trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.6961325966850829,
  "eval_steps": 22,
  "global_step": 154,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.011049723756906077,
      "grad_norm": 0.21540871262550354,
      "learning_rate": 0.0005,
      "loss": 0.7758,
      "step": 1
    },
    {
      "epoch": 0.2430939226519337,
      "grad_norm": 0.3093005120754242,
      "learning_rate": 0.0005,
      "loss": 0.6515,
      "step": 22
    },
    {
      "epoch": 0.2430939226519337,
      "eval_loss": 0.6094172596931458,
      "eval_runtime": 264.4017,
      "eval_samples_per_second": 0.318,
      "eval_steps_per_second": 0.318,
      "step": 22
    },
    {
      "epoch": 0.4861878453038674,
      "grad_norm": 0.21729212999343872,
      "learning_rate": 0.0005,
      "loss": 0.5957,
      "step": 44
    },
    {
      "epoch": 0.4861878453038674,
      "eval_loss": 0.5930584669113159,
      "eval_runtime": 264.3144,
      "eval_samples_per_second": 0.318,
      "eval_steps_per_second": 0.318,
      "step": 44
    },
    {
      "epoch": 0.7292817679558011,
      "grad_norm": 0.259674608707428,
      "learning_rate": 0.0005,
      "loss": 0.5947,
      "step": 66
    },
    {
      "epoch": 0.7292817679558011,
      "eval_loss": 0.5857287049293518,
      "eval_runtime": 264.292,
      "eval_samples_per_second": 0.318,
      "eval_steps_per_second": 0.318,
      "step": 66
    },
    {
      "epoch": 0.9723756906077348,
      "grad_norm": 0.26313361525535583,
      "learning_rate": 0.0005,
      "loss": 0.5608,
      "step": 88
    },
    {
      "epoch": 0.9723756906077348,
      "eval_loss": 0.5758072733879089,
      "eval_runtime": 264.4228,
      "eval_samples_per_second": 0.318,
      "eval_steps_per_second": 0.318,
      "step": 88
    },
    {
      "epoch": 1.2099447513812154,
      "grad_norm": 0.3272446095943451,
      "learning_rate": 0.0005,
      "loss": 0.4753,
      "step": 110
    },
    {
      "epoch": 1.2099447513812154,
      "eval_loss": 0.5889958143234253,
      "eval_runtime": 264.2995,
      "eval_samples_per_second": 0.318,
      "eval_steps_per_second": 0.318,
      "step": 110
    },
    {
      "epoch": 1.4530386740331491,
      "grad_norm": 0.3307793438434601,
      "learning_rate": 0.0005,
      "loss": 0.4744,
      "step": 132
    },
    {
      "epoch": 1.4530386740331491,
      "eval_loss": 0.5877426862716675,
      "eval_runtime": 264.3746,
      "eval_samples_per_second": 0.318,
      "eval_steps_per_second": 0.318,
      "step": 132
    },
    {
      "epoch": 1.6961325966850829,
      "grad_norm": 0.2999575436115265,
      "learning_rate": 0.0005,
      "loss": 0.4756,
      "step": 154
    },
    {
      "epoch": 1.6961325966850829,
      "eval_loss": 0.5917853116989136,
      "eval_runtime": 264.5295,
      "eval_samples_per_second": 0.318,
      "eval_steps_per_second": 0.318,
      "step": 154
    }
  ],
  "logging_steps": 22,
  "max_steps": 270,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 22,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 6.843914701539533e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}