last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 8.0,
  "eval_steps": 500,
  "global_step": 2624,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_gen_len": 10.017182130584192,
      "eval_loss": 0.6921027898788452,
      "eval_rouge1": 34.9112,
      "eval_rouge2": 26.7503,
      "eval_rougeL": 31.4124,
      "eval_rougeLsum": 31.7295,
      "eval_runtime": 139.485,
      "eval_samples_per_second": 2.086,
      "eval_steps_per_second": 0.265,
      "step": 328
    },
    {
      "epoch": 1.52,
      "learning_rate": 0.0003125,
      "loss": 6.8746,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_gen_len": 10.845360824742269,
      "eval_loss": 0.6025161147117615,
      "eval_rouge1": 33.9134,
      "eval_rouge2": 25.3236,
      "eval_rougeL": 30.1968,
      "eval_rougeLsum": 30.472,
      "eval_runtime": 151.0112,
      "eval_samples_per_second": 1.927,
      "eval_steps_per_second": 0.245,
      "step": 656
    },
    {
      "epoch": 3.0,
      "eval_gen_len": 11.872852233676976,
      "eval_loss": 0.5686607956886292,
      "eval_rouge1": 31.6178,
      "eval_rouge2": 22.9463,
      "eval_rougeL": 27.8758,
      "eval_rougeLsum": 28.3572,
      "eval_runtime": 165.9096,
      "eval_samples_per_second": 1.754,
      "eval_steps_per_second": 0.223,
      "step": 984
    },
    {
      "epoch": 3.05,
      "learning_rate": 0.0004853133981334999,
      "loss": 0.6462,
      "step": 1000
    },
    {
      "epoch": 4.0,
      "eval_gen_len": 12.130584192439862,
      "eval_loss": 0.5354723334312439,
      "eval_rouge1": 30.8157,
      "eval_rouge2": 22.1783,
      "eval_rougeL": 27.1641,
      "eval_rougeLsum": 27.569,
      "eval_runtime": 160.3562,
      "eval_samples_per_second": 1.815,
      "eval_steps_per_second": 0.231,
      "step": 1312
    },
    {
      "epoch": 4.57,
      "learning_rate": 0.00033927030930209996,
      "loss": 0.5618,
      "step": 1500
    },
    {
      "epoch": 5.0,
      "eval_gen_len": 12.50171821305842,
      "eval_loss": 0.5159749388694763,
      "eval_rouge1": 29.9183,
      "eval_rouge2": 21.0842,
      "eval_rougeL": 26.1671,
      "eval_rougeLsum": 26.5965,
      "eval_runtime": 158.4069,
      "eval_samples_per_second": 1.837,
      "eval_steps_per_second": 0.234,
      "step": 1640
    },
    {
      "epoch": 6.0,
      "eval_gen_len": 12.508591065292096,
      "eval_loss": 0.5024679899215698,
      "eval_rouge1": 29.7823,
      "eval_rouge2": 21.1443,
      "eval_rougeL": 26.0286,
      "eval_rougeLsum": 26.5215,
      "eval_runtime": 154.9067,
      "eval_samples_per_second": 1.879,
      "eval_steps_per_second": 0.239,
      "step": 1968
    },
    {
      "epoch": 6.1,
      "learning_rate": 0.0001310131517407316,
      "loss": 0.498,
      "step": 2000
    },
    {
      "epoch": 7.0,
      "eval_gen_len": 12.896907216494846,
      "eval_loss": 0.4978101849555969,
      "eval_rouge1": 29.1043,
      "eval_rouge2": 20.2391,
      "eval_rougeL": 25.3347,
      "eval_rougeLsum": 25.804,
      "eval_runtime": 152.5447,
      "eval_samples_per_second": 1.908,
      "eval_steps_per_second": 0.243,
      "step": 2296
    },
    {
      "epoch": 7.62,
      "learning_rate": 5.680048251060194e-06,
      "loss": 0.4551,
      "step": 2500
    },
    {
      "epoch": 8.0,
      "eval_gen_len": 12.738831615120274,
      "eval_loss": 0.497775673866272,
      "eval_rouge1": 29.5362,
      "eval_rouge2": 20.6621,
      "eval_rougeL": 25.7689,
      "eval_rougeLsum": 26.2351,
      "eval_runtime": 159.8028,
      "eval_samples_per_second": 1.821,
      "eval_steps_per_second": 0.232,
      "step": 2624
    }
  ],
  "logging_steps": 500,
  "max_steps": 2624,
  "num_train_epochs": 8,
  "save_steps": 500,
  "total_flos": 1.966513744458547e+16,
  "trial_name": null,
  "trial_params": null
}