last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0011969119671248179,
  "eval_steps": 8,
  "global_step": 30,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 3.9897065570827266e-05,
      "eval_loss": 12.323113441467285,
      "eval_runtime": 1380.3928,
      "eval_samples_per_second": 7.646,
      "eval_steps_per_second": 3.823,
      "step": 1
    },
    {
      "epoch": 0.0001196911967124818,
      "grad_norm": 37.664207458496094,
      "learning_rate": 6e-05,
      "loss": 12.5524,
      "step": 3
    },
    {
      "epoch": 0.0002393823934249636,
      "grad_norm": 37.92738723754883,
      "learning_rate": 0.00012,
      "loss": 9.0915,
      "step": 6
    },
    {
      "epoch": 0.0003191765245666181,
      "eval_loss": 0.5433661937713623,
      "eval_runtime": 1387.7552,
      "eval_samples_per_second": 7.605,
      "eval_steps_per_second": 3.803,
      "step": 8
    },
    {
      "epoch": 0.0003590735901374454,
      "grad_norm": 0.09790990501642227,
      "learning_rate": 0.00018,
      "loss": 1.1546,
      "step": 9
    },
    {
      "epoch": 0.0004787647868499272,
      "grad_norm": 6.068345069885254,
      "learning_rate": 0.00019510565162951537,
      "loss": 0.1465,
      "step": 12
    },
    {
      "epoch": 0.0005984559835624089,
      "grad_norm": 5.6719207763671875,
      "learning_rate": 0.00017071067811865476,
      "loss": 0.2842,
      "step": 15
    },
    {
      "epoch": 0.0006383530491332362,
      "eval_loss": 0.1020064577460289,
      "eval_runtime": 1387.6514,
      "eval_samples_per_second": 7.606,
      "eval_steps_per_second": 3.803,
      "step": 16
    },
    {
      "epoch": 0.0007181471802748908,
      "grad_norm": 16.075775146484375,
      "learning_rate": 0.00013090169943749476,
      "loss": 0.2982,
      "step": 18
    },
    {
      "epoch": 0.0008378383769873726,
      "grad_norm": 0.2642863392829895,
      "learning_rate": 8.435655349597689e-05,
      "loss": 0.094,
      "step": 21
    },
    {
      "epoch": 0.0009575295736998544,
      "grad_norm": 8.427238464355469,
      "learning_rate": 4.12214747707527e-05,
      "loss": 0.3002,
      "step": 24
    },
    {
      "epoch": 0.0009575295736998544,
      "eval_loss": 0.1472158133983612,
      "eval_runtime": 1387.514,
      "eval_samples_per_second": 7.606,
      "eval_steps_per_second": 3.803,
      "step": 24
    },
    {
      "epoch": 0.0010772207704123362,
      "grad_norm": 0.03284990042448044,
      "learning_rate": 1.0899347581163221e-05,
      "loss": 0.1525,
      "step": 27
    },
    {
      "epoch": 0.0011969119671248179,
      "grad_norm": 0.01697017066180706,
      "learning_rate": 0.0,
      "loss": 0.0896,
      "step": 30
    }
  ],
  "logging_steps": 3,
  "max_steps": 30,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 70,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.109738548887552e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}