trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 9.940438871473354,
  "eval_steps": 500,
  "global_step": 1590,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.6269592476489029,
      "grad_norm": 0.4195554256439209,
      "learning_rate": 0.0002,
      "loss": 0.6385,
      "step": 100
    },
    {
      "epoch": 1.250783699059561,
      "grad_norm": 0.3866879343986511,
      "learning_rate": 0.0002,
      "loss": 0.4144,
      "step": 200
    },
    {
      "epoch": 1.877742946708464,
      "grad_norm": 0.34548789262771606,
      "learning_rate": 0.0002,
      "loss": 0.3602,
      "step": 300
    },
    {
      "epoch": 2.501567398119122,
      "grad_norm": 0.40215349197387695,
      "learning_rate": 0.0002,
      "loss": 0.3016,
      "step": 400
    },
    {
      "epoch": 3.1253918495297803,
      "grad_norm": 0.4264250099658966,
      "learning_rate": 0.0002,
      "loss": 0.2818,
      "step": 500
    },
    {
      "epoch": 3.7523510971786833,
      "grad_norm": 0.38338279724121094,
      "learning_rate": 0.0002,
      "loss": 0.2397,
      "step": 600
    },
    {
      "epoch": 4.376175548589342,
      "grad_norm": 0.4163173735141754,
      "learning_rate": 0.0002,
      "loss": 0.2157,
      "step": 700
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.644266664981842,
      "learning_rate": 0.0002,
      "loss": 0.2047,
      "step": 800
    },
    {
      "epoch": 5.6269592476489025,
      "grad_norm": 0.4428400695323944,
      "learning_rate": 0.0002,
      "loss": 0.1671,
      "step": 900
    },
    {
      "epoch": 6.250783699059561,
      "grad_norm": 0.3752143383026123,
      "learning_rate": 0.0002,
      "loss": 0.1598,
      "step": 1000
    },
    {
      "epoch": 6.877742946708464,
      "grad_norm": 0.40711358189582825,
      "learning_rate": 0.0002,
      "loss": 0.1446,
      "step": 1100
    },
    {
      "epoch": 7.501567398119122,
      "grad_norm": 0.45269152522087097,
      "learning_rate": 0.0002,
      "loss": 0.1257,
      "step": 1200
    },
    {
      "epoch": 8.12539184952978,
      "grad_norm": 0.5310996174812317,
      "learning_rate": 0.0002,
      "loss": 0.1216,
      "step": 1300
    },
    {
      "epoch": 8.752351097178684,
      "grad_norm": 0.3715245723724365,
      "learning_rate": 0.0002,
      "loss": 0.109,
      "step": 1400
    },
    {
      "epoch": 9.376175548589341,
      "grad_norm": 0.4838401973247528,
      "learning_rate": 0.0002,
      "loss": 0.1009,
      "step": 1500
    },
    {
      "epoch": 9.940438871473354,
      "step": 1590,
      "total_flos": 3.043890401016545e+17,
      "train_loss": 0.2311508838485622,
      "train_runtime": 4878.1888,
      "train_samples_per_second": 5.225,
      "train_steps_per_second": 0.326
    }
  ],
  "logging_steps": 100,
  "max_steps": 1590,
  "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": 3.043890401016545e+17,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}