trainer_state.json

{
  "best_metric": 10.557,
  "best_model_checkpoint": "./results/10k/es-arn/es-en/checkpoint-3750",
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 3750,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.4,
      "grad_norm": 10.127298355102539,
      "learning_rate": 0.00026024,
      "loss": 4.5734,
      "step": 500
    },
    {
      "epoch": 0.8,
      "grad_norm": 8.763890266418457,
      "learning_rate": 0.00022023999999999996,
      "loss": 3.7455,
      "step": 1000
    },
    {
      "epoch": 1.0,
      "eval_++": 20.834,
      "eval_bleu": 2.66,
      "eval_chrf": 22.142,
      "eval_gen_len": 20.45,
      "eval_loss": 3.3113255500793457,
      "eval_runtime": 591.1108,
      "eval_samples_per_second": 2.115,
      "eval_sentbleu": 2.66,
      "eval_steps_per_second": 0.266,
      "step": 1250
    },
    {
      "epoch": 1.2,
      "grad_norm": 7.676456928253174,
      "learning_rate": 0.00018024,
      "loss": 3.1429,
      "step": 1500
    },
    {
      "epoch": 1.6,
      "grad_norm": 10.029934883117676,
      "learning_rate": 0.00014023999999999997,
      "loss": 2.7968,
      "step": 2000
    },
    {
      "epoch": 2.0,
      "grad_norm": 8.800146102905273,
      "learning_rate": 0.00010023999999999999,
      "loss": 2.6369,
      "step": 2500
    },
    {
      "epoch": 2.0,
      "eval_++": 29.052,
      "eval_bleu": 8.203,
      "eval_chrf": 30.432,
      "eval_gen_len": 10.555,
      "eval_loss": 2.8542518615722656,
      "eval_runtime": 90.4892,
      "eval_samples_per_second": 13.814,
      "eval_sentbleu": 8.203,
      "eval_steps_per_second": 1.735,
      "step": 2500
    },
    {
      "epoch": 2.4,
      "grad_norm": 11.469346046447754,
      "learning_rate": 6.024e-05,
      "loss": 1.9387,
      "step": 3000
    },
    {
      "epoch": 2.8,
      "grad_norm": 7.59063720703125,
      "learning_rate": 2.0239999999999996e-05,
      "loss": 1.8717,
      "step": 3500
    },
    {
      "epoch": 3.0,
      "eval_++": 32.599,
      "eval_bleu": 10.557,
      "eval_chrf": 34.097,
      "eval_gen_len": 9.137,
      "eval_loss": 2.7572648525238037,
      "eval_runtime": 56.5425,
      "eval_samples_per_second": 22.107,
      "eval_sentbleu": 10.557,
      "eval_steps_per_second": 2.777,
      "step": 3750
    }
  ],
  "logging_steps": 500,
  "max_steps": 3750,
  "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": 100767127044096.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}