trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 7.0,
  "eval_steps": 500,
  "global_step": 2464,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_gen_len": 109.027,
      "eval_loss": 1.2415885925292969,
      "eval_rouge1": 0.5269433666229263,
      "eval_rouge2": 0.28037183395108156,
      "eval_rougeL": 0.35995139966897394,
      "eval_rougeLsum": 0.37649885084214285,
      "eval_runtime": 723.8312,
      "eval_samples_per_second": 1.942,
      "eval_steps_per_second": 0.122,
      "step": 352
    },
    {
      "epoch": 1.4204545454545454,
      "grad_norm": 3.200246810913086,
      "learning_rate": 4.759018352770229e-06,
      "loss": 1.3036,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_gen_len": 112.4431,
      "eval_loss": 1.177475094795227,
      "eval_rouge1": 0.5430391605521028,
      "eval_rouge2": 0.2972281166428693,
      "eval_rougeL": 0.37256436707598495,
      "eval_rougeLsum": 0.3884270291699813,
      "eval_runtime": 682.3343,
      "eval_samples_per_second": 2.061,
      "eval_steps_per_second": 0.129,
      "step": 704
    },
    {
      "epoch": 2.840909090909091,
      "grad_norm": 3.206270217895508,
      "learning_rate": 4.075611866937373e-06,
      "loss": 1.1054,
      "step": 1000
    },
    {
      "epoch": 3.0,
      "eval_gen_len": 108.6607,
      "eval_loss": 1.1586213111877441,
      "eval_rouge1": 0.5453459493029312,
      "eval_rouge2": 0.30012808902733723,
      "eval_rougeL": 0.3766083565770627,
      "eval_rougeLsum": 0.39252447019524606,
      "eval_runtime": 840.2954,
      "eval_samples_per_second": 1.673,
      "eval_steps_per_second": 0.105,
      "step": 1056
    },
    {
      "epoch": 4.0,
      "eval_gen_len": 106.1166,
      "eval_loss": 1.1364487409591675,
      "eval_rouge1": 0.5476054862493738,
      "eval_rouge2": 0.3011862619458042,
      "eval_rougeL": 0.37811611035117715,
      "eval_rougeLsum": 0.3946684605387395,
      "eval_runtime": 576.235,
      "eval_samples_per_second": 2.44,
      "eval_steps_per_second": 0.153,
      "step": 1408
    },
    {
      "epoch": 4.261363636363637,
      "grad_norm": 3.138392925262451,
      "learning_rate": 3.0836134096397642e-06,
      "loss": 1.0104,
      "step": 1500
    },
    {
      "epoch": 5.0,
      "eval_gen_len": 105.2859,
      "eval_loss": 1.1392455101013184,
      "eval_rouge1": 0.5465154278801767,
      "eval_rouge2": 0.3007552050130359,
      "eval_rougeL": 0.3773198844643692,
      "eval_rougeLsum": 0.3932806868540929,
      "eval_runtime": 1033.7454,
      "eval_samples_per_second": 1.36,
      "eval_steps_per_second": 0.085,
      "step": 1760
    },
    {
      "epoch": 5.681818181818182,
      "grad_norm": 3.235567092895508,
      "learning_rate": 1.977311159362942e-06,
      "loss": 0.9495,
      "step": 2000
    },
    {
      "epoch": 6.0,
      "eval_gen_len": 104.6494,
      "eval_loss": 1.125599980354309,
      "eval_rouge1": 0.5492129139731206,
      "eval_rouge2": 0.30319883423234095,
      "eval_rougeL": 0.3789239095282904,
      "eval_rougeLsum": 0.39465416548270177,
      "eval_runtime": 562.8701,
      "eval_samples_per_second": 2.498,
      "eval_steps_per_second": 0.156,
      "step": 2112
    },
    {
      "epoch": 7.0,
      "eval_gen_len": 103.5974,
      "eval_loss": 1.1215816736221313,
      "eval_rouge1": 0.5489838414026048,
      "eval_rouge2": 0.30370662819639743,
      "eval_rougeL": 0.3803871339480277,
      "eval_rougeLsum": 0.3957810407664568,
      "eval_runtime": 623.0852,
      "eval_samples_per_second": 2.257,
      "eval_steps_per_second": 0.141,
      "step": 2464
    }
  ],
  "logging_steps": 500,
  "max_steps": 3520,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 9.445404124486042e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}