trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.987012987012987,
  "eval_steps": 500,
  "global_step": 19,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.05194805194805195,
      "grad_norm": 4.959436893463135,
      "learning_rate": 2e-05,
      "loss": 2.0897,
      "step": 1
    },
    {
      "epoch": 0.1038961038961039,
      "grad_norm": 3.942784547805786,
      "learning_rate": 4e-05,
      "loss": 2.3866,
      "step": 2
    },
    {
      "epoch": 0.15584415584415584,
      "grad_norm": 4.221513748168945,
      "learning_rate": 6e-05,
      "loss": 2.4587,
      "step": 3
    },
    {
      "epoch": 0.2077922077922078,
      "grad_norm": 3.284791946411133,
      "learning_rate": 8e-05,
      "loss": 2.1799,
      "step": 4
    },
    {
      "epoch": 0.2597402597402597,
      "grad_norm": 5.71047306060791,
      "learning_rate": 0.0001,
      "loss": 2.1779,
      "step": 5
    },
    {
      "epoch": 0.3116883116883117,
      "grad_norm": 3.1526741981506348,
      "learning_rate": 0.00012,
      "loss": 2.1932,
      "step": 6
    },
    {
      "epoch": 0.36363636363636365,
      "grad_norm": 3.1342685222625732,
      "learning_rate": 0.00014,
      "loss": 1.4442,
      "step": 7
    },
    {
      "epoch": 0.4155844155844156,
      "grad_norm": 5.338088035583496,
      "learning_rate": 0.00016,
      "loss": 2.0,
      "step": 8
    },
    {
      "epoch": 0.4675324675324675,
      "grad_norm": 4.116346836090088,
      "learning_rate": 0.00018,
      "loss": 1.5286,
      "step": 9
    },
    {
      "epoch": 0.5194805194805194,
      "grad_norm": 2.5058088302612305,
      "learning_rate": 0.0002,
      "loss": 1.4238,
      "step": 10
    },
    {
      "epoch": 0.5714285714285714,
      "grad_norm": 4.636209487915039,
      "learning_rate": 0.0001999317060143023,
      "loss": 1.511,
      "step": 11
    },
    {
      "epoch": 0.6233766233766234,
      "grad_norm": 3.1061031818389893,
      "learning_rate": 0.00019972691733857883,
      "loss": 1.3241,
      "step": 12
    },
    {
      "epoch": 0.6753246753246753,
      "grad_norm": 2.3636441230773926,
      "learning_rate": 0.0001993859136895274,
      "loss": 1.044,
      "step": 13
    },
    {
      "epoch": 0.7272727272727273,
      "grad_norm": 2.5445024967193604,
      "learning_rate": 0.0001989091608371146,
      "loss": 0.9548,
      "step": 14
    },
    {
      "epoch": 0.7792207792207793,
      "grad_norm": 2.371872901916504,
      "learning_rate": 0.0001982973099683902,
      "loss": 1.0614,
      "step": 15
    },
    {
      "epoch": 0.8311688311688312,
      "grad_norm": 5.0257039070129395,
      "learning_rate": 0.00019755119679804367,
      "loss": 1.1408,
      "step": 16
    },
    {
      "epoch": 0.8831168831168831,
      "grad_norm": 1.9052318334579468,
      "learning_rate": 0.00019667184042691875,
      "loss": 0.8993,
      "step": 17
    },
    {
      "epoch": 0.935064935064935,
      "grad_norm": 2.2592344284057617,
      "learning_rate": 0.0001956604419500441,
      "loss": 1.0319,
      "step": 18
    },
    {
      "epoch": 0.987012987012987,
      "grad_norm": 2.264328718185425,
      "learning_rate": 0.00019451838281608197,
      "loss": 1.483,
      "step": 19
    }
  ],
  "logging_steps": 1,
  "max_steps": 95,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 19,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 6.307716021406925e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}