trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.0,
  "global_step": 1280,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.08,
      "learning_rate": 5e-05,
      "loss": 10.9274,
      "step": 50
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.0001,
      "loss": 9.7447,
      "step": 100
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0001,
      "loss": 7.7718,
      "step": 150
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.0001,
      "loss": 6.8433,
      "step": 200
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.0001,
      "loss": 6.399,
      "step": 250
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.0001,
      "loss": 6.1445,
      "step": 300
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.0001,
      "loss": 5.9761,
      "step": 350
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.0001,
      "loss": 5.8578,
      "step": 400
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.0001,
      "loss": 5.7772,
      "step": 450
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.0001,
      "loss": 5.7197,
      "step": 500
    },
    {
      "epoch": 0.86,
      "learning_rate": 0.0001,
      "loss": 5.633,
      "step": 550
    },
    {
      "epoch": 0.94,
      "learning_rate": 0.0001,
      "loss": 5.6076,
      "step": 600
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.0001,
      "loss": 5.5512,
      "step": 650
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0001,
      "loss": 5.5049,
      "step": 700
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0001,
      "loss": 5.4586,
      "step": 750
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.0001,
      "loss": 5.4495,
      "step": 800
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.0001,
      "loss": 5.4038,
      "step": 850
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.0001,
      "loss": 5.3907,
      "step": 900
    },
    {
      "epoch": 1.48,
      "learning_rate": 0.0001,
      "loss": 5.3733,
      "step": 950
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.0001,
      "loss": 5.3187,
      "step": 1000
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.0001,
      "loss": 5.2799,
      "step": 1050
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.0001,
      "loss": 5.2512,
      "step": 1100
    },
    {
      "epoch": 1.8,
      "learning_rate": 0.0001,
      "loss": 5.2537,
      "step": 1150
    },
    {
      "epoch": 1.88,
      "learning_rate": 0.0001,
      "loss": 5.2417,
      "step": 1200
    },
    {
      "epoch": 1.95,
      "learning_rate": 0.0001,
      "loss": 5.1957,
      "step": 1250
    },
    {
      "epoch": 2.0,
      "step": 1280,
      "total_flos": 1889639433529344.0,
      "train_loss": 6.061862027645111,
      "train_runtime": 2497.1163,
      "train_samples_per_second": 65.604,
      "train_steps_per_second": 0.513
    }
  ],
  "max_steps": 1280,
  "num_train_epochs": 2,
  "total_flos": 1889639433529344.0,
  "trial_name": null,
  "trial_params": null
}