last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.027359781121751026,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0027359781121751026,
      "grad_norm": 1.4246985912322998,
      "learning_rate": 2e-05,
      "loss": 9.8447,
      "step": 1
    },
    {
      "epoch": 0.0027359781121751026,
      "eval_loss": 2.319706916809082,
      "eval_runtime": 6.1664,
      "eval_samples_per_second": 24.974,
      "eval_steps_per_second": 12.487,
      "step": 1
    },
    {
      "epoch": 0.005471956224350205,
      "grad_norm": 1.2818539142608643,
      "learning_rate": 4e-05,
      "loss": 9.3268,
      "step": 2
    },
    {
      "epoch": 0.008207934336525308,
      "grad_norm": 1.2641276121139526,
      "learning_rate": 6e-05,
      "loss": 8.3441,
      "step": 3
    },
    {
      "epoch": 0.008207934336525308,
      "eval_loss": 2.3167338371276855,
      "eval_runtime": 6.2245,
      "eval_samples_per_second": 24.741,
      "eval_steps_per_second": 12.37,
      "step": 3
    },
    {
      "epoch": 0.01094391244870041,
      "grad_norm": 1.6363928318023682,
      "learning_rate": 8e-05,
      "loss": 9.9866,
      "step": 4
    },
    {
      "epoch": 0.013679890560875513,
      "grad_norm": 1.4630907773971558,
      "learning_rate": 0.0001,
      "loss": 8.2748,
      "step": 5
    },
    {
      "epoch": 0.016415868673050615,
      "grad_norm": 1.7034509181976318,
      "learning_rate": 0.00012,
      "loss": 8.8614,
      "step": 6
    },
    {
      "epoch": 0.016415868673050615,
      "eval_loss": 2.269967555999756,
      "eval_runtime": 6.2348,
      "eval_samples_per_second": 24.7,
      "eval_steps_per_second": 12.35,
      "step": 6
    },
    {
      "epoch": 0.019151846785225718,
      "grad_norm": 1.8625192642211914,
      "learning_rate": 0.00014,
      "loss": 9.148,
      "step": 7
    },
    {
      "epoch": 0.02188782489740082,
      "grad_norm": 1.919395089149475,
      "learning_rate": 0.00016,
      "loss": 8.318,
      "step": 8
    },
    {
      "epoch": 0.024623803009575923,
      "grad_norm": 1.9851558208465576,
      "learning_rate": 0.00018,
      "loss": 7.9338,
      "step": 9
    },
    {
      "epoch": 0.024623803009575923,
      "eval_loss": 2.1127398014068604,
      "eval_runtime": 6.3213,
      "eval_samples_per_second": 24.362,
      "eval_steps_per_second": 12.181,
      "step": 9
    },
    {
      "epoch": 0.027359781121751026,
      "grad_norm": 2.3724186420440674,
      "learning_rate": 0.0002,
      "loss": 7.5162,
      "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": 1873043106299904.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}