{
  "best_global_step": 270,
  "best_metric": 0.9539986252784729,
  "best_model_checkpoint": "j05hr3d/LoRA-Q-30B-R8-NoComments/checkpoint-270",
  "epoch": 10.0,
  "eval_steps": 30,
  "global_step": 280,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.35714285714285715,
      "grad_norm": 0.3198702037334442,
      "learning_rate": 2e-05,
      "loss": 1.2794,
      "step": 10
    },
    {
      "epoch": 0.7142857142857143,
      "grad_norm": 0.34039074182510376,
      "learning_rate": 1.9932881112974298e-05,
      "loss": 1.4269,
      "step": 20
    },
    {
      "epoch": 1.0714285714285714,
      "grad_norm": 0.43611446022987366,
      "learning_rate": 1.9732425440896298e-05,
      "loss": 1.0811,
      "step": 30
    },
    {
      "epoch": 1.0714285714285714,
      "eval_loss": 1.0560672283172607,
      "eval_runtime": 334.6185,
      "eval_samples_per_second": 0.069,
      "eval_steps_per_second": 0.036,
      "step": 30
    },
    {
      "epoch": 1.4285714285714286,
      "grad_norm": 0.288340300321579,
      "learning_rate": 1.9401323856087573e-05,
      "loss": 1.3697,
      "step": 40
    },
    {
      "epoch": 1.7857142857142856,
      "grad_norm": 0.26042041182518005,
      "learning_rate": 1.894402099252109e-05,
      "loss": 1.215,
      "step": 50
    },
    {
      "epoch": 2.142857142857143,
      "grad_norm": 0.1834608018398285,
      "learning_rate": 1.8366655582044096e-05,
      "loss": 1.0913,
      "step": 60
    },
    {
      "epoch": 2.142857142857143,
      "eval_loss": 1.0087836980819702,
      "eval_runtime": 332.2708,
      "eval_samples_per_second": 0.069,
      "eval_steps_per_second": 0.036,
      "step": 60
    },
    {
      "epoch": 2.5,
      "grad_norm": 0.13914090394973755,
      "learning_rate": 1.7676978049408262e-05,
      "loss": 1.4309,
      "step": 70
    },
    {
      "epoch": 2.857142857142857,
      "grad_norm": 0.5253108143806458,
      "learning_rate": 1.6884246472293018e-05,
      "loss": 1.0997,
      "step": 80
    },
    {
      "epoch": 3.2142857142857144,
      "grad_norm": 0.13747122883796692,
      "learning_rate": 1.5999102302931585e-05,
      "loss": 0.9658,
      "step": 90
    },
    {
      "epoch": 3.2142857142857144,
      "eval_loss": 0.9858900904655457,
      "eval_runtime": 331.9894,
      "eval_samples_per_second": 0.069,
      "eval_steps_per_second": 0.036,
      "step": 90
    },
    {
      "epoch": 3.571428571428571,
      "grad_norm": 0.14882412552833557,
      "learning_rate": 1.503342751962493e-05,
      "loss": 1.4497,
      "step": 100
    },
    {
      "epoch": 3.928571428571429,
      "grad_norm": 0.1565730720758438,
      "learning_rate": 1.4000185125709919e-05,
      "loss": 0.937,
      "step": 110
    },
    {
      "epoch": 4.285714285714286,
      "grad_norm": 0.1836182028055191,
      "learning_rate": 1.2913245137088024e-05,
      "loss": 1.0058,
      "step": 120
    },
    {
      "epoch": 4.285714285714286,
      "eval_loss": 0.9723401069641113,
      "eval_runtime": 331.9207,
      "eval_samples_per_second": 0.069,
      "eval_steps_per_second": 0.036,
      "step": 120
    },
    {
      "epoch": 4.642857142857143,
      "grad_norm": 0.14099319279193878,
      "learning_rate": 1.178719839421925e-05,
      "loss": 1.4094,
      "step": 130
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.12579289078712463,
      "learning_rate": 1.0637160697927651e-05,
      "loss": 0.9489,
      "step": 140
    },
    {
      "epoch": 5.357142857142857,
      "grad_norm": 0.07630683481693268,
      "learning_rate": 9.478569898255765e-06,
      "loss": 1.0988,
      "step": 150
    },
    {
      "epoch": 5.357142857142857,
      "eval_loss": 0.9634329080581665,
      "eval_runtime": 333.0314,
      "eval_samples_per_second": 0.069,
      "eval_steps_per_second": 0.036,
      "step": 150
    },
    {
      "epoch": 5.714285714285714,
      "grad_norm": 0.18268221616744995,
      "learning_rate": 8.326978660202034e-06,
      "loss": 1.2879,
      "step": 160
    },
    {
      "epoch": 6.071428571428571,
      "grad_norm": 0.21017637848854065,
      "learning_rate": 7.197845688207805e-06,
      "loss": 0.9358,
      "step": 170
    },
    {
      "epoch": 6.428571428571429,
      "grad_norm": 0.08567366749048233,
      "learning_rate": 6.106328211949928e-06,
      "loss": 1.2672,
      "step": 180
    },
    {
      "epoch": 6.428571428571429,
      "eval_loss": 0.958134651184082,
      "eval_runtime": 333.6422,
      "eval_samples_per_second": 0.069,
      "eval_steps_per_second": 0.036,
      "step": 180
    },
    {
      "epoch": 6.785714285714286,
      "grad_norm": 0.17129184305667877,
      "learning_rate": 5.067078519063514e-06,
      "loss": 1.1062,
      "step": 190
    },
    {
      "epoch": 7.142857142857143,
      "grad_norm": 0.14120297133922577,
      "learning_rate": 4.094047266094225e-06,
      "loss": 0.9977,
      "step": 200
    },
    {
      "epoch": 7.5,
      "grad_norm": 0.10870719701051712,
      "learning_rate": 3.2002962079901743e-06,
      "loss": 1.3735,
      "step": 210
    },
    {
      "epoch": 7.5,
      "eval_loss": 0.9554634094238281,
      "eval_runtime": 334.4155,
      "eval_samples_per_second": 0.069,
      "eval_steps_per_second": 0.036,
      "step": 210
    },
    {
      "epoch": 7.857142857142857,
      "grad_norm": 0.4395056366920471,
      "learning_rate": 2.3978228600109564e-06,
      "loss": 1.0173,
      "step": 220
    },
    {
      "epoch": 8.214285714285714,
      "grad_norm": 0.16671957075595856,
      "learning_rate": 1.6973994457534026e-06,
      "loss": 0.8987,
      "step": 230
    },
    {
      "epoch": 8.571428571428571,
      "grad_norm": 0.134662464261055,
      "learning_rate": 1.1084282932198543e-06,
      "loss": 1.4106,
      "step": 240
    },
    {
      "epoch": 8.571428571428571,
      "eval_loss": 0.9541482925415039,
      "eval_runtime": 335.4858,
      "eval_samples_per_second": 0.069,
      "eval_steps_per_second": 0.036,
      "step": 240
    },
    {
      "epoch": 8.928571428571429,
      "grad_norm": 0.15510539710521698,
      "learning_rate": 6.388156200599726e-07,
      "loss": 0.8852,
      "step": 250
    },
    {
      "epoch": 9.285714285714286,
      "grad_norm": 0.1709097921848297,
      "learning_rate": 2.9486540226488556e-07,
      "loss": 0.958,
      "step": 260
    },
    {
      "epoch": 9.642857142857142,
      "grad_norm": 0.22939978539943695,
      "learning_rate": 8.119475099673035e-08,
      "loss": 1.385,
      "step": 270
    },
    {
      "epoch": 9.642857142857142,
      "eval_loss": 0.9539986252784729,
      "eval_runtime": 334.0911,
      "eval_samples_per_second": 0.069,
      "eval_steps_per_second": 0.036,
      "step": 270
    },
    {
      "epoch": 10.0,
      "grad_norm": 0.12558279931545258,
      "learning_rate": 6.719335161364804e-10,
      "loss": 0.917,
      "step": 280
    },
    {
      "epoch": 10.0,
      "step": 280,
      "total_flos": 1.6868978137340314e+18,
      "train_loss": 1.1517673764910017,
      "train_runtime": 34809.0643,
      "train_samples_per_second": 0.064,
      "train_steps_per_second": 0.008,
      "train_tokens_per_second": 527.164
    },
    {
      "epoch": 10.0,
      "eval_loss": 0.9539986252784729,
      "eval_runtime": 337.0962,
      "eval_samples_per_second": 0.068,
      "eval_steps_per_second": 0.036,
      "step": 280
    }
  ],
  "logging_steps": 10,
  "max_steps": 280,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 30,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 2,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 0
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.6868978137340314e+18,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}