trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.9372937293729375,
  "eval_steps": 500,
  "global_step": 111,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.264026402640264,
      "grad_norm": 12.022938086535472,
      "learning_rate": 5e-06,
      "loss": 1.1583,
      "step": 10
    },
    {
      "epoch": 0.528052805280528,
      "grad_norm": 2.088240001039349,
      "learning_rate": 5e-06,
      "loss": 1.0448,
      "step": 20
    },
    {
      "epoch": 0.7920792079207921,
      "grad_norm": 1.3000719336915976,
      "learning_rate": 5e-06,
      "loss": 0.9949,
      "step": 30
    },
    {
      "epoch": 0.976897689768977,
      "eval_loss": 0.9789440631866455,
      "eval_runtime": 40.3765,
      "eval_samples_per_second": 25.213,
      "eval_steps_per_second": 0.396,
      "step": 37
    },
    {
      "epoch": 1.0594059405940595,
      "grad_norm": 0.9713352191576023,
      "learning_rate": 5e-06,
      "loss": 0.9862,
      "step": 40
    },
    {
      "epoch": 1.3234323432343233,
      "grad_norm": 0.9308222764215573,
      "learning_rate": 5e-06,
      "loss": 0.9269,
      "step": 50
    },
    {
      "epoch": 1.5874587458745875,
      "grad_norm": 0.7459606179464745,
      "learning_rate": 5e-06,
      "loss": 0.9154,
      "step": 60
    },
    {
      "epoch": 1.8514851485148514,
      "grad_norm": 1.2070339110976707,
      "learning_rate": 5e-06,
      "loss": 0.9063,
      "step": 70
    },
    {
      "epoch": 1.9834983498349836,
      "eval_loss": 0.9386385679244995,
      "eval_runtime": 40.383,
      "eval_samples_per_second": 25.209,
      "eval_steps_per_second": 0.396,
      "step": 75
    },
    {
      "epoch": 2.118811881188119,
      "grad_norm": 0.8653617914235142,
      "learning_rate": 5e-06,
      "loss": 0.9088,
      "step": 80
    },
    {
      "epoch": 2.382838283828383,
      "grad_norm": 0.7418041958945455,
      "learning_rate": 5e-06,
      "loss": 0.852,
      "step": 90
    },
    {
      "epoch": 2.6468646864686467,
      "grad_norm": 0.7211883671655479,
      "learning_rate": 5e-06,
      "loss": 0.8509,
      "step": 100
    },
    {
      "epoch": 2.910891089108911,
      "grad_norm": 0.9407139732490507,
      "learning_rate": 5e-06,
      "loss": 0.8472,
      "step": 110
    },
    {
      "epoch": 2.9372937293729375,
      "eval_loss": 0.9289506077766418,
      "eval_runtime": 39.7511,
      "eval_samples_per_second": 25.609,
      "eval_steps_per_second": 0.403,
      "step": 111
    },
    {
      "epoch": 2.9372937293729375,
      "step": 111,
      "total_flos": 185719754588160.0,
      "train_loss": 0.9437385083318831,
      "train_runtime": 6794.3487,
      "train_samples_per_second": 8.537,
      "train_steps_per_second": 0.016
    }
  ],
  "logging_steps": 10,
  "max_steps": 111,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "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": 185719754588160.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}