last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.001453646836501072,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00014536468365010721,
      "grad_norm": 12.12260913848877,
      "learning_rate": 2e-05,
      "loss": 5.4024,
      "step": 1
    },
    {
      "epoch": 0.00014536468365010721,
      "eval_loss": 4.953571796417236,
      "eval_runtime": 572.8822,
      "eval_samples_per_second": 5.057,
      "eval_steps_per_second": 2.529,
      "step": 1
    },
    {
      "epoch": 0.00029072936730021443,
      "grad_norm": 11.454992294311523,
      "learning_rate": 4e-05,
      "loss": 4.8248,
      "step": 2
    },
    {
      "epoch": 0.00043609405095032164,
      "grad_norm": 12.662338256835938,
      "learning_rate": 6e-05,
      "loss": 3.6741,
      "step": 3
    },
    {
      "epoch": 0.00043609405095032164,
      "eval_loss": 4.851102828979492,
      "eval_runtime": 557.8236,
      "eval_samples_per_second": 5.193,
      "eval_steps_per_second": 2.598,
      "step": 3
    },
    {
      "epoch": 0.0005814587346004289,
      "grad_norm": 9.953781127929688,
      "learning_rate": 8e-05,
      "loss": 3.7173,
      "step": 4
    },
    {
      "epoch": 0.000726823418250536,
      "grad_norm": 13.284978866577148,
      "learning_rate": 0.0001,
      "loss": 5.2433,
      "step": 5
    },
    {
      "epoch": 0.0008721881019006433,
      "grad_norm": 6.807127952575684,
      "learning_rate": 0.00012,
      "loss": 2.7762,
      "step": 6
    },
    {
      "epoch": 0.0008721881019006433,
      "eval_loss": 3.951345443725586,
      "eval_runtime": 1158.7862,
      "eval_samples_per_second": 2.5,
      "eval_steps_per_second": 1.25,
      "step": 6
    },
    {
      "epoch": 0.0010175527855507504,
      "grad_norm": 6.472811222076416,
      "learning_rate": 0.00014,
      "loss": 2.2953,
      "step": 7
    },
    {
      "epoch": 0.0011629174692008577,
      "grad_norm": 18.097318649291992,
      "learning_rate": 0.00016,
      "loss": 4.6898,
      "step": 8
    },
    {
      "epoch": 0.0013082821528509648,
      "grad_norm": 10.140506744384766,
      "learning_rate": 0.00018,
      "loss": 3.1614,
      "step": 9
    },
    {
      "epoch": 0.0013082821528509648,
      "eval_loss": 2.8068807125091553,
      "eval_runtime": 1166.945,
      "eval_samples_per_second": 2.483,
      "eval_steps_per_second": 1.242,
      "step": 9
    },
    {
      "epoch": 0.001453646836501072,
      "grad_norm": 7.146152496337891,
      "learning_rate": 0.0002,
      "loss": 2.4525,
      "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": 621231451668480.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}