last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.02877697841726619,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0028776978417266188,
      "grad_norm": 32.09505081176758,
      "learning_rate": 2e-05,
      "loss": 8.1718,
      "step": 1
    },
    {
      "epoch": 0.0028776978417266188,
      "eval_loss": 8.294729232788086,
      "eval_runtime": 7.2285,
      "eval_samples_per_second": 10.237,
      "eval_steps_per_second": 10.237,
      "step": 1
    },
    {
      "epoch": 0.0057553956834532375,
      "grad_norm": 36.42708969116211,
      "learning_rate": 4e-05,
      "loss": 8.4587,
      "step": 2
    },
    {
      "epoch": 0.008633093525179856,
      "grad_norm": 35.663421630859375,
      "learning_rate": 6e-05,
      "loss": 7.7727,
      "step": 3
    },
    {
      "epoch": 0.008633093525179856,
      "eval_loss": 6.748553276062012,
      "eval_runtime": 7.2035,
      "eval_samples_per_second": 10.273,
      "eval_steps_per_second": 10.273,
      "step": 3
    },
    {
      "epoch": 0.011510791366906475,
      "grad_norm": Infinity,
      "learning_rate": 6e-05,
      "loss": 6.9544,
      "step": 4
    },
    {
      "epoch": 0.014388489208633094,
      "grad_norm": 37.38779067993164,
      "learning_rate": 8e-05,
      "loss": 6.7915,
      "step": 5
    },
    {
      "epoch": 0.017266187050359712,
      "grad_norm": 31.44697380065918,
      "learning_rate": 0.0001,
      "loss": 5.4368,
      "step": 6
    },
    {
      "epoch": 0.017266187050359712,
      "eval_loss": 3.5580296516418457,
      "eval_runtime": 7.235,
      "eval_samples_per_second": 10.228,
      "eval_steps_per_second": 10.228,
      "step": 6
    },
    {
      "epoch": 0.02014388489208633,
      "grad_norm": 28.851085662841797,
      "learning_rate": 0.00012,
      "loss": 3.0727,
      "step": 7
    },
    {
      "epoch": 0.02302158273381295,
      "grad_norm": 20.200868606567383,
      "learning_rate": 0.00014,
      "loss": 2.0007,
      "step": 8
    },
    {
      "epoch": 0.025899280575539568,
      "grad_norm": 27.33905792236328,
      "learning_rate": 0.00016,
      "loss": 1.6268,
      "step": 9
    },
    {
      "epoch": 0.025899280575539568,
      "eval_loss": 0.9558103680610657,
      "eval_runtime": 7.2197,
      "eval_samples_per_second": 10.25,
      "eval_steps_per_second": 10.25,
      "step": 9
    },
    {
      "epoch": 0.02877697841726619,
      "grad_norm": 23.0020809173584,
      "learning_rate": 0.00018,
      "loss": 1.1909,
      "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": 1967550690754560.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}