trainer_state.json

{
  "best_metric": 1.5800185203552246,
  "best_model_checkpoint": "chinese_roberta_L-2_H-128-finetuned-MC-hyper/checkpoint-125",
  "epoch": 2.0,
  "global_step": 250,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.16,
      "learning_rate": 0.0001608,
      "loss": 1.5282,
      "step": 20
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0001368,
      "loss": 1.5563,
      "step": 40
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00011279999999999999,
      "loss": 1.5631,
      "step": 60
    },
    {
      "epoch": 0.64,
      "learning_rate": 8.879999999999999e-05,
      "loss": 1.5201,
      "step": 80
    },
    {
      "epoch": 0.8,
      "learning_rate": 6.479999999999999e-05,
      "loss": 1.526,
      "step": 100
    },
    {
      "epoch": 0.96,
      "learning_rate": 4.08e-05,
      "loss": 1.5107,
      "step": 120
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.30000001192092896,
      "eval_loss": 1.5800185203552246,
      "eval_runtime": 0.4321,
      "eval_samples_per_second": 1157.093,
      "eval_steps_per_second": 37.027,
      "step": 125
    },
    {
      "epoch": 1.12,
      "learning_rate": 1.68e-05,
      "loss": 1.4737,
      "step": 140
    },
    {
      "epoch": 1.28,
      "learning_rate": 0.0,
      "loss": 1.4647,
      "step": 160
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.0,
      "loss": 1.4731,
      "step": 180
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0,
      "loss": 1.4969,
      "step": 200
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.0,
      "loss": 1.4998,
      "step": 220
    },
    {
      "epoch": 1.92,
      "learning_rate": 0.0,
      "loss": 1.495,
      "step": 240
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.2939999997615814,
      "eval_loss": 1.5831648111343384,
      "eval_runtime": 0.4307,
      "eval_samples_per_second": 1160.788,
      "eval_steps_per_second": 37.145,
      "step": 250
    },
    {
      "epoch": 2.0,
      "step": 250,
      "total_flos": 4691397715200.0,
      "train_loss": 1.5075831718444823,
      "train_runtime": 20.08,
      "train_samples_per_second": 398.407,
      "train_steps_per_second": 12.45
    }
  ],
  "max_steps": 250,
  "num_train_epochs": 2,
  "total_flos": 4691397715200.0,
  "trial_name": null,
  "trial_params": null
}