last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0006520286240565961,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 6.520286240565961e-05,
      "grad_norm": 2.601851463317871,
      "learning_rate": 2e-05,
      "loss": 12.7637,
      "step": 1
    },
    {
      "epoch": 6.520286240565961e-05,
      "eval_loss": 3.082449197769165,
      "eval_runtime": 180.4522,
      "eval_samples_per_second": 35.788,
      "eval_steps_per_second": 17.894,
      "step": 1
    },
    {
      "epoch": 0.00013040572481131922,
      "grad_norm": 2.136512041091919,
      "learning_rate": 4e-05,
      "loss": 12.7267,
      "step": 2
    },
    {
      "epoch": 0.00019560858721697882,
      "grad_norm": 2.148167848587036,
      "learning_rate": 6e-05,
      "loss": 13.1466,
      "step": 3
    },
    {
      "epoch": 0.00019560858721697882,
      "eval_loss": 3.0817880630493164,
      "eval_runtime": 179.5791,
      "eval_samples_per_second": 35.962,
      "eval_steps_per_second": 17.981,
      "step": 3
    },
    {
      "epoch": 0.00026081144962263844,
      "grad_norm": 3.0716772079467773,
      "learning_rate": 8e-05,
      "loss": 12.9326,
      "step": 4
    },
    {
      "epoch": 0.00032601431202829807,
      "grad_norm": 2.6990952491760254,
      "learning_rate": 0.0001,
      "loss": 11.8885,
      "step": 5
    },
    {
      "epoch": 0.00039121717443395764,
      "grad_norm": 3.010658025741577,
      "learning_rate": 0.00012,
      "loss": 12.3449,
      "step": 6
    },
    {
      "epoch": 0.00039121717443395764,
      "eval_loss": 3.029665231704712,
      "eval_runtime": 179.8671,
      "eval_samples_per_second": 35.904,
      "eval_steps_per_second": 17.952,
      "step": 6
    },
    {
      "epoch": 0.00045642003683961726,
      "grad_norm": 4.398700714111328,
      "learning_rate": 0.00014,
      "loss": 12.7579,
      "step": 7
    },
    {
      "epoch": 0.0005216228992452769,
      "grad_norm": 5.402773380279541,
      "learning_rate": 0.00016,
      "loss": 12.2506,
      "step": 8
    },
    {
      "epoch": 0.0005868257616509365,
      "grad_norm": 6.6318535804748535,
      "learning_rate": 0.00018,
      "loss": 11.6895,
      "step": 9
    },
    {
      "epoch": 0.0005868257616509365,
      "eval_loss": 2.801266670227051,
      "eval_runtime": 180.3457,
      "eval_samples_per_second": 35.809,
      "eval_steps_per_second": 17.905,
      "step": 9
    },
    {
      "epoch": 0.0006520286240565961,
      "grad_norm": 9.701369285583496,
      "learning_rate": 0.0002,
      "loss": 11.8366,
      "step": 10
    }
  ],
  "logging_steps": 1,
  "max_steps": 10,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 3,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1632478030725120.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}