trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 10.0,
  "global_step": 3860,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_bleu": 0.0256,
      "eval_gen_len": 19.0,
      "eval_loss": 0.9372363090515137,
      "eval_runtime": 38.1578,
      "eval_samples_per_second": 20.206,
      "eval_steps_per_second": 1.284,
      "step": 386
    },
    {
      "epoch": 1.3,
      "learning_rate": 4.352331606217617e-05,
      "loss": 1.4765,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_bleu": 0.0303,
      "eval_gen_len": 19.0,
      "eval_loss": 0.8219472169876099,
      "eval_runtime": 38.1767,
      "eval_samples_per_second": 20.196,
      "eval_steps_per_second": 1.284,
      "step": 772
    },
    {
      "epoch": 2.59,
      "learning_rate": 3.704663212435233e-05,
      "loss": 0.9443,
      "step": 1000
    },
    {
      "epoch": 3.0,
      "eval_bleu": 0.022,
      "eval_gen_len": 19.0,
      "eval_loss": 0.7708175778388977,
      "eval_runtime": 37.8173,
      "eval_samples_per_second": 20.388,
      "eval_steps_per_second": 1.296,
      "step": 1158
    },
    {
      "epoch": 3.89,
      "learning_rate": 3.05699481865285e-05,
      "loss": 0.8436,
      "step": 1500
    },
    {
      "epoch": 4.0,
      "eval_bleu": 0.0562,
      "eval_gen_len": 19.0,
      "eval_loss": 0.7287958860397339,
      "eval_runtime": 38.1538,
      "eval_samples_per_second": 20.208,
      "eval_steps_per_second": 1.284,
      "step": 1544
    },
    {
      "epoch": 5.0,
      "eval_bleu": 0.0469,
      "eval_gen_len": 19.0,
      "eval_loss": 0.7091419696807861,
      "eval_runtime": 37.9015,
      "eval_samples_per_second": 20.342,
      "eval_steps_per_second": 1.293,
      "step": 1930
    },
    {
      "epoch": 5.18,
      "learning_rate": 2.4093264248704665e-05,
      "loss": 0.7916,
      "step": 2000
    },
    {
      "epoch": 6.0,
      "eval_bleu": 0.0464,
      "eval_gen_len": 19.0,
      "eval_loss": 0.6926471590995789,
      "eval_runtime": 38.039,
      "eval_samples_per_second": 20.269,
      "eval_steps_per_second": 1.288,
      "step": 2316
    },
    {
      "epoch": 6.48,
      "learning_rate": 1.761658031088083e-05,
      "loss": 0.7568,
      "step": 2500
    },
    {
      "epoch": 7.0,
      "eval_bleu": 0.0536,
      "eval_gen_len": 19.0,
      "eval_loss": 0.6799036860466003,
      "eval_runtime": 38.099,
      "eval_samples_per_second": 20.237,
      "eval_steps_per_second": 1.286,
      "step": 2702
    },
    {
      "epoch": 7.77,
      "learning_rate": 1.1139896373056995e-05,
      "loss": 0.7403,
      "step": 3000
    },
    {
      "epoch": 8.0,
      "eval_bleu": 0.0547,
      "eval_gen_len": 19.0,
      "eval_loss": 0.6737177968025208,
      "eval_runtime": 37.9859,
      "eval_samples_per_second": 20.297,
      "eval_steps_per_second": 1.29,
      "step": 3088
    },
    {
      "epoch": 9.0,
      "eval_bleu": 0.0532,
      "eval_gen_len": 19.0,
      "eval_loss": 0.6666902303695679,
      "eval_runtime": 38.1676,
      "eval_samples_per_second": 20.2,
      "eval_steps_per_second": 1.284,
      "step": 3474
    },
    {
      "epoch": 9.07,
      "learning_rate": 4.663212435233161e-06,
      "loss": 0.7215,
      "step": 3500
    },
    {
      "epoch": 10.0,
      "eval_bleu": 0.056,
      "eval_gen_len": 19.0,
      "eval_loss": 0.6674807667732239,
      "eval_runtime": 38.091,
      "eval_samples_per_second": 20.241,
      "eval_steps_per_second": 1.286,
      "step": 3860
    },
    {
      "epoch": 10.0,
      "step": 3860,
      "total_flos": 2.192693390180352e+16,
      "train_loss": 0.8798027631532342,
      "train_runtime": 2043.5708,
      "train_samples_per_second": 30.173,
      "train_steps_per_second": 1.889
    }
  ],
  "max_steps": 3860,
  "num_train_epochs": 10,
  "total_flos": 2.192693390180352e+16,
  "trial_name": null,
  "trial_params": null
}