last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.00012930630400558604,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.2930630400558603e-05,
      "grad_norm": 5.58854866027832,
      "learning_rate": 2e-05,
      "loss": 5.4094,
      "step": 1
    },
    {
      "epoch": 1.2930630400558603e-05,
      "eval_loss": 5.511589050292969,
      "eval_runtime": 873.0487,
      "eval_samples_per_second": 37.298,
      "eval_steps_per_second": 18.65,
      "step": 1
    },
    {
      "epoch": 2.5861260801117205e-05,
      "grad_norm": 5.671387195587158,
      "learning_rate": 4e-05,
      "loss": 5.7626,
      "step": 2
    },
    {
      "epoch": 3.879189120167581e-05,
      "grad_norm": 4.992819309234619,
      "learning_rate": 6e-05,
      "loss": 5.3517,
      "step": 3
    },
    {
      "epoch": 3.879189120167581e-05,
      "eval_loss": 5.490509986877441,
      "eval_runtime": 868.1433,
      "eval_samples_per_second": 37.509,
      "eval_steps_per_second": 18.755,
      "step": 3
    },
    {
      "epoch": 5.172252160223441e-05,
      "grad_norm": 5.726319789886475,
      "learning_rate": 8e-05,
      "loss": 5.3797,
      "step": 4
    },
    {
      "epoch": 6.465315200279302e-05,
      "grad_norm": 6.387721061706543,
      "learning_rate": 0.0001,
      "loss": 6.0841,
      "step": 5
    },
    {
      "epoch": 7.758378240335163e-05,
      "grad_norm": 5.756309509277344,
      "learning_rate": 0.00012,
      "loss": 5.4144,
      "step": 6
    },
    {
      "epoch": 7.758378240335163e-05,
      "eval_loss": 5.243377208709717,
      "eval_runtime": 866.0356,
      "eval_samples_per_second": 37.6,
      "eval_steps_per_second": 18.801,
      "step": 6
    },
    {
      "epoch": 9.051441280391022e-05,
      "grad_norm": 4.841323375701904,
      "learning_rate": 0.00014,
      "loss": 5.4706,
      "step": 7
    },
    {
      "epoch": 0.00010344504320446882,
      "grad_norm": 4.377363681793213,
      "learning_rate": 0.00016,
      "loss": 4.6354,
      "step": 8
    },
    {
      "epoch": 0.00011637567360502743,
      "grad_norm": 4.830871105194092,
      "learning_rate": 0.00018,
      "loss": 4.9439,
      "step": 9
    },
    {
      "epoch": 0.00011637567360502743,
      "eval_loss": 4.783145427703857,
      "eval_runtime": 867.392,
      "eval_samples_per_second": 37.541,
      "eval_steps_per_second": 18.771,
      "step": 9
    },
    {
      "epoch": 0.00012930630400558604,
      "grad_norm": 6.583218574523926,
      "learning_rate": 0.0002,
      "loss": 4.9939,
      "step": 10
    }
  ],
  "logging_steps": 1,
  "max_steps": 10,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 5,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 178076327608320.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}