trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 5.0,
  "eval_steps": 10,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.8,
      "grad_norm": 3.2206196784973145,
      "learning_rate": 0.0001,
      "loss": 2.8223,
      "step": 1
    },
    {
      "epoch": 1.0,
      "grad_norm": 3.775913715362549,
      "learning_rate": 0.0002,
      "loss": 2.8268,
      "step": 2
    },
    {
      "epoch": 1.0,
      "eval_loss": 2.808971643447876,
      "eval_runtime": 0.5089,
      "eval_samples_per_second": 1.965,
      "eval_steps_per_second": 1.965,
      "step": 2
    },
    {
      "epoch": 1.8,
      "grad_norm": 2.4820940494537354,
      "learning_rate": 0.000175,
      "loss": 2.6863,
      "step": 3
    },
    {
      "epoch": 2.0,
      "grad_norm": 2.1421539783477783,
      "learning_rate": 0.00015000000000000001,
      "loss": 2.4484,
      "step": 4
    },
    {
      "epoch": 2.0,
      "eval_loss": 2.5902364253997803,
      "eval_runtime": 0.5114,
      "eval_samples_per_second": 1.955,
      "eval_steps_per_second": 1.955,
      "step": 4
    },
    {
      "epoch": 2.8,
      "grad_norm": 1.6769822835922241,
      "learning_rate": 0.000125,
      "loss": 2.3966,
      "step": 5
    },
    {
      "epoch": 3.0,
      "grad_norm": 2.2351601123809814,
      "learning_rate": 0.0001,
      "loss": 2.3563,
      "step": 6
    },
    {
      "epoch": 3.0,
      "eval_loss": 2.461848020553589,
      "eval_runtime": 0.5161,
      "eval_samples_per_second": 1.938,
      "eval_steps_per_second": 1.938,
      "step": 6
    },
    {
      "epoch": 3.8,
      "grad_norm": 1.7321960926055908,
      "learning_rate": 7.500000000000001e-05,
      "loss": 2.2687,
      "step": 7
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.719794750213623,
      "learning_rate": 5e-05,
      "loss": 2.1538,
      "step": 8
    },
    {
      "epoch": 4.0,
      "eval_loss": 2.367581844329834,
      "eval_runtime": 0.5201,
      "eval_samples_per_second": 1.923,
      "eval_steps_per_second": 1.923,
      "step": 8
    },
    {
      "epoch": 4.8,
      "grad_norm": 1.1918765306472778,
      "learning_rate": 2.5e-05,
      "loss": 2.1569,
      "step": 9
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.5512747764587402,
      "learning_rate": 0.0,
      "loss": 2.1331,
      "step": 10
    },
    {
      "epoch": 5.0,
      "eval_loss": 2.331345319747925,
      "eval_runtime": 0.5274,
      "eval_samples_per_second": 1.896,
      "eval_steps_per_second": 1.896,
      "step": 10
    },
    {
      "epoch": 5.0,
      "step": 10,
      "total_flos": 549306960445440.0,
      "train_loss": 2.424940276145935,
      "train_runtime": 55.452,
      "train_samples_per_second": 1.443,
      "train_steps_per_second": 0.18
    }
  ],
  "logging_steps": 1,
  "max_steps": 10,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 549306960445440.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}