last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0010319118747258984,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00010319118747258984,
      "grad_norm": 2.621241569519043,
      "learning_rate": 2e-05,
      "loss": 2.1835,
      "step": 1
    },
    {
      "epoch": 0.00010319118747258984,
      "eval_loss": NaN,
      "eval_runtime": 945.6357,
      "eval_samples_per_second": 4.316,
      "eval_steps_per_second": 2.158,
      "step": 1
    },
    {
      "epoch": 0.00020638237494517968,
      "grad_norm": 2.2484052181243896,
      "learning_rate": 4e-05,
      "loss": 1.576,
      "step": 2
    },
    {
      "epoch": 0.00030957356241776953,
      "grad_norm": 3.5844671726226807,
      "learning_rate": 6e-05,
      "loss": 2.6,
      "step": 3
    },
    {
      "epoch": 0.00030957356241776953,
      "eval_loss": NaN,
      "eval_runtime": 660.2385,
      "eval_samples_per_second": 6.181,
      "eval_steps_per_second": 3.091,
      "step": 3
    },
    {
      "epoch": 0.00041276474989035937,
      "grad_norm": NaN,
      "learning_rate": 8e-05,
      "loss": 3.505,
      "step": 4
    },
    {
      "epoch": 0.0005159559373629492,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 3.741,
      "step": 5
    },
    {
      "epoch": 0.0006191471248355391,
      "grad_norm": 2.921973466873169,
      "learning_rate": 0.00012,
      "loss": 1.8695,
      "step": 6
    },
    {
      "epoch": 0.0006191471248355391,
      "eval_loss": NaN,
      "eval_runtime": 748.5948,
      "eval_samples_per_second": 5.452,
      "eval_steps_per_second": 2.726,
      "step": 6
    },
    {
      "epoch": 0.0007223383123081289,
      "grad_norm": 4.352147579193115,
      "learning_rate": 0.00014,
      "loss": 2.0507,
      "step": 7
    },
    {
      "epoch": 0.0008255294997807187,
      "grad_norm": 2.661231756210327,
      "learning_rate": 0.00016,
      "loss": 1.3541,
      "step": 8
    },
    {
      "epoch": 0.0009287206872533086,
      "grad_norm": 2.879901170730591,
      "learning_rate": 0.00018,
      "loss": 1.9633,
      "step": 9
    },
    {
      "epoch": 0.0009287206872533086,
      "eval_loss": NaN,
      "eval_runtime": 910.3312,
      "eval_samples_per_second": 4.483,
      "eval_steps_per_second": 2.242,
      "step": 9
    },
    {
      "epoch": 0.0010319118747258984,
      "grad_norm": 3.5780322551727295,
      "learning_rate": 0.0002,
      "loss": 2.0646,
      "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": 5225796096491520.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}