trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 10,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.1,
      "grad_norm": 1.1105867624282837,
      "learning_rate": 1e-05,
      "loss": 2.023,
      "step": 1
    },
    {
      "epoch": 0.2,
      "grad_norm": 1.0291038751602173,
      "learning_rate": 9e-06,
      "loss": 1.9548,
      "step": 2
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.9957126975059509,
      "learning_rate": 8.000000000000001e-06,
      "loss": 1.9376,
      "step": 3
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.9828665852546692,
      "learning_rate": 7e-06,
      "loss": 1.975,
      "step": 4
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.9356377720832825,
      "learning_rate": 6e-06,
      "loss": 1.9225,
      "step": 5
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.9131035208702087,
      "learning_rate": 5e-06,
      "loss": 1.9317,
      "step": 6
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.8676958084106445,
      "learning_rate": 4.000000000000001e-06,
      "loss": 1.8976,
      "step": 7
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.84479159116745,
      "learning_rate": 3e-06,
      "loss": 1.8474,
      "step": 8
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.849846363067627,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 1.897,
      "step": 9
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.7898678779602051,
      "learning_rate": 1.0000000000000002e-06,
      "loss": 1.7788,
      "step": 10
    },
    {
      "epoch": 1.0,
      "eval_loss": 1.870275616645813,
      "eval_runtime": 5.1424,
      "eval_samples_per_second": 0.389,
      "eval_steps_per_second": 0.194,
      "step": 10
    }
  ],
  "logging_steps": 1.0,
  "max_steps": 10,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 0,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 6.019423562576691e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}