trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.3964949843384481,
  "global_step": 5000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.02,
      "learning_rate": 5e-06,
      "loss": 0.7497,
      "step": 300
    },
    {
      "epoch": 0.05,
      "learning_rate": 1e-05,
      "loss": 0.7216,
      "step": 600
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.5000000000000002e-05,
      "loss": 0.7234,
      "step": 900
    },
    {
      "epoch": 0.1,
      "learning_rate": 2e-05,
      "loss": 0.7101,
      "step": 1200
    },
    {
      "epoch": 0.12,
      "learning_rate": 1.9965904738461534e-05,
      "loss": 0.7076,
      "step": 1500
    },
    {
      "epoch": 0.14,
      "learning_rate": 1.9864303873200655e-05,
      "loss": 0.7049,
      "step": 1800
    },
    {
      "epoch": 0.17,
      "learning_rate": 1.9695211009121407e-05,
      "loss": 0.7015,
      "step": 2100
    },
    {
      "epoch": 0.19,
      "learning_rate": 1.946000599403684e-05,
      "loss": 0.7053,
      "step": 2400
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.9160292703247836e-05,
      "loss": 0.7002,
      "step": 2700
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.8798114897361602e-05,
      "loss": 0.6937,
      "step": 3000
    },
    {
      "epoch": 0.26,
      "learning_rate": 1.8375942285781164e-05,
      "loss": 0.6946,
      "step": 3300
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.7896653685627762e-05,
      "loss": 0.696,
      "step": 3600
    },
    {
      "epoch": 0.31,
      "learning_rate": 1.736351739093632e-05,
      "loss": 0.6872,
      "step": 3900
    },
    {
      "epoch": 0.33,
      "learning_rate": 1.678016888598748e-05,
      "loss": 0.6893,
      "step": 4200
    },
    {
      "epoch": 0.36,
      "learning_rate": 1.6150586054750085e-05,
      "loss": 0.6893,
      "step": 4500
    },
    {
      "epoch": 0.38,
      "learning_rate": 1.5479062055482384e-05,
      "loss": 0.6888,
      "step": 4800
    }
  ],
  "max_steps": 12610,
  "num_train_epochs": 1,
  "total_flos": 2.872350690204713e+18,
  "trial_name": null,
  "trial_params": null
}