last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0017139429257005741,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0001713942925700574,
      "grad_norm": 0.6748816967010498,
      "learning_rate": 2e-05,
      "loss": 5.2217,
      "step": 1
    },
    {
      "epoch": 0.0001713942925700574,
      "eval_loss": 1.2500425577163696,
      "eval_runtime": 159.8786,
      "eval_samples_per_second": 15.368,
      "eval_steps_per_second": 7.687,
      "step": 1
    },
    {
      "epoch": 0.0003427885851401148,
      "grad_norm": 0.48369231820106506,
      "learning_rate": 4e-05,
      "loss": 5.07,
      "step": 2
    },
    {
      "epoch": 0.0005141828777101723,
      "grad_norm": 0.6595766544342041,
      "learning_rate": 6e-05,
      "loss": 4.7583,
      "step": 3
    },
    {
      "epoch": 0.0005141828777101723,
      "eval_loss": 1.249487042427063,
      "eval_runtime": 159.9385,
      "eval_samples_per_second": 15.362,
      "eval_steps_per_second": 7.684,
      "step": 3
    },
    {
      "epoch": 0.0006855771702802296,
      "grad_norm": 0.5301584601402283,
      "learning_rate": 8e-05,
      "loss": 4.1981,
      "step": 4
    },
    {
      "epoch": 0.0008569714628502871,
      "grad_norm": 0.9308584928512573,
      "learning_rate": 0.0001,
      "loss": 4.4825,
      "step": 5
    },
    {
      "epoch": 0.0010283657554203446,
      "grad_norm": 0.6746999621391296,
      "learning_rate": 0.00012,
      "loss": 4.6614,
      "step": 6
    },
    {
      "epoch": 0.0010283657554203446,
      "eval_loss": 1.2453702688217163,
      "eval_runtime": 159.9636,
      "eval_samples_per_second": 15.36,
      "eval_steps_per_second": 7.683,
      "step": 6
    },
    {
      "epoch": 0.0011997600479904018,
      "grad_norm": 0.6202865242958069,
      "learning_rate": 0.00014,
      "loss": 4.9219,
      "step": 7
    },
    {
      "epoch": 0.0013711543405604593,
      "grad_norm": 0.6713095903396606,
      "learning_rate": 0.00016,
      "loss": 4.8562,
      "step": 8
    },
    {
      "epoch": 0.0015425486331305167,
      "grad_norm": 0.7715242505073547,
      "learning_rate": 0.00018,
      "loss": 4.275,
      "step": 9
    },
    {
      "epoch": 0.0015425486331305167,
      "eval_loss": 1.2166459560394287,
      "eval_runtime": 160.1929,
      "eval_samples_per_second": 15.338,
      "eval_steps_per_second": 7.672,
      "step": 9
    },
    {
      "epoch": 0.0017139429257005741,
      "grad_norm": 1.0439811944961548,
      "learning_rate": 0.0002,
      "loss": 5.0375,
      "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": 4246709207040000.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}