{
  "best_metric": 2.025848865509033,
  "best_model_checkpoint": "./checkpoints/barthez-deft-sciences_de_l_information/checkpoint-424",
  "epoch": 20.0,
  "global_step": 2120,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.94,
      "learning_rate": 2.869811320754717e-05,
      "loss": 3.3405,
      "step": 100
    },
    {
      "epoch": 1.0,
      "eval_gen_len": 14.9714,
      "eval_loss": 2.368244171142578,
      "eval_rouge1": 31.3511,
      "eval_rouge2": 12.1973,
      "eval_rougeL": 25.6977,
      "eval_rougeLsum": 25.6851,
      "eval_runtime": 3.4096,
      "eval_samples_per_second": 20.53,
      "eval_steps_per_second": 5.279,
      "step": 106
    },
    {
      "epoch": 1.89,
      "learning_rate": 2.728301886792453e-05,
      "loss": 2.4219,
      "step": 200
    },
    {
      "epoch": 2.0,
      "eval_gen_len": 14.0429,
      "eval_loss": 2.1890599727630615,
      "eval_rouge1": 30.1154,
      "eval_rouge2": 13.3459,
      "eval_rougeL": 25.4854,
      "eval_rougeLsum": 25.5403,
      "eval_runtime": 3.216,
      "eval_samples_per_second": 21.766,
      "eval_steps_per_second": 5.597,
      "step": 212
    },
    {
      "epoch": 2.83,
      "learning_rate": 2.586792452830189e-05,
      "loss": 2.0789,
      "step": 300
    },
    {
      "epoch": 3.0,
      "eval_gen_len": 15.2,
      "eval_loss": 2.0993804931640625,
      "eval_rouge1": 32.153,
      "eval_rouge2": 15.3865,
      "eval_rougeL": 26.1859,
      "eval_rougeLsum": 26.1672,
      "eval_runtime": 3.3607,
      "eval_samples_per_second": 20.829,
      "eval_steps_per_second": 5.356,
      "step": 318
    },
    {
      "epoch": 3.77,
      "learning_rate": 2.4452830188679248e-05,
      "loss": 1.869,
      "step": 400
    },
    {
      "epoch": 4.0,
      "eval_gen_len": 16.9857,
      "eval_loss": 2.025848865509033,
      "eval_rouge1": 34.5797,
      "eval_rouge2": 16.4194,
      "eval_rougeL": 27.6909,
      "eval_rougeLsum": 27.7201,
      "eval_runtime": 3.3358,
      "eval_samples_per_second": 20.985,
      "eval_steps_per_second": 5.396,
      "step": 424
    },
    {
      "epoch": 4.72,
      "learning_rate": 2.3037735849056604e-05,
      "loss": 1.6569,
      "step": 500
    },
    {
      "epoch": 5.0,
      "eval_gen_len": 15.2429,
      "eval_loss": 2.0417497158050537,
      "eval_rouge1": 34.3854,
      "eval_rouge2": 16.5237,
      "eval_rougeL": 28.7036,
      "eval_rougeLsum": 28.8258,
      "eval_runtime": 3.3091,
      "eval_samples_per_second": 21.154,
      "eval_steps_per_second": 5.44,
      "step": 530
    },
    {
      "epoch": 5.66,
      "learning_rate": 2.162264150943396e-05,
      "loss": 1.5414,
      "step": 600
    },
    {
      "epoch": 6.0,
      "eval_gen_len": 16.0143,
      "eval_loss": 2.050337791442871,
      "eval_rouge1": 33.1768,
      "eval_rouge2": 15.4851,
      "eval_rougeL": 27.2818,
      "eval_rougeLsum": 27.2884,
      "eval_runtime": 3.4107,
      "eval_samples_per_second": 20.524,
      "eval_steps_per_second": 5.278,
      "step": 636
    },
    {
      "epoch": 6.6,
      "learning_rate": 2.020754716981132e-05,
      "loss": 1.4461,
      "step": 700
    },
    {
      "epoch": 7.0,
      "eval_gen_len": 16.6857,
      "eval_loss": 2.029313802719116,
      "eval_rouge1": 35.4273,
      "eval_rouge2": 16.118,
      "eval_rougeL": 27.3622,
      "eval_rougeLsum": 27.393,
      "eval_runtime": 3.3635,
      "eval_samples_per_second": 20.812,
      "eval_steps_per_second": 5.352,
      "step": 742
    },
    {
      "epoch": 7.55,
      "learning_rate": 1.879245283018868e-05,
      "loss": 1.3435,
      "step": 800
    },
    {
      "epoch": 8.0,
      "eval_gen_len": 17.2,
      "eval_loss": 2.033561944961548,
      "eval_rouge1": 35.3471,
      "eval_rouge2": 15.9695,
      "eval_rougeL": 27.668,
      "eval_rougeLsum": 27.6749,
      "eval_runtime": 3.4311,
      "eval_samples_per_second": 20.402,
      "eval_steps_per_second": 5.246,
      "step": 848
    },
    {
      "epoch": 8.49,
      "learning_rate": 1.737735849056604e-05,
      "loss": 1.2624,
      "step": 900
    },
    {
      "epoch": 9.0,
      "eval_gen_len": 17.1857,
      "eval_loss": 2.077875852584839,
      "eval_rouge1": 35.9201,
      "eval_rouge2": 17.2547,
      "eval_rougeL": 27.409,
      "eval_rougeLsum": 27.3293,
      "eval_runtime": 3.4,
      "eval_samples_per_second": 20.588,
      "eval_steps_per_second": 5.294,
      "step": 954
    },
    {
      "epoch": 9.43,
      "learning_rate": 1.5962264150943395e-05,
      "loss": 1.1807,
      "step": 1000
    },
    {
      "epoch": 10.0,
      "eval_gen_len": 17.1286,
      "eval_loss": 2.130077600479126,
      "eval_rouge1": 35.7061,
      "eval_rouge2": 15.9138,
      "eval_rougeL": 27.3968,
      "eval_rougeLsum": 27.4716,
      "eval_runtime": 3.3605,
      "eval_samples_per_second": 20.83,
      "eval_steps_per_second": 5.356,
      "step": 1060
    },
    {
      "epoch": 10.38,
      "learning_rate": 1.4547169811320755e-05,
      "loss": 1.0972,
      "step": 1100
    },
    {
      "epoch": 11.0,
      "eval_gen_len": 17.1429,
      "eval_loss": 2.1725852489471436,
      "eval_rouge1": 34.3194,
      "eval_rouge2": 16.1313,
      "eval_rougeL": 27.0367,
      "eval_rougeLsum": 27.0737,
      "eval_runtime": 3.3905,
      "eval_samples_per_second": 20.646,
      "eval_steps_per_second": 5.309,
      "step": 1166
    },
    {
      "epoch": 11.32,
      "learning_rate": 1.3132075471698114e-05,
      "loss": 1.0224,
      "step": 1200
    },
    {
      "epoch": 12.0,
      "eval_gen_len": 16.6571,
      "eval_loss": 2.1703763008117676,
      "eval_rouge1": 34.9278,
      "eval_rouge2": 16.7958,
      "eval_rougeL": 27.8754,
      "eval_rougeLsum": 27.932,
      "eval_runtime": 3.3798,
      "eval_samples_per_second": 20.711,
      "eval_steps_per_second": 5.326,
      "step": 1272
    },
    {
      "epoch": 12.26,
      "learning_rate": 1.1716981132075474e-05,
      "loss": 1.0181,
      "step": 1300
    },
    {
      "epoch": 13.0,
      "eval_gen_len": 16.7571,
      "eval_loss": 2.245802879333496,
      "eval_rouge1": 34.472,
      "eval_rouge2": 15.9111,
      "eval_rougeL": 28.2938,
      "eval_rougeLsum": 28.2946,
      "eval_runtime": 3.367,
      "eval_samples_per_second": 20.79,
      "eval_steps_per_second": 5.346,
      "step": 1378
    },
    {
      "epoch": 13.21,
      "learning_rate": 1.030188679245283e-05,
      "loss": 0.9769,
      "step": 1400
    },
    {
      "epoch": 14.0,
      "eval_gen_len": 16.5429,
      "eval_loss": 2.3404934406280518,
      "eval_rouge1": 35.1592,
      "eval_rouge2": 16.3135,
      "eval_rougeL": 29.0956,
      "eval_rougeLsum": 29.0858,
      "eval_runtime": 3.3904,
      "eval_samples_per_second": 20.647,
      "eval_steps_per_second": 5.309,
      "step": 1484
    },
    {
      "epoch": 14.15,
      "learning_rate": 8.900943396226416e-06,
      "loss": 0.8866,
      "step": 1500
    },
    {
      "epoch": 15.0,
      "eval_gen_len": 16.2429,
      "eval_loss": 2.3303470611572266,
      "eval_rouge1": 34.8732,
      "eval_rouge2": 15.6709,
      "eval_rougeL": 27.5858,
      "eval_rougeLsum": 27.6169,
      "eval_runtime": 3.4313,
      "eval_samples_per_second": 20.401,
      "eval_steps_per_second": 5.246,
      "step": 1590
    },
    {
      "epoch": 15.09,
      "learning_rate": 7.485849056603774e-06,
      "loss": 0.8888,
      "step": 1600
    },
    {
      "epoch": 16.0,
      "eval_gen_len": 17.5143,
      "eval_loss": 2.297647476196289,
      "eval_rouge1": 35.3034,
      "eval_rouge2": 16.8011,
      "eval_rougeL": 27.7988,
      "eval_rougeLsum": 27.7569,
      "eval_runtime": 3.3934,
      "eval_samples_per_second": 20.628,
      "eval_steps_per_second": 5.304,
      "step": 1696
    },
    {
      "epoch": 16.04,
      "learning_rate": 6.070754716981133e-06,
      "loss": 0.8194,
      "step": 1700
    },
    {
      "epoch": 16.98,
      "learning_rate": 4.6556603773584905e-06,
      "loss": 0.8358,
      "step": 1800
    },
    {
      "epoch": 17.0,
      "eval_gen_len": 16.8143,
      "eval_loss": 2.334933042526245,
      "eval_rouge1": 35.505,
      "eval_rouge2": 16.8851,
      "eval_rougeL": 28.3651,
      "eval_rougeLsum": 28.413,
      "eval_runtime": 3.4202,
      "eval_samples_per_second": 20.467,
      "eval_steps_per_second": 5.263,
      "step": 1802
    },
    {
      "epoch": 17.92,
      "learning_rate": 3.240566037735849e-06,
      "loss": 0.8026,
      "step": 1900
    },
    {
      "epoch": 18.0,
      "eval_gen_len": 16.6143,
      "eval_loss": 2.373809814453125,
      "eval_rouge1": 35.2328,
      "eval_rouge2": 17.0358,
      "eval_rougeL": 28.544,
      "eval_rougeLsum": 28.6211,
      "eval_runtime": 3.4477,
      "eval_samples_per_second": 20.303,
      "eval_steps_per_second": 5.221,
      "step": 1908
    },
    {
      "epoch": 18.87,
      "learning_rate": 1.8254716981132076e-06,
      "loss": 0.7487,
      "step": 2000
    },
    {
      "epoch": 19.0,
      "eval_gen_len": 16.7286,
      "eval_loss": 2.4102871417999268,
      "eval_rouge1": 34.0793,
      "eval_rouge2": 15.4468,
      "eval_rougeL": 27.8057,
      "eval_rougeLsum": 27.8586,
      "eval_runtime": 3.4317,
      "eval_samples_per_second": 20.398,
      "eval_steps_per_second": 5.245,
      "step": 2014
    },
    {
      "epoch": 19.81,
      "learning_rate": 4.1037735849056606e-07,
      "loss": 0.7722,
      "step": 2100
    },
    {
      "epoch": 20.0,
      "eval_gen_len": 16.9286,
      "eval_loss": 2.3990561962127686,
      "eval_rouge1": 34.8116,
      "eval_rouge2": 15.8706,
      "eval_rougeL": 27.9173,
      "eval_rougeLsum": 27.983,
      "eval_runtime": 3.3605,
      "eval_samples_per_second": 20.83,
      "eval_steps_per_second": 5.356,
      "step": 2120
    },
    {
      "epoch": 20.0,
      "step": 2120,
      "total_flos": 1197078338174976.0,
      "train_loss": 1.328409050995449,
      "train_runtime": 326.8588,
      "train_samples_per_second": 25.944,
      "train_steps_per_second": 6.486
    }
  ],
  "max_steps": 2120,
  "num_train_epochs": 20,
  "total_flos": 1197078338174976.0,
  "trial_name": null,
  "trial_params": null
}