{
  "best_metric": 40.4443,
  "best_model_checkpoint": "M2M100_enfr_FT/checkpoint-160000",
  "epoch": 1.9716817212781428,
  "global_step": 160000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.2,
      "learning_rate": 1.975361680365747e-05,
      "loss": 0.2246,
      "step": 16000
    },
    {
      "epoch": 0.2,
      "eval_bleu": 35.632,
      "eval_gen_len": 45.0096,
      "eval_loss": 0.13251402974128723,
      "eval_runtime": 285.0264,
      "eval_samples_per_second": 3.659,
      "eval_steps_per_second": 0.46,
      "step": 16000
    },
    {
      "epoch": 0.39,
      "learning_rate": 1.950723360731494e-05,
      "loss": 0.1526,
      "step": 32000
    },
    {
      "epoch": 0.39,
      "eval_bleu": 36.2601,
      "eval_gen_len": 45.5829,
      "eval_loss": 0.12710681557655334,
      "eval_runtime": 288.5488,
      "eval_samples_per_second": 3.615,
      "eval_steps_per_second": 0.454,
      "step": 32000
    },
    {
      "epoch": 0.59,
      "learning_rate": 1.926085041097241e-05,
      "loss": 0.1444,
      "step": 48000
    },
    {
      "epoch": 0.59,
      "eval_bleu": 37.0395,
      "eval_gen_len": 44.7641,
      "eval_loss": 0.12176303565502167,
      "eval_runtime": 283.8778,
      "eval_samples_per_second": 3.674,
      "eval_steps_per_second": 0.461,
      "step": 48000
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.9014498022156776e-05,
      "loss": 0.1381,
      "step": 64000
    },
    {
      "epoch": 0.79,
      "eval_bleu": 37.6851,
      "eval_gen_len": 45.0633,
      "eval_loss": 0.11842654645442963,
      "eval_runtime": 281.5142,
      "eval_samples_per_second": 3.705,
      "eval_steps_per_second": 0.465,
      "step": 64000
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.8768114825814246e-05,
      "loss": 0.135,
      "step": 80000
    },
    {
      "epoch": 0.99,
      "eval_bleu": 38.8165,
      "eval_gen_len": 45.2244,
      "eval_loss": 0.11733223497867584,
      "eval_runtime": 282.3739,
      "eval_samples_per_second": 3.694,
      "eval_steps_per_second": 0.464,
      "step": 80000
    },
    {
      "epoch": 1.18,
      "learning_rate": 1.852174703323516e-05,
      "loss": 0.1205,
      "step": 96000
    },
    {
      "epoch": 1.18,
      "eval_bleu": 39.4115,
      "eval_gen_len": 45.0767,
      "eval_loss": 0.11619798094034195,
      "eval_runtime": 279.629,
      "eval_samples_per_second": 3.73,
      "eval_steps_per_second": 0.468,
      "step": 96000
    },
    {
      "epoch": 1.38,
      "learning_rate": 1.827537924065608e-05,
      "loss": 0.1203,
      "step": 112000
    },
    {
      "epoch": 1.38,
      "eval_bleu": 38.9974,
      "eval_gen_len": 45.0451,
      "eval_loss": 0.11450415849685669,
      "eval_runtime": 283.1117,
      "eval_samples_per_second": 3.684,
      "eval_steps_per_second": 0.463,
      "step": 112000
    },
    {
      "epoch": 1.58,
      "learning_rate": 1.802899604431355e-05,
      "loss": 0.1199,
      "step": 128000
    },
    {
      "epoch": 1.58,
      "eval_bleu": 38.9989,
      "eval_gen_len": 45.0268,
      "eval_loss": 0.1141299158334732,
      "eval_runtime": 283.9785,
      "eval_samples_per_second": 3.673,
      "eval_steps_per_second": 0.461,
      "step": 128000
    },
    {
      "epoch": 1.77,
      "learning_rate": 1.7782628251734466e-05,
      "loss": 0.1187,
      "step": 144000
    },
    {
      "epoch": 1.77,
      "eval_bleu": 39.3658,
      "eval_gen_len": 44.8562,
      "eval_loss": 0.11224538832902908,
      "eval_runtime": 281.1031,
      "eval_samples_per_second": 3.71,
      "eval_steps_per_second": 0.466,
      "step": 144000
    },
    {
      "epoch": 1.97,
      "learning_rate": 1.7536275862918832e-05,
      "loss": 0.1178,
      "step": 160000
    },
    {
      "epoch": 1.97,
      "eval_bleu": 40.4443,
      "eval_gen_len": 44.8341,
      "eval_loss": 0.11058922857046127,
      "eval_runtime": 283.3732,
      "eval_samples_per_second": 3.681,
      "eval_steps_per_second": 0.462,
      "step": 160000
    }
  ],
  "max_steps": 1298384,
  "num_train_epochs": 16,
  "total_flos": 5.178177255993508e+17,
  "trial_name": null,
  "trial_params": null
}