trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 10.0,
  "global_step": 3980,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_bleu": 0.0463,
      "eval_gen_len": 19.0,
      "eval_loss": 0.9985501170158386,
      "eval_runtime": 39.5467,
      "eval_samples_per_second": 20.078,
      "eval_steps_per_second": 1.264,
      "step": 398
    },
    {
      "epoch": 1.26,
      "learning_rate": 4.3718592964824124e-05,
      "loss": 1.43,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_bleu": 0.5462,
      "eval_gen_len": 19.0,
      "eval_loss": 0.8809053301811218,
      "eval_runtime": 38.6627,
      "eval_samples_per_second": 20.537,
      "eval_steps_per_second": 1.293,
      "step": 796
    },
    {
      "epoch": 2.51,
      "learning_rate": 3.7437185929648245e-05,
      "loss": 0.9934,
      "step": 1000
    },
    {
      "epoch": 3.0,
      "eval_bleu": 0.4119,
      "eval_gen_len": 19.0,
      "eval_loss": 0.8291094899177551,
      "eval_runtime": 38.6686,
      "eval_samples_per_second": 20.533,
      "eval_steps_per_second": 1.293,
      "step": 1194
    },
    {
      "epoch": 3.77,
      "learning_rate": 3.1155778894472366e-05,
      "loss": 0.8998,
      "step": 1500
    },
    {
      "epoch": 4.0,
      "eval_bleu": 0.4437,
      "eval_gen_len": 19.0,
      "eval_loss": 0.7965371608734131,
      "eval_runtime": 38.8377,
      "eval_samples_per_second": 20.444,
      "eval_steps_per_second": 1.287,
      "step": 1592
    },
    {
      "epoch": 5.0,
      "eval_bleu": 0.3303,
      "eval_gen_len": 19.0,
      "eval_loss": 0.7747035026550293,
      "eval_runtime": 38.8115,
      "eval_samples_per_second": 20.458,
      "eval_steps_per_second": 1.288,
      "step": 1990
    },
    {
      "epoch": 5.03,
      "learning_rate": 2.4874371859296484e-05,
      "loss": 0.851,
      "step": 2000
    },
    {
      "epoch": 6.0,
      "eval_bleu": 0.3764,
      "eval_gen_len": 19.0,
      "eval_loss": 0.7562447190284729,
      "eval_runtime": 39.1663,
      "eval_samples_per_second": 20.273,
      "eval_steps_per_second": 1.277,
      "step": 2388
    },
    {
      "epoch": 6.28,
      "learning_rate": 1.8592964824120602e-05,
      "loss": 0.8139,
      "step": 2500
    },
    {
      "epoch": 7.0,
      "eval_bleu": 0.3244,
      "eval_gen_len": 19.0,
      "eval_loss": 0.7414721250534058,
      "eval_runtime": 38.5786,
      "eval_samples_per_second": 20.581,
      "eval_steps_per_second": 1.296,
      "step": 2786
    },
    {
      "epoch": 7.54,
      "learning_rate": 1.2311557788944725e-05,
      "loss": 0.7865,
      "step": 3000
    },
    {
      "epoch": 8.0,
      "eval_bleu": 0.4334,
      "eval_gen_len": 19.0,
      "eval_loss": 0.7340883016586304,
      "eval_runtime": 38.6301,
      "eval_samples_per_second": 20.554,
      "eval_steps_per_second": 1.294,
      "step": 3184
    },
    {
      "epoch": 8.79,
      "learning_rate": 6.030150753768844e-06,
      "loss": 0.7767,
      "step": 3500
    },
    {
      "epoch": 9.0,
      "eval_bleu": 0.3918,
      "eval_gen_len": 19.0,
      "eval_loss": 0.7294268608093262,
      "eval_runtime": 39.2323,
      "eval_samples_per_second": 20.238,
      "eval_steps_per_second": 1.274,
      "step": 3582
    },
    {
      "epoch": 10.0,
      "eval_bleu": 0.3979,
      "eval_gen_len": 19.0,
      "eval_loss": 0.7257152199745178,
      "eval_runtime": 38.4568,
      "eval_samples_per_second": 20.647,
      "eval_steps_per_second": 1.3,
      "step": 3980
    },
    {
      "epoch": 10.0,
      "step": 3980,
      "total_flos": 2.259548135129088e+16,
      "train_loss": 0.9154524760030622,
      "train_runtime": 2099.4873,
      "train_samples_per_second": 30.265,
      "train_steps_per_second": 1.896
    }
  ],
  "max_steps": 3980,
  "num_train_epochs": 10,
  "total_flos": 2.259548135129088e+16,
  "trial_name": null,
  "trial_params": null
}