{
  "best_global_step": 936,
  "best_metric": 3.4453418254852295,
  "best_model_checkpoint": "igbo_en_translation_t4_optimized/checkpoint-936",
  "epoch": 2.992,
  "eval_steps": 500,
  "global_step": 936,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.16,
      "grad_norm": 105667.1875,
      "learning_rate": 4.738247863247863e-05,
      "loss": 3.908,
      "step": 50
    },
    {
      "epoch": 0.32,
      "grad_norm": 183695.078125,
      "learning_rate": 4.4711538461538466e-05,
      "loss": 3.6837,
      "step": 100
    },
    {
      "epoch": 0.48,
      "grad_norm": 105609.046875,
      "learning_rate": 4.2040598290598295e-05,
      "loss": 3.6434,
      "step": 150
    },
    {
      "epoch": 0.64,
      "grad_norm": 90329.390625,
      "learning_rate": 3.9369658119658124e-05,
      "loss": 3.5711,
      "step": 200
    },
    {
      "epoch": 0.8,
      "grad_norm": 75909.1015625,
      "learning_rate": 3.6698717948717946e-05,
      "loss": 3.5542,
      "step": 250
    },
    {
      "epoch": 0.96,
      "grad_norm": 101548.625,
      "learning_rate": 3.402777777777778e-05,
      "loss": 3.5248,
      "step": 300
    },
    {
      "epoch": 1.0,
      "eval_loss": 3.4996626377105713,
      "eval_rouge1": 18.13845106805843,
      "eval_rouge2": 4.382326032460742,
      "eval_rougeL": 16.25107422982945,
      "eval_rougeLsum": 16.185034213563082,
      "eval_runtime": 7.0086,
      "eval_samples_per_second": 27.538,
      "eval_steps_per_second": 1.855,
      "step": 313
    },
    {
      "epoch": 1.1184,
      "grad_norm": 85483.953125,
      "learning_rate": 3.135683760683761e-05,
      "loss": 3.4393,
      "step": 350
    },
    {
      "epoch": 1.2784,
      "grad_norm": 84569.3203125,
      "learning_rate": 2.8685897435897437e-05,
      "loss": 3.4524,
      "step": 400
    },
    {
      "epoch": 1.4384000000000001,
      "grad_norm": 77848.4375,
      "learning_rate": 2.601495726495727e-05,
      "loss": 3.4723,
      "step": 450
    },
    {
      "epoch": 1.5984,
      "grad_norm": 73118.3046875,
      "learning_rate": 2.3344017094017095e-05,
      "loss": 3.4171,
      "step": 500
    },
    {
      "epoch": 1.7584,
      "grad_norm": 384926.9375,
      "learning_rate": 2.0673076923076924e-05,
      "loss": 3.4344,
      "step": 550
    },
    {
      "epoch": 1.9184,
      "grad_norm": 92710.7734375,
      "learning_rate": 1.8002136752136753e-05,
      "loss": 3.4022,
      "step": 600
    },
    {
      "epoch": 2.0,
      "eval_loss": 3.4569602012634277,
      "eval_rouge1": 17.603753476397717,
      "eval_rouge2": 4.407651351529355,
      "eval_rougeL": 15.912143520543449,
      "eval_rougeLsum": 15.874740254358638,
      "eval_runtime": 6.7217,
      "eval_samples_per_second": 28.713,
      "eval_steps_per_second": 1.934,
      "step": 626
    },
    {
      "epoch": 2.0768,
      "grad_norm": 78027.09375,
      "learning_rate": 1.5331196581196582e-05,
      "loss": 3.3792,
      "step": 650
    },
    {
      "epoch": 2.2368,
      "grad_norm": 88675.40625,
      "learning_rate": 1.2660256410256411e-05,
      "loss": 3.3763,
      "step": 700
    },
    {
      "epoch": 2.3968,
      "grad_norm": 77131.90625,
      "learning_rate": 9.98931623931624e-06,
      "loss": 3.4146,
      "step": 750
    },
    {
      "epoch": 2.5568,
      "grad_norm": 108723.25,
      "learning_rate": 7.318376068376069e-06,
      "loss": 3.3924,
      "step": 800
    },
    {
      "epoch": 2.7168,
      "grad_norm": 76727.1484375,
      "learning_rate": 4.647435897435897e-06,
      "loss": 3.3895,
      "step": 850
    },
    {
      "epoch": 2.8768000000000002,
      "grad_norm": 82499.703125,
      "learning_rate": 1.9764957264957265e-06,
      "loss": 3.3928,
      "step": 900
    },
    {
      "epoch": 2.992,
      "eval_loss": 3.4453418254852295,
      "eval_rouge1": 17.902942345295795,
      "eval_rouge2": 4.319690451903335,
      "eval_rougeL": 15.969199830396672,
      "eval_rougeLsum": 15.935294540519285,
      "eval_runtime": 6.777,
      "eval_samples_per_second": 28.479,
      "eval_steps_per_second": 1.918,
      "step": 936
    }
  ],
  "logging_steps": 50,
  "max_steps": 936,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 4049426700042240.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}