last-checkpoint/trainer_state.json

{
  "best_metric": 2.2401010990142822,
  "best_model_checkpoint": "dq158/pingusPongus/checkpoint-2371",
  "epoch": 2.9993674889310564,
  "eval_steps": 500,
  "global_step": 2371,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.63,
      "learning_rate": 8.49777221180502e-05,
      "loss": 2.8794,
      "step": 500
    },
    {
      "epoch": 1.0,
      "eval_bleu": 1.0,
      "eval_brevity_penalty": 1.0,
      "eval_length_ratio": 1.0,
      "eval_loss": 2.392251491546631,
      "eval_precisions": [
        1.0,
        1.0,
        1.0,
        1.0
      ],
      "eval_reference_length": 53098,
      "eval_runtime": 1401.0268,
      "eval_samples_per_second": 2.006,
      "eval_steps_per_second": 0.126,
      "eval_translation_length": 53098,
      "step": 790
    },
    {
      "epoch": 1.27,
      "learning_rate": 9.445572420019074e-05,
      "loss": 2.5652,
      "step": 1000
    },
    {
      "epoch": 1.9,
      "learning_rate": 9.999999999999999e-05,
      "loss": 2.468,
      "step": 1500
    },
    {
      "epoch": 2.0,
      "eval_bleu": 1.0,
      "eval_brevity_penalty": 1.0,
      "eval_length_ratio": 1.0,
      "eval_loss": 2.288722038269043,
      "eval_precisions": [
        1.0,
        1.0,
        1.0,
        1.0
      ],
      "eval_reference_length": 53391,
      "eval_runtime": 1340.9293,
      "eval_samples_per_second": 2.096,
      "eval_steps_per_second": 0.131,
      "eval_translation_length": 53391,
      "step": 1581
    },
    {
      "epoch": 2.53,
      "learning_rate": 0.0001,
      "loss": 2.4165,
      "step": 2000
    },
    {
      "epoch": 3.0,
      "eval_bleu": 1.0,
      "eval_brevity_penalty": 1.0,
      "eval_length_ratio": 1.0,
      "eval_loss": 2.2401010990142822,
      "eval_precisions": [
        1.0,
        1.0,
        1.0,
        1.0
      ],
      "eval_reference_length": 53401,
      "eval_runtime": 1342.5303,
      "eval_samples_per_second": 2.094,
      "eval_steps_per_second": 0.131,
      "eval_translation_length": 53401,
      "step": 2371
    }
  ],
  "logging_steps": 500,
  "max_steps": 23700,
  "num_train_epochs": 30,
  "save_steps": 500,
  "total_flos": 1158039818207232.0,
  "trial_name": null,
  "trial_params": null
}