last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.018001800180018002,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0018001800180018,
      "grad_norm": NaN,
      "learning_rate": 2e-05,
      "loss": 12.5554,
      "step": 1
    },
    {
      "epoch": 0.0018001800180018,
      "eval_loss": NaN,
      "eval_runtime": 12.4601,
      "eval_samples_per_second": 18.78,
      "eval_steps_per_second": 9.39,
      "step": 1
    },
    {
      "epoch": 0.0036003600360036,
      "grad_norm": NaN,
      "learning_rate": 4e-05,
      "loss": 0.0,
      "step": 2
    },
    {
      "epoch": 0.0054005400540054005,
      "grad_norm": NaN,
      "learning_rate": 6e-05,
      "loss": 0.0,
      "step": 3
    },
    {
      "epoch": 0.0054005400540054005,
      "eval_loss": NaN,
      "eval_runtime": 11.2137,
      "eval_samples_per_second": 20.867,
      "eval_steps_per_second": 10.434,
      "step": 3
    },
    {
      "epoch": 0.0072007200720072,
      "grad_norm": NaN,
      "learning_rate": 8e-05,
      "loss": 0.0,
      "step": 4
    },
    {
      "epoch": 0.009000900090009001,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 10.3744,
      "step": 5
    },
    {
      "epoch": 0.010801080108010801,
      "grad_norm": NaN,
      "learning_rate": 0.00012,
      "loss": 0.0,
      "step": 6
    },
    {
      "epoch": 0.010801080108010801,
      "eval_loss": NaN,
      "eval_runtime": 11.2625,
      "eval_samples_per_second": 20.777,
      "eval_steps_per_second": 10.388,
      "step": 6
    },
    {
      "epoch": 0.012601260126012601,
      "grad_norm": NaN,
      "learning_rate": 0.00014,
      "loss": 11.9441,
      "step": 7
    },
    {
      "epoch": 0.0144014401440144,
      "grad_norm": NaN,
      "learning_rate": 0.00016,
      "loss": 5.9032,
      "step": 8
    },
    {
      "epoch": 0.016201620162016202,
      "grad_norm": NaN,
      "learning_rate": 0.00018,
      "loss": 0.0,
      "step": 9
    },
    {
      "epoch": 0.016201620162016202,
      "eval_loss": NaN,
      "eval_runtime": 11.2555,
      "eval_samples_per_second": 20.79,
      "eval_steps_per_second": 10.395,
      "step": 9
    },
    {
      "epoch": 0.018001800180018002,
      "grad_norm": NaN,
      "learning_rate": 0.0002,
      "loss": 2.8104,
      "step": 10
    }
  ],
  "logging_steps": 1,
  "max_steps": 10,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 3,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2935721891463168.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}