trainer_state.json

{
  "best_metric": 0.492574961905093,
  "best_model_checkpoint": "0-unsup-wc-64-768-3e-6-1300/checkpoint-1100",
  "epoch": 1.024428684003152,
  "global_step": 1300,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.08,
      "learning_rate": 3e-06,
      "loss": 261.1663,
      "step": 100
    },
    {
      "epoch": 0.16,
      "learning_rate": 2.75e-06,
      "loss": 109.7871,
      "step": 200
    },
    {
      "epoch": 0.24,
      "learning_rate": 2.5e-06,
      "loss": 56.6627,
      "step": 300
    },
    {
      "epoch": 0.32,
      "learning_rate": 2.25e-06,
      "loss": 6.4156,
      "step": 400
    },
    {
      "epoch": 0.39,
      "learning_rate": 2e-06,
      "loss": 0.997,
      "step": 500
    },
    {
      "epoch": 0.47,
      "learning_rate": 1.7500000000000002e-06,
      "loss": 0.5423,
      "step": 600
    },
    {
      "epoch": 0.55,
      "learning_rate": 1.5e-06,
      "loss": 0.3075,
      "step": 700
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.25e-06,
      "loss": 0.2421,
      "step": 800
    },
    {
      "epoch": 0.71,
      "learning_rate": 1e-06,
      "loss": 0.2157,
      "step": 900
    },
    {
      "epoch": 0.79,
      "learning_rate": 7.5e-07,
      "loss": 0.1681,
      "step": 1000
    },
    {
      "epoch": 0.87,
      "learning_rate": 5e-07,
      "loss": 0.1869,
      "step": 1100
    },
    {
      "epoch": 0.95,
      "learning_rate": 2.5e-07,
      "loss": 0.1285,
      "step": 1200
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.0,
      "loss": 0.1196,
      "step": 1300
    }
  ],
  "max_steps": 1300,
  "num_train_epochs": 2,
  "total_flos": 516606747070464.0,
  "trial_name": null,
  "trial_params": null
}