trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.5,
  "eval_steps": 3,
  "global_step": 57,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.008771929824561403,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 2.0802,
      "step": 1
    },
    {
      "epoch": 0.02631578947368421,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 2.0494,
      "step": 3
    },
    {
      "epoch": 0.05263157894736842,
      "grad_norm": 2.4723047601512564,
      "learning_rate": 6.000000000000001e-07,
      "loss": 2.0771,
      "step": 6
    },
    {
      "epoch": 0.07894736842105263,
      "grad_norm": 2.106898891910819,
      "learning_rate": 1.8e-06,
      "loss": 2.0245,
      "step": 9
    },
    {
      "epoch": 0.10526315789473684,
      "grad_norm": 0.8386924455230856,
      "learning_rate": 2.9432692307692307e-06,
      "loss": 2.0568,
      "step": 12
    },
    {
      "epoch": 0.13157894736842105,
      "grad_norm": 1.1653346026055935,
      "learning_rate": 2.7730769230769233e-06,
      "loss": 2.0326,
      "step": 15
    },
    {
      "epoch": 0.15789473684210525,
      "grad_norm": 1.5574035133942843,
      "learning_rate": 2.6028846153846155e-06,
      "loss": 2.0396,
      "step": 18
    },
    {
      "epoch": 0.18421052631578946,
      "grad_norm": 0.9621595083981531,
      "learning_rate": 2.4326923076923077e-06,
      "loss": 2.0336,
      "step": 21
    },
    {
      "epoch": 0.21052631578947367,
      "grad_norm": 0.9955527535333841,
      "learning_rate": 2.2625e-06,
      "loss": 1.9571,
      "step": 24
    },
    {
      "epoch": 0.23684210526315788,
      "grad_norm": 0.7832834940985813,
      "learning_rate": 2.092307692307692e-06,
      "loss": 1.9792,
      "step": 27
    },
    {
      "epoch": 0.2631578947368421,
      "grad_norm": 0.7516778470264893,
      "learning_rate": 1.9221153846153848e-06,
      "loss": 2.015,
      "step": 30
    },
    {
      "epoch": 0.2894736842105263,
      "grad_norm": 1.4307615046721256,
      "learning_rate": 1.7519230769230768e-06,
      "loss": 1.9845,
      "step": 33
    },
    {
      "epoch": 0.3157894736842105,
      "grad_norm": 0.7110172377205767,
      "learning_rate": 1.581730769230769e-06,
      "loss": 1.974,
      "step": 36
    },
    {
      "epoch": 0.34210526315789475,
      "grad_norm": 0.9472807779995442,
      "learning_rate": 1.4115384615384616e-06,
      "loss": 1.9848,
      "step": 39
    },
    {
      "epoch": 0.3684210526315789,
      "grad_norm": 0.6928271721519345,
      "learning_rate": 1.2413461538461538e-06,
      "loss": 1.9453,
      "step": 42
    },
    {
      "epoch": 0.39473684210526316,
      "grad_norm": 0.9136097540650397,
      "learning_rate": 1.071153846153846e-06,
      "loss": 1.9987,
      "step": 45
    },
    {
      "epoch": 0.42105263157894735,
      "grad_norm": 0.6671789202988747,
      "learning_rate": 9.009615384615385e-07,
      "loss": 2.0054,
      "step": 48
    },
    {
      "epoch": 0.4473684210526316,
      "grad_norm": 1.2888605208856772,
      "learning_rate": 7.307692307692307e-07,
      "loss": 1.9706,
      "step": 51
    },
    {
      "epoch": 0.47368421052631576,
      "grad_norm": 0.9597150120115726,
      "learning_rate": 6.740384615384617e-07,
      "loss": 1.9874,
      "step": 54
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.7428542672176522,
      "learning_rate": 5.038461538461539e-07,
      "loss": 2.0058,
      "step": 57
    },
    {
      "epoch": 0.5,
      "step": 57,
      "total_flos": 227154377834496.0,
      "train_loss": 2.006924344782244,
      "train_runtime": 17818.5287,
      "train_samples_per_second": 0.409,
      "train_steps_per_second": 0.003
    }
  ],
  "logging_steps": 3,
  "max_steps": 57,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 12,
  "total_flos": 227154377834496.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}