last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.04337218758471131,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00433721875847113,
      "grad_norm": 213.65277099609375,
      "learning_rate": 1e-05,
      "loss": 128.6171,
      "step": 1
    },
    {
      "epoch": 0.00433721875847113,
      "eval_loss": NaN,
      "eval_runtime": 101.3618,
      "eval_samples_per_second": 3.838,
      "eval_steps_per_second": 1.924,
      "step": 1
    },
    {
      "epoch": 0.00867443751694226,
      "grad_norm": 232.8147735595703,
      "learning_rate": 2e-05,
      "loss": 131.4164,
      "step": 2
    },
    {
      "epoch": 0.013011656275413392,
      "grad_norm": 241.21810913085938,
      "learning_rate": 3e-05,
      "loss": 131.4731,
      "step": 3
    },
    {
      "epoch": 0.013011656275413392,
      "eval_loss": NaN,
      "eval_runtime": 81.9267,
      "eval_samples_per_second": 4.748,
      "eval_steps_per_second": 2.38,
      "step": 3
    },
    {
      "epoch": 0.01734887503388452,
      "grad_norm": 234.3598175048828,
      "learning_rate": 4e-05,
      "loss": 129.1868,
      "step": 4
    },
    {
      "epoch": 0.021686093792355653,
      "grad_norm": 219.65846252441406,
      "learning_rate": 5e-05,
      "loss": 121.4155,
      "step": 5
    },
    {
      "epoch": 0.026023312550826783,
      "grad_norm": 224.65374755859375,
      "learning_rate": 6e-05,
      "loss": 114.3207,
      "step": 6
    },
    {
      "epoch": 0.026023312550826783,
      "eval_loss": NaN,
      "eval_runtime": 81.9854,
      "eval_samples_per_second": 4.745,
      "eval_steps_per_second": 2.378,
      "step": 6
    },
    {
      "epoch": 0.030360531309297913,
      "grad_norm": 146.9974365234375,
      "learning_rate": 7e-05,
      "loss": 94.7878,
      "step": 7
    },
    {
      "epoch": 0.03469775006776904,
      "grad_norm": 144.07135009765625,
      "learning_rate": 8e-05,
      "loss": 81.1473,
      "step": 8
    },
    {
      "epoch": 0.03903496882624018,
      "grad_norm": 177.5219268798828,
      "learning_rate": 9e-05,
      "loss": 70.3032,
      "step": 9
    },
    {
      "epoch": 0.03903496882624018,
      "eval_loss": NaN,
      "eval_runtime": 82.0919,
      "eval_samples_per_second": 4.739,
      "eval_steps_per_second": 2.375,
      "step": 9
    },
    {
      "epoch": 0.04337218758471131,
      "grad_norm": 192.78652954101562,
      "learning_rate": 0.0001,
      "loss": 53.0197,
      "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": 9.14912188563456e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}