{
  "best_metric": 7.9168,
  "best_model_checkpoint": "/content/tst-translation/checkpoint-1600",
  "epoch": 20.0,
  "eval_steps": 200,
  "global_step": 2540,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.574803149606299,
      "grad_norm": 2.453336238861084,
      "learning_rate": 0.00046062992125984255,
      "loss": 4.5126,
      "step": 200
    },
    {
      "epoch": 1.574803149606299,
      "eval_bleu": 1.0891,
      "eval_gen_len": 120.8315,
      "eval_loss": 2.747382402420044,
      "eval_runtime": 505.8332,
      "eval_samples_per_second": 1.971,
      "eval_steps_per_second": 0.125,
      "step": 200
    },
    {
      "epoch": 3.1496062992125986,
      "grad_norm": 2.0367627143859863,
      "learning_rate": 0.00042125984251968504,
      "loss": 2.4414,
      "step": 400
    },
    {
      "epoch": 3.1496062992125986,
      "eval_bleu": 5.0172,
      "eval_gen_len": 54.7623,
      "eval_loss": 2.5120186805725098,
      "eval_runtime": 340.8894,
      "eval_samples_per_second": 2.925,
      "eval_steps_per_second": 0.185,
      "step": 400
    },
    {
      "epoch": 4.724409448818898,
      "grad_norm": 2.0367233753204346,
      "learning_rate": 0.00038188976377952753,
      "loss": 1.724,
      "step": 600
    },
    {
      "epoch": 4.724409448818898,
      "eval_bleu": 5.2115,
      "eval_gen_len": 61.985,
      "eval_loss": 2.4089620113372803,
      "eval_runtime": 368.5588,
      "eval_samples_per_second": 2.705,
      "eval_steps_per_second": 0.171,
      "step": 600
    },
    {
      "epoch": 6.299212598425197,
      "grad_norm": 1.8066316843032837,
      "learning_rate": 0.00034251968503937007,
      "loss": 1.2755,
      "step": 800
    },
    {
      "epoch": 6.299212598425197,
      "eval_bleu": 7.0203,
      "eval_gen_len": 46.1414,
      "eval_loss": 2.551859140396118,
      "eval_runtime": 215.1683,
      "eval_samples_per_second": 4.634,
      "eval_steps_per_second": 0.293,
      "step": 800
    },
    {
      "epoch": 7.874015748031496,
      "grad_norm": 1.8381917476654053,
      "learning_rate": 0.0003031496062992126,
      "loss": 0.97,
      "step": 1000
    },
    {
      "epoch": 7.874015748031496,
      "eval_bleu": 7.005,
      "eval_gen_len": 56.5266,
      "eval_loss": 2.5974559783935547,
      "eval_runtime": 308.9884,
      "eval_samples_per_second": 3.227,
      "eval_steps_per_second": 0.204,
      "step": 1000
    },
    {
      "epoch": 9.448818897637794,
      "grad_norm": 1.8475762605667114,
      "learning_rate": 0.0002637795275590551,
      "loss": 0.7251,
      "step": 1200
    },
    {
      "epoch": 9.448818897637794,
      "eval_bleu": 7.6235,
      "eval_gen_len": 52.6841,
      "eval_loss": 2.791808605194092,
      "eval_runtime": 280.2999,
      "eval_samples_per_second": 3.557,
      "eval_steps_per_second": 0.225,
      "step": 1200
    },
    {
      "epoch": 11.023622047244094,
      "grad_norm": 1.7481825351715088,
      "learning_rate": 0.00022440944881889764,
      "loss": 0.584,
      "step": 1400
    },
    {
      "epoch": 11.023622047244094,
      "eval_bleu": 7.3273,
      "eval_gen_len": 49.9659,
      "eval_loss": 2.8952395915985107,
      "eval_runtime": 205.6302,
      "eval_samples_per_second": 4.849,
      "eval_steps_per_second": 0.306,
      "step": 1400
    },
    {
      "epoch": 12.598425196850394,
      "grad_norm": 1.5020047426223755,
      "learning_rate": 0.00018503937007874016,
      "loss": 0.4358,
      "step": 1600
    },
    {
      "epoch": 12.598425196850394,
      "eval_bleu": 7.9168,
      "eval_gen_len": 51.4945,
      "eval_loss": 3.120616912841797,
      "eval_runtime": 243.389,
      "eval_samples_per_second": 4.096,
      "eval_steps_per_second": 0.259,
      "step": 1600
    },
    {
      "epoch": 14.173228346456693,
      "grad_norm": 1.6936888694763184,
      "learning_rate": 0.00014566929133858267,
      "loss": 0.3619,
      "step": 1800
    },
    {
      "epoch": 14.173228346456693,
      "eval_bleu": 7.9096,
      "eval_gen_len": 50.5517,
      "eval_loss": 3.264512538909912,
      "eval_runtime": 220.31,
      "eval_samples_per_second": 4.525,
      "eval_steps_per_second": 0.286,
      "step": 1800
    },
    {
      "epoch": 15.748031496062993,
      "grad_norm": 1.4287927150726318,
      "learning_rate": 0.0001062992125984252,
      "loss": 0.2933,
      "step": 2000
    },
    {
      "epoch": 15.748031496062993,
      "eval_bleu": 7.9015,
      "eval_gen_len": 49.6169,
      "eval_loss": 3.3970730304718018,
      "eval_runtime": 195.9873,
      "eval_samples_per_second": 5.087,
      "eval_steps_per_second": 0.321,
      "step": 2000
    },
    {
      "epoch": 17.322834645669293,
      "grad_norm": 1.6358137130737305,
      "learning_rate": 6.692913385826773e-05,
      "loss": 0.2447,
      "step": 2200
    },
    {
      "epoch": 17.322834645669293,
      "eval_bleu": 7.8441,
      "eval_gen_len": 49.6911,
      "eval_loss": 3.5039305686950684,
      "eval_runtime": 201.4971,
      "eval_samples_per_second": 4.948,
      "eval_steps_per_second": 0.313,
      "step": 2200
    },
    {
      "epoch": 18.89763779527559,
      "grad_norm": 1.3624520301818848,
      "learning_rate": 2.7559055118110236e-05,
      "loss": 0.2151,
      "step": 2400
    },
    {
      "epoch": 18.89763779527559,
      "eval_bleu": 7.8198,
      "eval_gen_len": 50.1153,
      "eval_loss": 3.5556399822235107,
      "eval_runtime": 195.3596,
      "eval_samples_per_second": 5.103,
      "eval_steps_per_second": 0.322,
      "step": 2400
    },
    {
      "epoch": 20.0,
      "step": 2540,
      "total_flos": 2811538357714944.0,
      "train_loss": 1.096170819650485,
      "train_runtime": 5288.9286,
      "train_samples_per_second": 3.827,
      "train_steps_per_second": 0.48
    }
  ],
  "logging_steps": 200,
  "max_steps": 2540,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 20,
  "save_steps": 200,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2811538357714944.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}