last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 14.285714285714286,
  "eval_steps": 20,
  "global_step": 100,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.14285714285714285,
      "eval_loss": 3.573837995529175,
      "eval_runtime": 2.7173,
      "eval_samples_per_second": 552.387,
      "eval_steps_per_second": 4.784,
      "step": 1
    },
    {
      "epoch": 1.4285714285714286,
      "grad_norm": Infinity,
      "learning_rate": 0.00019863613034027224,
      "loss": 3.0145,
      "step": 10
    },
    {
      "epoch": 2.857142857142857,
      "grad_norm": Infinity,
      "learning_rate": 0.0001879473751206489,
      "loss": 3.0123,
      "step": 20
    },
    {
      "epoch": 2.857142857142857,
      "eval_loss": 3.611345052719116,
      "eval_runtime": 2.6681,
      "eval_samples_per_second": 562.578,
      "eval_steps_per_second": 4.872,
      "step": 20
    },
    {
      "epoch": 4.285714285714286,
      "grad_norm": Infinity,
      "learning_rate": 0.00016772815716257412,
      "loss": 3.0185,
      "step": 30
    },
    {
      "epoch": 5.714285714285714,
      "grad_norm": Infinity,
      "learning_rate": 0.00014016954246529696,
      "loss": 3.0228,
      "step": 40
    },
    {
      "epoch": 5.714285714285714,
      "eval_loss": 3.6001811027526855,
      "eval_runtime": 2.6186,
      "eval_samples_per_second": 573.214,
      "eval_steps_per_second": 4.965,
      "step": 40
    },
    {
      "epoch": 7.142857142857143,
      "grad_norm": Infinity,
      "learning_rate": 0.00010825793454723325,
      "loss": 3.0106,
      "step": 50
    },
    {
      "epoch": 8.571428571428571,
      "grad_norm": Infinity,
      "learning_rate": 7.54514512859201e-05,
      "loss": 3.0176,
      "step": 60
    },
    {
      "epoch": 8.571428571428571,
      "eval_loss": 3.629997491836548,
      "eval_runtime": 2.5632,
      "eval_samples_per_second": 585.596,
      "eval_steps_per_second": 5.072,
      "step": 60
    },
    {
      "epoch": 10.0,
      "grad_norm": Infinity,
      "learning_rate": 4.530518418775733e-05,
      "loss": 3.0127,
      "step": 70
    },
    {
      "epoch": 11.428571428571429,
      "grad_norm": Infinity,
      "learning_rate": 2.1085949060360654e-05,
      "loss": 3.0121,
      "step": 80
    },
    {
      "epoch": 11.428571428571429,
      "eval_loss": 3.595003366470337,
      "eval_runtime": 2.7208,
      "eval_samples_per_second": 551.675,
      "eval_steps_per_second": 4.778,
      "step": 80
    },
    {
      "epoch": 12.857142857142858,
      "grad_norm": Infinity,
      "learning_rate": 5.418275829936537e-06,
      "loss": 3.0185,
      "step": 90
    },
    {
      "epoch": 14.285714285714286,
      "grad_norm": Infinity,
      "learning_rate": 0.0,
      "loss": 3.0197,
      "step": 100
    },
    {
      "epoch": 14.285714285714286,
      "eval_loss": 3.5995988845825195,
      "eval_runtime": 2.6731,
      "eval_samples_per_second": 561.527,
      "eval_steps_per_second": 4.863,
      "step": 100
    }
  ],
  "logging_steps": 10,
  "max_steps": 100,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 15,
  "save_steps": 20,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2.032554215473152e+16,
  "train_batch_size": 30,
  "trial_name": null,
  "trial_params": null
}