last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.00782472613458529,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.000782472613458529,
      "grad_norm": 46.71413803100586,
      "learning_rate": 2e-05,
      "loss": 12.1042,
      "step": 1
    },
    {
      "epoch": 0.000782472613458529,
      "eval_loss": 2.958329439163208,
      "eval_runtime": 18.3707,
      "eval_samples_per_second": 29.34,
      "eval_steps_per_second": 14.697,
      "step": 1
    },
    {
      "epoch": 0.001564945226917058,
      "grad_norm": 35.501705169677734,
      "learning_rate": 4e-05,
      "loss": 11.5379,
      "step": 2
    },
    {
      "epoch": 0.002347417840375587,
      "grad_norm": 42.39483642578125,
      "learning_rate": 6e-05,
      "loss": 15.6484,
      "step": 3
    },
    {
      "epoch": 0.002347417840375587,
      "eval_loss": 2.855407238006592,
      "eval_runtime": 18.3609,
      "eval_samples_per_second": 29.356,
      "eval_steps_per_second": 14.705,
      "step": 3
    },
    {
      "epoch": 0.003129890453834116,
      "grad_norm": 30.048139572143555,
      "learning_rate": 8e-05,
      "loss": 10.6754,
      "step": 4
    },
    {
      "epoch": 0.003912363067292645,
      "grad_norm": 38.46082305908203,
      "learning_rate": 0.0001,
      "loss": 10.405,
      "step": 5
    },
    {
      "epoch": 0.004694835680751174,
      "grad_norm": 40.440921783447266,
      "learning_rate": 0.00012,
      "loss": 9.7267,
      "step": 6
    },
    {
      "epoch": 0.004694835680751174,
      "eval_loss": 2.4337451457977295,
      "eval_runtime": 18.4261,
      "eval_samples_per_second": 29.252,
      "eval_steps_per_second": 14.653,
      "step": 6
    },
    {
      "epoch": 0.005477308294209703,
      "grad_norm": 29.447078704833984,
      "learning_rate": 0.00014,
      "loss": 8.2425,
      "step": 7
    },
    {
      "epoch": 0.006259780907668232,
      "grad_norm": 26.56793212890625,
      "learning_rate": 0.00016,
      "loss": 9.423,
      "step": 8
    },
    {
      "epoch": 0.007042253521126761,
      "grad_norm": 28.494226455688477,
      "learning_rate": 0.00018,
      "loss": 7.4282,
      "step": 9
    },
    {
      "epoch": 0.007042253521126761,
      "eval_loss": 2.320192813873291,
      "eval_runtime": 18.4372,
      "eval_samples_per_second": 29.234,
      "eval_steps_per_second": 14.644,
      "step": 9
    },
    {
      "epoch": 0.00782472613458529,
      "grad_norm": 31.02028465270996,
      "learning_rate": 0.0002,
      "loss": 7.8054,
      "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": 1752669785948160.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}