last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.027510316368638238,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.002751031636863824,
      "grad_norm": 1.327262043952942,
      "learning_rate": 2e-05,
      "loss": 1.2687,
      "step": 1
    },
    {
      "epoch": 0.002751031636863824,
      "eval_loss": 1.20850670337677,
      "eval_runtime": 5.1788,
      "eval_samples_per_second": 29.543,
      "eval_steps_per_second": 14.868,
      "step": 1
    },
    {
      "epoch": 0.005502063273727648,
      "grad_norm": NaN,
      "learning_rate": 2e-05,
      "loss": 1.3054,
      "step": 2
    },
    {
      "epoch": 0.008253094910591471,
      "grad_norm": 0.9249797463417053,
      "learning_rate": 4e-05,
      "loss": 1.0607,
      "step": 3
    },
    {
      "epoch": 0.008253094910591471,
      "eval_loss": 1.203626036643982,
      "eval_runtime": 5.1635,
      "eval_samples_per_second": 29.631,
      "eval_steps_per_second": 14.912,
      "step": 3
    },
    {
      "epoch": 0.011004126547455296,
      "grad_norm": 0.8976566791534424,
      "learning_rate": 6e-05,
      "loss": 1.1355,
      "step": 4
    },
    {
      "epoch": 0.013755158184319119,
      "grad_norm": 1.1046843528747559,
      "learning_rate": 8e-05,
      "loss": 1.2204,
      "step": 5
    },
    {
      "epoch": 0.016506189821182942,
      "grad_norm": 0.8999706506729126,
      "learning_rate": 0.0001,
      "loss": 0.9594,
      "step": 6
    },
    {
      "epoch": 0.016506189821182942,
      "eval_loss": 1.1022193431854248,
      "eval_runtime": 5.1541,
      "eval_samples_per_second": 29.685,
      "eval_steps_per_second": 14.939,
      "step": 6
    },
    {
      "epoch": 0.01925722145804677,
      "grad_norm": 1.0555391311645508,
      "learning_rate": 0.00012,
      "loss": 1.1325,
      "step": 7
    },
    {
      "epoch": 0.02200825309491059,
      "grad_norm": NaN,
      "learning_rate": 0.00012,
      "loss": 1.009,
      "step": 8
    },
    {
      "epoch": 0.024759284731774415,
      "grad_norm": 0.9899677634239197,
      "learning_rate": 0.00014,
      "loss": 1.1344,
      "step": 9
    },
    {
      "epoch": 0.024759284731774415,
      "eval_loss": 0.988933265209198,
      "eval_runtime": 5.1614,
      "eval_samples_per_second": 29.643,
      "eval_steps_per_second": 14.918,
      "step": 9
    },
    {
      "epoch": 0.027510316368638238,
      "grad_norm": 0.9675346612930298,
      "learning_rate": 0.00016,
      "loss": 0.9712,
      "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": 1023167913000960.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}