{
  "best_metric": 42.7412,
  "best_model_checkpoint": "opus_base_adapt_wce_gloss_unsampled_bands_7_ubweight_1.5/checkpoint-48000",
  "epoch": 2.365930599369085,
  "global_step": 48000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.2,
      "learning_rate": 1.9753918572555207e-05,
      "loss": 0.2075,
      "step": 4000
    },
    {
      "epoch": 0.2,
      "eval_bleu": 35.693,
      "eval_gen_len": 39.837,
      "eval_loss": 0.10906277596950531,
      "eval_runtime": 203.6881,
      "eval_samples_per_second": 5.121,
      "eval_steps_per_second": 0.162,
      "step": 4000
    },
    {
      "epoch": 0.39,
      "learning_rate": 1.950746746845426e-05,
      "loss": 0.1742,
      "step": 8000
    },
    {
      "epoch": 0.39,
      "eval_bleu": 37.6497,
      "eval_gen_len": 37.93,
      "eval_loss": 0.10657652467489243,
      "eval_runtime": 221.5331,
      "eval_samples_per_second": 4.708,
      "eval_steps_per_second": 0.149,
      "step": 8000
    },
    {
      "epoch": 0.59,
      "learning_rate": 1.9261016364353314e-05,
      "loss": 0.1672,
      "step": 12000
    },
    {
      "epoch": 0.59,
      "eval_bleu": 41.3565,
      "eval_gen_len": 34.6529,
      "eval_loss": 0.10534726828336716,
      "eval_runtime": 168.8583,
      "eval_samples_per_second": 6.177,
      "eval_steps_per_second": 0.195,
      "step": 12000
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.9014565260252367e-05,
      "loss": 0.1617,
      "step": 16000
    },
    {
      "epoch": 0.79,
      "eval_bleu": 41.4881,
      "eval_gen_len": 35.4343,
      "eval_loss": 0.10427802801132202,
      "eval_runtime": 136.9034,
      "eval_samples_per_second": 7.619,
      "eval_steps_per_second": 0.241,
      "step": 16000
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.876811415615142e-05,
      "loss": 0.1593,
      "step": 20000
    },
    {
      "epoch": 0.99,
      "eval_bleu": 41.6185,
      "eval_gen_len": 35.9434,
      "eval_loss": 0.10372872650623322,
      "eval_runtime": 158.7854,
      "eval_samples_per_second": 6.569,
      "eval_steps_per_second": 0.208,
      "step": 20000
    },
    {
      "epoch": 1.18,
      "learning_rate": 1.8521786277602524e-05,
      "loss": 0.1475,
      "step": 24000
    },
    {
      "epoch": 1.18,
      "eval_bleu": 41.9106,
      "eval_gen_len": 35.1831,
      "eval_loss": 0.10363597422838211,
      "eval_runtime": 127.398,
      "eval_samples_per_second": 8.187,
      "eval_steps_per_second": 0.259,
      "step": 24000
    },
    {
      "epoch": 1.38,
      "learning_rate": 1.8275520011829655e-05,
      "loss": 0.1473,
      "step": 28000
    },
    {
      "epoch": 1.38,
      "eval_bleu": 41.0151,
      "eval_gen_len": 36.2119,
      "eval_loss": 0.10283295065164566,
      "eval_runtime": 132.5317,
      "eval_samples_per_second": 7.87,
      "eval_steps_per_second": 0.249,
      "step": 28000
    },
    {
      "epoch": 1.58,
      "learning_rate": 1.8029130520504733e-05,
      "loss": 0.1469,
      "step": 32000
    },
    {
      "epoch": 1.58,
      "eval_bleu": 42.0417,
      "eval_gen_len": 34.9779,
      "eval_loss": 0.1022760346531868,
      "eval_runtime": 134.6277,
      "eval_samples_per_second": 7.747,
      "eval_steps_per_second": 0.245,
      "step": 32000
    },
    {
      "epoch": 1.77,
      "learning_rate": 1.7782741029179815e-05,
      "loss": 0.1455,
      "step": 36000
    },
    {
      "epoch": 1.77,
      "eval_bleu": 40.3879,
      "eval_gen_len": 37.6788,
      "eval_loss": 0.10188300907611847,
      "eval_runtime": 124.1502,
      "eval_samples_per_second": 8.401,
      "eval_steps_per_second": 0.266,
      "step": 36000
    },
    {
      "epoch": 1.97,
      "learning_rate": 1.7536413150630914e-05,
      "loss": 0.1447,
      "step": 40000
    },
    {
      "epoch": 1.97,
      "eval_bleu": 41.9744,
      "eval_gen_len": 35.4314,
      "eval_loss": 0.10198679566383362,
      "eval_runtime": 133.2603,
      "eval_samples_per_second": 7.827,
      "eval_steps_per_second": 0.248,
      "step": 40000
    },
    {
      "epoch": 2.17,
      "learning_rate": 1.729008527208202e-05,
      "loss": 0.1371,
      "step": 44000
    },
    {
      "epoch": 2.17,
      "eval_bleu": 42.1364,
      "eval_gen_len": 34.9569,
      "eval_loss": 0.10233627259731293,
      "eval_runtime": 158.4624,
      "eval_samples_per_second": 6.582,
      "eval_steps_per_second": 0.208,
      "step": 44000
    },
    {
      "epoch": 2.37,
      "learning_rate": 1.70436957807571e-05,
      "loss": 0.1362,
      "step": 48000
    },
    {
      "epoch": 2.37,
      "eval_bleu": 42.7412,
      "eval_gen_len": 34.8322,
      "eval_loss": 0.10221327096223831,
      "eval_runtime": 107.2167,
      "eval_samples_per_second": 9.728,
      "eval_steps_per_second": 0.308,
      "step": 48000
    }
  ],
  "max_steps": 324608,
  "num_train_epochs": 16,
  "total_flos": 8.279444701052928e+16,
  "trial_name": null,
  "trial_params": null
}