trainer_state.json

{
  "best_metric": 2.578360080718994,
  "best_model_checkpoint": "output/sum-41/checkpoint-108",
  "epoch": 6.0,
  "global_step": 108,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.28,
      "learning_rate": 0.00011269523002449659,
      "loss": 3.268,
      "step": 5
    },
    {
      "epoch": 0.56,
      "learning_rate": 5.668773501204858e-05,
      "loss": 2.9209,
      "step": 10
    },
    {
      "epoch": 0.83,
      "learning_rate": 9.190657300387505e-06,
      "loss": 2.9717,
      "step": 15
    },
    {
      "epoch": 1.0,
      "eval_loss": 2.928959369659424,
      "eval_runtime": 1.1458,
      "eval_samples_per_second": 22.691,
      "eval_steps_per_second": 3.491,
      "step": 18
    },
    {
      "epoch": 1.11,
      "learning_rate": 4.137086214086682e-06,
      "loss": 2.7011,
      "step": 20
    },
    {
      "epoch": 1.39,
      "learning_rate": 4.513741816785908e-05,
      "loss": 2.8934,
      "step": 25
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.00010290000000000001,
      "loss": 2.793,
      "step": 30
    },
    {
      "epoch": 1.94,
      "learning_rate": 0.00013615781185663748,
      "loss": 2.7355,
      "step": 35
    },
    {
      "epoch": 2.0,
      "eval_loss": 2.6775753498077393,
      "eval_runtime": 1.2274,
      "eval_samples_per_second": 22.813,
      "eval_steps_per_second": 3.259,
      "step": 36
    },
    {
      "epoch": 2.22,
      "learning_rate": 0.0001211506487979619,
      "loss": 2.6552,
      "step": 40
    },
    {
      "epoch": 2.5,
      "learning_rate": 6.860000000000001e-05,
      "loss": 2.4943,
      "step": 45
    },
    {
      "epoch": 2.78,
      "learning_rate": 1.6049351202038163e-05,
      "loss": 2.2804,
      "step": 50
    },
    {
      "epoch": 3.0,
      "eval_loss": 2.632376194000244,
      "eval_runtime": 1.2528,
      "eval_samples_per_second": 22.349,
      "eval_steps_per_second": 3.193,
      "step": 54
    },
    {
      "epoch": 3.06,
      "learning_rate": 1.0421881433625223e-06,
      "loss": 2.6341,
      "step": 55
    },
    {
      "epoch": 3.33,
      "learning_rate": 3.4300000000000014e-05,
      "loss": 2.3676,
      "step": 60
    },
    {
      "epoch": 3.61,
      "learning_rate": 9.206258183214083e-05,
      "loss": 2.2182,
      "step": 65
    },
    {
      "epoch": 3.89,
      "learning_rate": 0.00013306291378591332,
      "loss": 2.4212,
      "step": 70
    },
    {
      "epoch": 4.0,
      "eval_loss": 2.5963516235351562,
      "eval_runtime": 1.2572,
      "eval_samples_per_second": 22.272,
      "eval_steps_per_second": 3.182,
      "step": 72
    },
    {
      "epoch": 4.17,
      "learning_rate": 0.00012800934269961248,
      "loss": 2.2587,
      "step": 75
    },
    {
      "epoch": 4.44,
      "learning_rate": 8.051226498795145e-05,
      "loss": 2.1767,
      "step": 80
    },
    {
      "epoch": 4.72,
      "learning_rate": 2.4504769975503385e-05,
      "loss": 2.1842,
      "step": 85
    },
    {
      "epoch": 5.0,
      "learning_rate": 0.0,
      "loss": 2.2178,
      "step": 90
    },
    {
      "epoch": 5.0,
      "eval_loss": 2.578657388687134,
      "eval_runtime": 1.2377,
      "eval_samples_per_second": 22.622,
      "eval_steps_per_second": 3.232,
      "step": 90
    },
    {
      "epoch": 5.28,
      "learning_rate": 2.4504769975503317e-05,
      "loss": 2.0438,
      "step": 95
    },
    {
      "epoch": 5.56,
      "learning_rate": 8.051226498795124e-05,
      "loss": 2.1724,
      "step": 100
    },
    {
      "epoch": 5.83,
      "learning_rate": 0.00012800934269961248,
      "loss": 1.9973,
      "step": 105
    },
    {
      "epoch": 6.0,
      "eval_loss": 2.578360080718994,
      "eval_runtime": 1.2443,
      "eval_samples_per_second": 22.503,
      "eval_steps_per_second": 3.215,
      "step": 108
    }
  ],
  "max_steps": 198,
  "num_train_epochs": 11,
  "total_flos": 110787821568000.0,
  "trial_name": null,
  "trial_params": null
}