{
  "best_metric": 0.7841530054644809,
  "best_model_checkpoint": "./save_phi2_ft_lora/checkpoint-300",
  "epoch": 0.8571428571428571,
  "eval_steps": 100,
  "global_step": 300,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.02857142857142857,
      "grad_norm": 0.18488438427448273,
      "learning_rate": 5e-06,
      "loss": 1.9926,
      "step": 10
    },
    {
      "epoch": 0.05714285714285714,
      "grad_norm": 0.20217478275299072,
      "learning_rate": 1e-05,
      "loss": 2.0428,
      "step": 20
    },
    {
      "epoch": 0.08571428571428572,
      "grad_norm": 0.22868461906909943,
      "learning_rate": 1.5e-05,
      "loss": 2.1981,
      "step": 30
    },
    {
      "epoch": 0.11428571428571428,
      "grad_norm": 0.31630581617355347,
      "learning_rate": 2e-05,
      "loss": 2.1773,
      "step": 40
    },
    {
      "epoch": 0.14285714285714285,
      "grad_norm": 0.46189257502555847,
      "learning_rate": 2.5e-05,
      "loss": 2.0399,
      "step": 50
    },
    {
      "epoch": 0.17142857142857143,
      "grad_norm": 1.0216264724731445,
      "learning_rate": 3e-05,
      "loss": 2.1152,
      "step": 60
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.6140756607055664,
      "learning_rate": 3.5e-05,
      "loss": 2.0064,
      "step": 70
    },
    {
      "epoch": 0.22857142857142856,
      "grad_norm": 0.7276609539985657,
      "learning_rate": 4e-05,
      "loss": 1.9443,
      "step": 80
    },
    {
      "epoch": 0.2571428571428571,
      "grad_norm": 0.694882869720459,
      "learning_rate": 4.5e-05,
      "loss": 1.9579,
      "step": 90
    },
    {
      "epoch": 0.2857142857142857,
      "grad_norm": 0.7517616748809814,
      "learning_rate": 5e-05,
      "loss": 1.9891,
      "step": 100
    },
    {
      "epoch": 0.2857142857142857,
      "eval_accuracy": 0.7295081967213115,
      "step": 100
    },
    {
      "epoch": 0.2857142857142857,
      "eval_loss": 1.9115574359893799,
      "eval_runtime": 333.2149,
      "eval_samples_per_second": 1.098,
      "eval_steps_per_second": 1.098,
      "step": 100
    },
    {
      "epoch": 0.3142857142857143,
      "grad_norm": 0.8051323294639587,
      "learning_rate": 4.9166666666666665e-05,
      "loss": 1.8878,
      "step": 110
    },
    {
      "epoch": 0.34285714285714286,
      "grad_norm": 0.8047056198120117,
      "learning_rate": 4.8333333333333334e-05,
      "loss": 1.9324,
      "step": 120
    },
    {
      "epoch": 0.37142857142857144,
      "grad_norm": 0.7180447578430176,
      "learning_rate": 4.75e-05,
      "loss": 1.939,
      "step": 130
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.8593822717666626,
      "learning_rate": 4.666666666666667e-05,
      "loss": 1.821,
      "step": 140
    },
    {
      "epoch": 0.42857142857142855,
      "grad_norm": 0.837947428226471,
      "learning_rate": 4.5833333333333334e-05,
      "loss": 1.8796,
      "step": 150
    },
    {
      "epoch": 0.45714285714285713,
      "grad_norm": 0.8277564644813538,
      "learning_rate": 4.5e-05,
      "loss": 1.8104,
      "step": 160
    },
    {
      "epoch": 0.4857142857142857,
      "grad_norm": 0.8004774451255798,
      "learning_rate": 4.4166666666666665e-05,
      "loss": 1.9447,
      "step": 170
    },
    {
      "epoch": 0.5142857142857142,
      "grad_norm": 0.907889187335968,
      "learning_rate": 4.3333333333333334e-05,
      "loss": 1.9153,
      "step": 180
    },
    {
      "epoch": 0.5428571428571428,
      "grad_norm": 0.8122074604034424,
      "learning_rate": 4.25e-05,
      "loss": 1.8258,
      "step": 190
    },
    {
      "epoch": 0.5714285714285714,
      "grad_norm": 0.8643149137496948,
      "learning_rate": 4.166666666666667e-05,
      "loss": 1.9096,
      "step": 200
    },
    {
      "epoch": 0.5714285714285714,
      "eval_accuracy": 0.7759562841530054,
      "step": 200
    },
    {
      "epoch": 0.5714285714285714,
      "eval_loss": 1.8283634185791016,
      "eval_runtime": 333.2589,
      "eval_samples_per_second": 1.098,
      "eval_steps_per_second": 1.098,
      "step": 200
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.7891345024108887,
      "learning_rate": 4.0833333333333334e-05,
      "loss": 1.8198,
      "step": 210
    },
    {
      "epoch": 0.6285714285714286,
      "grad_norm": 0.9142560362815857,
      "learning_rate": 4e-05,
      "loss": 1.822,
      "step": 220
    },
    {
      "epoch": 0.6571428571428571,
      "grad_norm": 0.8956629037857056,
      "learning_rate": 3.9166666666666665e-05,
      "loss": 1.9947,
      "step": 230
    },
    {
      "epoch": 0.6857142857142857,
      "grad_norm": 0.8790289163589478,
      "learning_rate": 3.8333333333333334e-05,
      "loss": 1.8011,
      "step": 240
    },
    {
      "epoch": 0.7142857142857143,
      "grad_norm": 0.7958552837371826,
      "learning_rate": 3.7500000000000003e-05,
      "loss": 1.8926,
      "step": 250
    },
    {
      "epoch": 0.7428571428571429,
      "grad_norm": 0.924050509929657,
      "learning_rate": 3.6666666666666666e-05,
      "loss": 1.7819,
      "step": 260
    },
    {
      "epoch": 0.7714285714285715,
      "grad_norm": 0.8796949982643127,
      "learning_rate": 3.5833333333333335e-05,
      "loss": 1.9019,
      "step": 270
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.886051595211029,
      "learning_rate": 3.5e-05,
      "loss": 1.9703,
      "step": 280
    },
    {
      "epoch": 0.8285714285714286,
      "grad_norm": 0.927409291267395,
      "learning_rate": 3.4166666666666666e-05,
      "loss": 1.7759,
      "step": 290
    },
    {
      "epoch": 0.8571428571428571,
      "grad_norm": 0.9948813915252686,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 1.8922,
      "step": 300
    },
    {
      "epoch": 0.8571428571428571,
      "eval_accuracy": 0.7841530054644809,
      "step": 300
    },
    {
      "epoch": 0.8571428571428571,
      "eval_loss": 1.7880818843841553,
      "eval_runtime": 333.1304,
      "eval_samples_per_second": 1.099,
      "eval_steps_per_second": 1.099,
      "step": 300
    }
  ],
  "logging_steps": 10,
  "max_steps": 700,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 3.9441917804544e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}