per_second": 0.879,
      "step": 3220
    },
    {
      "epoch": 29.0,
      "learning_rate": 4.236521739130436e-06,
      "loss": 2.9761,
      "step": 3335
    },
    {
      "epoch": 29.0,
      "eval_loss": 2.97066068649292,
      "eval_runtime": 5.7458,
      "eval_samples_per_second": 6.962,
      "eval_steps_per_second": 0.87,
      "step": 3335
    },
    {
      "epoch": 30.0,
      "learning_rate": 4.038260869565218e-06,
      "loss": 3.0075,
      "step": 3450
    },
    {
      "epoch": 30.0,
      "eval_loss": 2.939246416091919,
      "eval_runtime": 5.722,
      "eval_samples_per_second": 6.991,
      "eval_steps_per_second": 0.874,
      "step": 3450
    },
    {
      "epoch": 31.0,
      "learning_rate": 3.838260869565218e-06,
      "loss": 3.0198,
      "step": 3565
    },
    {
      "epoch": 31.0,
      "eval_loss": 2.9121549129486084,
      "eval_runtime": 5.8674,
      "eval_samples_per_second": 6.817,
      "eval_steps_per_second": 0.852,
      "step": 3565
    },
    {
      "epoch": 32.0,
      "learning_rate": 3.6382608695652177e-06,
      "loss": 2.9782,
      "step": 3680
    },
    {
      "epoch": 32.0,
      "eval_loss": 2.947079658508301,
      "eval_runtime": 5.7346,
      "eval_samples_per_second": 6.975,
      "eval_steps_per_second": 0.872,
      "step": 3680
    },
    {
      "epoch": 33.0,
      "learning_rate": 3.4382608695652175e-06,
      "loss": 2.9773,
      "step": 3795
    },
    {
      "epoch": 33.0,
      "eval_loss": 3.030561923980713,
      "eval_runtime": 5.7197,
      "eval_samples_per_second": 6.993,
      "eval_steps_per_second": 0.874,
      "step": 3795
    },
    {
      "epoch": 34.0,
      "learning_rate": 3.2382608695652177e-06,
      "loss": 2.9528,
      "step": 3910
    },
    {
      "epoch": 34.0,
      "eval_loss": 2.8513386249542236,
      "eval_runtime": 5.6599,
      "eval_samples_per_second": 7.067,
      "eval_steps_per_second": 0.883,
      "step": 3910
    },
    {
      "epoch": 35.0,
      "learning_rate": 3.038260869565218e-06,
      "loss": 2.9228,
      "step": 4025
    },
    {
      "epoch": 35.0,
      "eval_loss": 2.899709701538086,
      "eval_runtime": 5.6812,
      "eval_samples_per_second": 7.041,
      "eval_steps_per_second": 0.88,
      "step": 4025
    },
    {
      "epoch": 36.0,
      "learning_rate": 2.8382608695652177e-06,
      "loss": 2.9221,
      "step": 4140
    },
    {
      "epoch": 36.0,
      "eval_loss": 2.8646340370178223,
      "eval_runtime": 5.6561,
      "eval_samples_per_second": 7.072,
      "eval_steps_per_second": 0.884,
      "step": 4140
    },
    {
      "epoch": 37.0,
      "learning_rate": 2.638260869565218e-06,
      "loss": 2.8933,
      "step": 4255
    },
    {
      "epoch": 37.0,
      "eval_loss": 2.8870697021484375,
      "eval_runtime": 5.7007,
      "eval_samples_per_second": 7.017,
      "eval_steps_per_second": 0.877,
      "step": 4255
    },
    {
      "epoch": 38.0,
      "learning_rate": 2.4382608695652177e-06,
      "loss": 2.8925,
      "step": 4370
    },
    {
      "epoch": 38.0,
      "eval_loss": 2.94067120552063,
      "eval_runtime": 5.6567,
      "eval_samples_per_second": 7.071,
      "eval_steps_per_second": 0.884,
      "step": 4370
    },
    {
      "epoch": 39.0,
      "learning_rate": 2.2382608695652175e-06,
      "loss": 2.9069,
      "step": 4485
    },
    {
      "epoch": 39.0,
      "eval_loss": 2.9625325202941895,
      "eval_runtime": 5.6615,
      "eval_samples_per_second": 7.065,
      "eval_steps_per_second": 0.883,
      "step": 4485
    },
    {
      "epoch": 40.0,
      "learning_rate": 2.0382608695652177e-06,
      "loss": 2.9246,
      "step": 4600
    },
    {
      "epoch": 40.0,
      "eval_loss": 2.994633197784424,
      "eval_runtime": 5.6894,
      "eval_samples_per_second": 7.031,
      "eval_steps_per_second": 0.879,
      "step": 4600
    },
    {
      "epoch": 41.0,
      "learning_rate": 1.8382608695652178e-06,
      "loss": 2.9089,
      "step": 4715
    },
    {
      "epoch": 41.0,
      "eval_loss": 2.8935651779174805,
      "eval_runtime": 5.7226,
      "eval_samples_per_second": 6.99,
      "eval_steps_per_second": 0.874,
      "step": 4715
    },
    {
      "epoch": 42.0,
      "learning_rate": 1.6382608695652176e-06,
      "loss": 2.8573,
      "step": 4830
    },
    {
      "epoch": 42.0,
      "eval_loss": 2.827199935913086,
      "eval_runtime": 5.6411,
      "eval_samples_per_second": 7.091,
      "eval_steps_per_second": 0.886,
      "step": 4830
    },
    {
      "epoch": 43.0,
      "learning_rate": 5.726956521739131e-06,
      "loss": 2.8768,
      "step": 4945
    },
    {
      "epoch": 43.0,
      "eval_loss": 2.9868218898773193,
      "eval_runtime": 5.7349,
      "eval_samples_per_second": 6.975,
      "eval_steps_per_second": 0.872,
      "step": 4945
    },
    {
      "epoch": 44.0,
      "learning_rate": 5.627826086956523e-06,
      "loss": 2.9666,
      "step": 5060
    },
    {
      "epoch": 44.0,
      "eval_loss": 2.9200377464294434,
      "eval_runtime": 5.6461,
      "eval_samples_per_second": 7.084,
      "eval_steps_per_second": 0.886,
      "step": 5060
    },
    {
      "epoch": 45.0,
      "learning_rate": 5.527826086956523e-06,
      "loss": 2.958,
      "step": 5175
    },
    {
      "epoch": 45.0,
      "eval_loss": 2.875474214553833,
      "eval_runtime": 5.664,
      "eval_samples_per_second": 7.062,
      "eval_steps_per_second": 0.883,
      "step": 5175
    },
    {
      "epoch": 46.0,
      "learning_rate": 5.427826086956523e-06,
      "loss": 2.8923,
      "step": 5290
    },
    {
      "epoch": 46.0,
      "eval_loss": 2.851766586303711,
      "eval_runtime": 5.6855,
      "eval_samples_per_second": 7.035,
      "eval_steps_per_second": 0.879,
      "step": 5290
    },
    {
      "epoch": 47.0,
      "learning_rate": 5.327826086956522e-06,
      "loss": 2.9204,
      "step": 5405
    },
    {
      "epoch": 47.0,
      "eval_loss": 2.9000306129455566,
      "eval_runtime": 5.6182,
      "eval_samples_per_second": 7.12,
      "eval_steps_per_second": 0.89,
      "step": 5405
    },
    {
      "epoch": 48.0,
      "learning_rate": 5.227826086956522e-06,
      "loss": 2.9644,
      "step": 5520
    },
    {
      "epoch": 48.0,
      "eval_loss": 2.8968658447265625,
      "eval_runtime": 5.677,
      "eval_samples_per_second": 7.046,
      "eval_steps_per_second": 0.881,
      "step": 5520
    },
    {
      "epoch": 49.0,
      "learning_rate": 5.127826086956522e-06,
      "loss": 2.9011,
      "step": 5635
    },
    {
      "epoch": 49.0,
      "eval_loss": 2.7918035984039307,
      "eval_runtime": 5.6945,
      "eval_samples_per_second": 7.024,
      "eval_steps_per_second": 0.878,
      "step": 5635
    },
    {
      "epoch": 50.0,
      "learning_rate": 5.028695652173914e-06,
      "loss": 2.9329,
      "step": 5750
    },
    {
      "epoch": 50.0,
      "eval_loss": 2.9139397144317627,
      "eval_runtime": 5.6814,
      "eval_samples_per_second": 7.041,
      "eval_steps_per_second": 0.88,
      "step": 5750
    },
    {
      "epoch": 51.0,
      "learning_rate": 4.9286956521739135e-06,
      "loss": 2.9031,
      "step": 5865
    },
    {
      "epoch": 51.0,
      "eval_loss": 2.779574155807495,
      "eval_runtime": 5.6982,
      "eval_samples_per_second": 7.02,
      "eval_steps_per_second": 0.877,
      "step": 5865
    },
    {
      "epoch": 52.0,
      "learning_rate": 4.828695652173914e-06,
      "loss": 2.9029,
      "step": 5980
    },
    {
      "epoch": 52.0,
      "eval_loss": 2.802475690841675,
      "eval_runtime": 5.6076,
      "eval_samples_per_second": 7.133,
      "eval_steps_per_second": 0.892,
      "step": 5980
    },
    {
      "epoch": 53.0,
      "learning_rate": 4.728695652173914e-06,
      "loss": 2.9555,
      "step": 6095
    },
    {
      "epoch": 53.0,
      "eval_loss": 2.9120869636535645,
      "eval_runtime": 5.7682,
      "eval_samples_per_second": 6.935,
      "eval_steps_per_second": 0.867,
      "step": 6095
    },
    {
      "epoch": 54.0,
      "learning_rate": 4.628695652173914e-06,
      "loss": 2.9366,
      "step": 6210
    },
    {
      "epoch": 54.0,
      "eval_loss": 2.9034695625305176,
      "eval_runtime": 5.7147,
      "eval_samples_per_second": 7.0,
      "eval_steps_per_second": 0.875,
      "step": 6210
    },
    {
      "epoch": 55.0,
      "learning_rate": 4.528695652173913e-06,
      "loss": 2.8871,
      "step": 6325
    },
    {
      "epoch": 55.0,
      "eval_loss": 2.87589168548584,
      "eval_runtime": 5.6683,
      "eval_samples_per_second": 7.057,
      "eval_steps_per_second": 0.882,
      "step": 6325
    },
    {
      "epoch": 56.0,
      "learning_rate": 4.428695652173913e-06,
      "loss": 2.863,
      "step": 6440
    },
    {
      "epoch": 56.0,
      "eval_loss": 2.8540170192718506,
      "eval_runtime": 5.7794,
      "eval_samples_per_second": 6.921,
      "eval_steps_per_second": 0.865,
      "step": 6440
    },
    {
      "epoch": 57.0,
      "learning_rate": 4.328695652173913e-06,
      "loss": 2.8897,
      "step": 6555
    },
    {
      "epoch": 57.0,
      "eval_loss": 2.8401310443878174,
      "eval_runtime": 5.6374,
      "eval_samples_per_second": 7.095,
      "eval_steps_per_second": 0.887,
      "step": 6555
    },
    {
      "epoch": 58.0,
      "learning_rate": 4.228695652173913e-06,
      "loss": 2.828,
      "step": 6670
    },
    {
      "epoch": 58.0,
      "eval_loss": 2.8589885234832764,
      "eval_runtime": 5.697,
      "eval_samples_per_second": 7.021,
      "eval_steps_per_second": 0.878,
      "step": 6670
    },
    {
      "epoch": 59.0,
      "learning_rate": 4.1286956521739135e-06,
      "loss": 2.8221,
      "step": 6785
    },
    {
      "epoch": 59.0,
      "eval_loss": 2.92549467086792,
      "eval_runtime": 5.7309,
      "eval_samples_per_second": 6.98,
      "eval_steps_per_second": 0.872,
      "step": 6785
    },
    {
      "epoch": 60.0,
      "learning_rate": 4.028695652173914e-06,
      "loss": 2.835,
      "step": 6900
    },
    {
      "epoch": 60.0,
      "eval_loss": 2.9808831214904785,
      "eval_runtime": 5.6748,
      "eval_samples_per_second": 7.049,
      "eval_steps_per_second": 0.881,
      "step": 6900
    },
    {
      "epoch": 61.0,
      "learning_rate": 3.928695652173914e-06,
      "loss": 2.886,
      "step": 7015
    },
    {
      "epoch": 61.0,
      "eval_loss": 2.9906742572784424,
      "eval_runtime": 5.7118,
      "eval_samples_per_second": 7.003,
      "eval_steps_per_second": 0.875,
      "step": 7015
    },
    {
      "epoch": 62.0,
      "learning_rate": 3.828695652173913e-06,
      "loss": 2.8227,
      "step": 7130
    },
    {
      "epoch": 62.0,
      "eval_loss": 2.8283145427703857,
      "eval_runtime": 5.7014,
      "eval_samples_per_second": 7.016,
      "eval_steps_per_second": 0.877,
      "step": 7130
    },
    {
      "epoch": 63.0,
      "learning_rate": 3.728695652173913e-06,
      "loss": 2.7864,
      "step": 7245
    },
    {
      "epoch": 63.0,
      "eval_loss": 2.8258347511291504,
      "eval_runtime": 5.6903,
      "eval_samples_per_second": 7.029,
      "eval_steps_per_second": 0.879,
      "step": 7245
    },
    {
      "epoch": 64.0,
      "learning_rate": 3.6286956521739132e-06,
      "loss": 2.8179,
      "step": 7360
    },
    {
      "epoch": 64.0,
      "eval_loss": 2.9504449367523193,
      "eval_runtime": 5.7303,
      "eval_samples_per_second": 6.98,
      "eval_steps_per_second": 0.873,
      "step": 7360
    },
    {
      "epoch": 65.0,
      "learning_rate": 3.5286956521739133e-06,
      "loss": 2.7944,
      "step": 7475
    },
    {
      "epoch": 65.0,
      "eval_loss": 2.8042430877685547,
      "eval_runtime": 5.7552,
      "eval_samples_per_second": 6.95,
      "eval_steps_per_second": 0.869,
      "step": 7475
    },
    {
      "epoch": 66.0,
      "learning_rate": 3.4286956521739134e-06,
      "loss": 2.7986,
      "step": 7590
    },
    {
      "epoch": 66.0,
      "eval_loss": 2.8307268619537354,
      "eval_runtime": 5.741,
      "eval_samples_per_second": 6.967,
      "eval_steps_per_second": 0.871,
      "step": 7590
    },
    {
      "epoch": 67.0,
      "learning_rate": 3.328695652173913e-06,
      "loss": 2.7567,
      "step": 7705
    },
    {
      "epoch": 67.0,
      "eval_loss": 2.805975914001465,
      "eval_runtime": 5.7979,
      "eval_samples_per_second": 6.899,
      "eval_steps_per_second": 0.862,
      "step": 7705
    },
    {
      "epoch": 68.0,
      "learning_rate": 3.2286956521739132e-06,
      "loss": 2.7552,
      "step": 7820
    },
    {
      "epoch": 68.0,
      "eval_loss": 2.7994372844696045,
      "eval_runtime": 5.7117,
      "eval_samples_per_second": 7.003,
      "eval_steps_per_second": 0.875,
      "step": 7820
    },
    {
      "epoch": 69.0,
      "learning_rate": 3.1286956521739133e-06,
      "loss": 2.7933,
      "step": 7935
    },
    {
      "epoch": 69.0,
      "eval_loss": 2.849256992340088,
      "eval_runtime": 5.7067,
      "eval_samples_per_second": 7.009,
      "eval_steps_per_second": 0.876,
      "step": 7935
    },
    {
      "epoch": 70.0,
      "learning_rate": 3.028695652173913e-06,
      "loss": 2.7393,
      "step": 8050
    },
    {
      "epoch": 70.0,
      "eval_loss": 2.8409152030944824,
      "eval_runtime": 5.6797,
      "eval_samples_per_second": 7.043,
      "eval_steps_per_second": 0.88,
      "step": 8050
    },
    {
      "epoch": 71.0,
      "learning_rate": 2.9286956521739136e-06,
      "loss": 2.7357,
      "step": 8165
    },
    {
      "epoch": 71.0,
      "eval_loss": 2.8086206912994385,
      "eval_runtime": 5.6798,
      "eval_samples_per_second": 7.043,
      "eval_steps_per_second": 0.88,
      "step": 8165
    },
    {
      "epoch": 72.0,
      "learning_rate": 2.8286956521739132e-06,
      "loss": 2.7264,
      "step": 8280
    },
    {
      "epoch": 72.0,
      "eval_loss": 2.7772560119628906,
      "eval_runtime": 5.7909,
      "eval_samples_per_second": 6.907,
      "eval_steps_per_second": 0.863,
      "step": 8280
    },
    {
      "epoch": 73.0,
      "learning_rate": 2.7286956521739134e-06,
      "loss": 2.7614,
      "step": 8395
    },
    {
      "epoch": 73.0,
      "eval_loss": 2.8937366008758545,
      "eval_runtime": 5.7573,
      "eval_samples_per_second": 6.948,
      "eval_steps_per_second": 0.868,
      "step": 8395
    },
    {
      "epoch": 74.0,
      "learning_rate": 2.6295652173913044e-06,
      "loss": 2.7279,
      "step": 8510
    },
    {
      "epoch": 74.0,
      "eval_loss": 2.8887228965759277,
      "eval_runtime": 5.7093,
      "eval_samples_per_second": 7.006,
      "eval_steps_per_second": 0.876,
      "step": 8510
    },
    {
      "epoch": 75.0,
      "learning_rate": 2.5295652173913045e-06,
      "loss": 2.745,
      "step": 8625
    },
    {
      "epoch": 75.0,
      "eval_loss": 2.827376127243042,
      "eval_runtime": 5.6568,
      "eval_samples_per_second": 7.071,
      "eval_steps_per_second": 0.884,
      "step": 8625
    },
    {
      "epoch": 76.0,
      "learning_rate": 2.4295652173913046e-06,
      "loss": 2.7225,
      "step": 8740
    },
    {
      "epoch": 76.0,
      "eval_loss": 2.7970547676086426,
      "eval_runtime": 5.7165,
      "eval_samples_per_second": 6.997,
      "eval_steps_per_second": 0.875,
      "step": 8740
    },
    {
      "epoch": 77.0,
      "learning_rate": 2.3295652173913043e-06,
      "loss": 2.7094,
      "step": 8855
    },
    {
      "epoch": 77.0,
      "eval_loss": 2.868479013442993,
      "eval_runtime": 5.6813,
      "eval_samples_per_second": 7.041,
      "eval_steps_per_second": 0.88,
      "step": 8855
    },
    {
      "epoch": 78.0,
      "learning_rate": 2.229565217391305e-06,
      "loss": 2.7306,
      "step": 8970
    },
    {
      "epoch": 78.0,
      "eval_loss": 2.8482155799865723,
      "eval_runtime": 5.7403,
      "eval_samples_per_second": 6.968,
      "eval_steps_per_second": 0.871,
      "step": 8970
    },
    {
      "epoch": 79.0,
      "learning_rate": 2.1295652173913045e-06,
      "loss": 2.6844,
      "step": 9085
    },
    {
      "epoch": 79.0,
      "eval_loss": 2.7371761798858643,
      "eval_runtime": 5.7039,
      "eval_samples_per_second": 7.013,
      "eval_steps_per_second": 0.877,
      "step": 9085
    },
    {
      "epoch": 80.0,
      "learning_rate": 2.0295652173913046e-06,
      "loss": 2.6949,
      "step": 9200
    },
    {
      "epoch": 80.0,
      "eval_loss": 2.81486177444458,
      "eval_runtime": 5.7433,
      "eval_samples_per_second": 6.965,
      "eval_steps_per_second": 0.871,
      "step": 9200
    },
    {
      "epoch": 81.0,
      "learning_rate": 1.9295652173913047e-06,
      "loss": 2.7342,
      "step": 9315
    },
    {
      "epoch": 81.0,
      "eval_loss": 2.76469087600708,
      "eval_runtime": 5.6684,
      "eval_samples_per_second": 7.057,
      "eval_steps_per_second": 0.882,
      "step": 9315
    },
    {
      "epoch": 82.0,
      "learning_rate": 1.8295652173913044e-06,
      "loss": 2.6813,
      "step": 9430
    },
    {
      "epoch": 82.0,
      "eval_loss": 2.7665936946868896,
      "eval_runtime": 5.6874,
      "eval_samples_per_second": 7.033,
      "eval_steps_per_second": 0.879,
      "step": 9430
    },
    {
      "epoch": 83.0,
      "learning_rate": 1.7295652173913043e-06,
      "loss": 2.7161,
      "step": 9545
    },
    {
      "epoch": 83.0,
      "eval_loss": 2.843693971633911,
      "eval_runtime": 5.7308,
      "eval_samples_per_second": 6.98,
      "eval_steps_per_second": 0.872,
      "step": 9545
    },
    {
      "epoch": 84.0,
      "learning_rate": 1.6295652173913046e-06,
      "loss": 2.6953,
      "step": 9660
    },
    {
      "epoch": 84.0,
      "eval_loss": 2.7895007133483887,
      "eval_runtime": 5.5926,
      "eval_samples_per_second": 7.152,
      "eval_steps_per_second": 0.894,
      "step": 9660
    },
    {
      "epoch": 85.0,
      "learning_rate": 1.5295652173913045e-06,
      "loss": 2.6714,
      "step": 9775
    },
    {
      "epoch": 85.0,
      "eval_loss": 2.768319606781006,
      "eval_runtime": 5.5781,
      "eval_samples_per_second": 7.171,
      "eval_steps_per_second": 0.896,
      "step": 9775
    },
    {
      "epoch": 86.0,
      "learning_rate": 1.4295652173913044e-06,
      "loss": 2.6611,
      "step": 9890
    },
    {
      "epoch": 86.0,
      "eval_loss": 2.700421094894409,
      "eval_runtime": 5.7156,
      "eval_samples_per_second": 6.998,
      "eval_steps_per_second": 0.875,
      "step": 9890
    },
    {
      "epoch": 87.0,
      "learning_rate": 1.3295652173913045e-06,
      "loss": 2.6714,
      "step": 10005
    },
    {
      "epoch": 87.0,
      "eval_loss": 2.7182838916778564,
      "eval_runtime": 5.6318,
      "eval_samples_per_second": 7.103,
      "eval_steps_per_second": 0.888,
      "step": 10005
    },
    {
      "epoch": 88.0,
      "learning_rate": 1.2295652173913044e-06,
      "loss": 2.6655,
      "step": 10120
    },
    {
      "epoch": 88.0,
      "eval_loss": 2.7042617797851562,
      "eval_runtime": 5.8135,
      "eval_samples_per_second": 6.881,
      "eval_steps_per_second": 0.86,
      "step": 10120
    },
    {
      "epoch": 89.0,
      "learning_rate": 1.1295652173913045e-06,
      "loss": 2.6509,
      "step": 10235
    },
    {
      "epoch": 89.0,
      "eval_loss": 2.770498514175415,
      "eval_runtime": 5.7555,
      "eval_samples_per_second": 6.95,
      "eval_steps_per_second": 0.869,
      "step": 10235
    },
    {
      "epoch": 90.0,
      "learning_rate": 1.0295652173913044e-06,
      "loss": 2.6266,
      "step": 10350
    },
    {
      "epoch": 90.0,
      "eval_loss": 2.7151668071746826,
      "eval_runtime": 5.6476,
      "eval_samples_per_second": 7.083,
      "eval_steps_per_second": 0.885,
      "step": 10350
    },
    {
      "epoch": 91.0,
      "learning_rate": 9.295652173913044e-07,
      "loss": 2.6677,
      "step": 10465
    },
    {
      "epoch": 91.0,
      "eval_loss": 2.729531764984131,
      "eval_runtime": 5.6362,
      "eval_samples_per_second": 7.097,
      "eval_steps_per_second": 0.887,
      "step": 10465
    },
    {
      "epoch": 92.0,
      "learning_rate": 8.295652173913043e-07,
      "loss": 2.6438,
      "step": 10580
    },
    {
      "epoch": 92.0,
      "eval_loss": 2.70180344581604,
      "eval_runtime": 5.6488,
      "eval_samples_per_second": 7.081,
      "eval_steps_per_second": 0.885,
      "step": 10580
    },
    {
      "epoch": 93.0,
      "learning_rate": 7.295652173913044e-07,
      "loss": 2.6267,
      "step": 10695
    },
    {
      "epoch": 93.0,
      "eval_loss": 2.7062716484069824,
      "eval_runtime": 5.5879,
      "eval_samples_per_second": 7.158,
      "eval_steps_per_second": 0.895,
      "step": 10695
    },
    {
      "epoch": 94.0,
      "learning_rate": 6.295652173913045e-07,
      "loss": 2.6286,
      "step": 10810
    },
    {
      "epoch": 94.0,
      "eval_loss": 2.779799222946167,
      "eval_runtime": 5.6552,
      "eval_samples_per_second": 7.073,
      "eval_steps_per_second": 0.884,
      "step": 10810
    },
    {
      "epoch": 95.0,
      "learning_rate": 5.295652173913045e-07,
      "loss": 2.6043,
      "step": 10925
    },
    {
      "epoch": 95.0,
      "eval_loss": 2.7712299823760986,
      "eval_runtime": 5.6131,
      "eval_samples_per_second": 7.126,
      "eval_steps_per_second": 0.891,
      "step": 10925
    },
    {
      "epoch": 96.0,
      "learning_rate": 4.295652173913044e-07,
      "loss": 2.6188,
      "step": 11040
    },
    {
      "epoch": 96.0,
      "eval_loss": 2.7614357471466064,
      "eval_runtime": 5.567,
      "eval_samples_per_second": 7.185,
      "eval_steps_per_second": 0.898,
      "step": 11040
    },
    {
      "epoch": 97.0,
      "learning_rate": 3.2956521739130436e-07,
      "loss": 2.6028,
      "step": 11155
    },
    {
      "epoch": 97.0,
      "eval_loss": 2.740495204925537,
      "eval_runtime": 5.621,
      "eval_samples_per_second": 7.116,
      "eval_steps_per_second": 0.89,
      "step": 11155
    },
    {
      "epoch": 98.0,
      "learning_rate": 2.2956521739130436e-07,
      "loss": 2.621,
      "step": 11270
    },
    {
      "epoch": 98.0,
      "eval_loss": 2.741487503051758,
      "eval_runtime": 5.6116,
      "eval_samples_per_second": 7.128,
      "eval_steps_per_second": 0.891,
      "step": 11270
    },
    {
      "epoch": 99.0,
      "learning_rate": 1.2956521739130434e-07,
      "loss": 2.61,
      "step": 11385
    },
    {
      "epoch": 99.0,
      "eval_loss": 2.741542100906372,
      "eval_runtime": 5.7228,
      "eval_samples_per_second": 6.99,
      "eval_steps_per_second": 0.874,
      "step": 11385
    },
    {
      "epoch": 100.0,
      "learning_rate": 2.9565217391304353e-08,
      "loss": 2.6164,
      "step": 11500
    },
    {
      "epoch": 100.0,
      "eval_loss": 2.7389094829559326,
      "eval_runtime": 5.5344,
      "eval_samples_per_second": 7.228,
      "eval_steps_per_second": 0.903,
      "step": 11500
    },
    {
      "epoch": 100.0,
      "step": 11500,
      "total_flos": 2.295532415232e+19,
      "train_loss": 1.6043826771611753,
      "train_runtime": 5669.8878,
      "train_samples_per_second": 8.113,
      "train_steps_per_second": 2.028
    }
  ],
  "logging_steps": 500,
  "max_steps": 11500,
  "num_train_epochs": 100,
  "save_steps": 500,
  "total_flos": 2.295532415232e+19,
  "trial_name": null,
  "trial_params": null
}