{
  "best_global_step": 237,
  "best_metric": 0.9935535559662932,
  "best_model_checkpoint": "ckpt/checkpoint-237",
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 237,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.08917197452229299,
      "grad_norm": 36.30582809448242,
      "learning_rate": 5.2173913043478265e-06,
      "loss": 1.1125,
      "step": 7
    },
    {
      "epoch": 0.17834394904458598,
      "grad_norm": 35.81989669799805,
      "learning_rate": 1.1304347826086957e-05,
      "loss": 0.8511,
      "step": 14
    },
    {
      "epoch": 0.267515923566879,
      "grad_norm": 6.439170837402344,
      "learning_rate": 1.739130434782609e-05,
      "loss": 0.6037,
      "step": 21
    },
    {
      "epoch": 0.35668789808917195,
      "grad_norm": 3.1305227279663086,
      "learning_rate": 1.9626168224299065e-05,
      "loss": 0.3597,
      "step": 28
    },
    {
      "epoch": 0.445859872611465,
      "grad_norm": 4.9628095626831055,
      "learning_rate": 1.8971962616822433e-05,
      "loss": 0.2122,
      "step": 35
    },
    {
      "epoch": 0.535031847133758,
      "grad_norm": 3.5573127269744873,
      "learning_rate": 1.8317757009345797e-05,
      "loss": 0.1176,
      "step": 42
    },
    {
      "epoch": 0.6242038216560509,
      "grad_norm": 2.0348973274230957,
      "learning_rate": 1.766355140186916e-05,
      "loss": 0.0712,
      "step": 49
    },
    {
      "epoch": 0.7133757961783439,
      "grad_norm": 6.060424327850342,
      "learning_rate": 1.7009345794392526e-05,
      "loss": 0.0753,
      "step": 56
    },
    {
      "epoch": 0.802547770700637,
      "grad_norm": 3.5385732650756836,
      "learning_rate": 1.635514018691589e-05,
      "loss": 0.0575,
      "step": 63
    },
    {
      "epoch": 0.89171974522293,
      "grad_norm": 2.5258700847625732,
      "learning_rate": 1.5700934579439254e-05,
      "loss": 0.039,
      "step": 70
    },
    {
      "epoch": 0.9808917197452229,
      "grad_norm": 0.7659706473350525,
      "learning_rate": 1.5046728971962619e-05,
      "loss": 0.0656,
      "step": 77
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9894872195867869,
      "eval_f1": 0.9894880819914448,
      "eval_loss": 0.03790666535496712,
      "eval_runtime": 2530.3213,
      "eval_samples_per_second": 22.744,
      "eval_steps_per_second": 0.356,
      "step": 79
    },
    {
      "epoch": 1.0636942675159236,
      "grad_norm": 0.6808683276176453,
      "learning_rate": 1.4392523364485981e-05,
      "loss": 0.0528,
      "step": 84
    },
    {
      "epoch": 1.1528662420382165,
      "grad_norm": 26.949949264526367,
      "learning_rate": 1.3738317757009347e-05,
      "loss": 0.099,
      "step": 91
    },
    {
      "epoch": 1.2420382165605095,
      "grad_norm": 1.5863674879074097,
      "learning_rate": 1.308411214953271e-05,
      "loss": 0.0336,
      "step": 98
    },
    {
      "epoch": 1.3312101910828025,
      "grad_norm": 2.183351755142212,
      "learning_rate": 1.2429906542056076e-05,
      "loss": 0.0173,
      "step": 105
    },
    {
      "epoch": 1.4203821656050954,
      "grad_norm": 2.4681448936462402,
      "learning_rate": 1.177570093457944e-05,
      "loss": 0.017,
      "step": 112
    },
    {
      "epoch": 1.5095541401273884,
      "grad_norm": 6.7833476066589355,
      "learning_rate": 1.1121495327102804e-05,
      "loss": 0.0249,
      "step": 119
    },
    {
      "epoch": 1.5987261146496814,
      "grad_norm": 0.06058590114116669,
      "learning_rate": 1.0467289719626168e-05,
      "loss": 0.0244,
      "step": 126
    },
    {
      "epoch": 1.6878980891719744,
      "grad_norm": 0.17764343321323395,
      "learning_rate": 9.813084112149533e-06,
      "loss": 0.0357,
      "step": 133
    },
    {
      "epoch": 1.7770700636942676,
      "grad_norm": 12.890459060668945,
      "learning_rate": 9.158878504672899e-06,
      "loss": 0.0439,
      "step": 140
    },
    {
      "epoch": 1.8662420382165605,
      "grad_norm": 7.447948455810547,
      "learning_rate": 8.504672897196263e-06,
      "loss": 0.0415,
      "step": 147
    },
    {
      "epoch": 1.9554140127388535,
      "grad_norm": 0.40099361538887024,
      "learning_rate": 7.850467289719627e-06,
      "loss": 0.0141,
      "step": 154
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9897478670350484,
      "eval_f1": 0.9897461756211305,
      "eval_loss": 0.03573331609368324,
      "eval_runtime": 2458.8654,
      "eval_samples_per_second": 23.405,
      "eval_steps_per_second": 0.366,
      "step": 158
    },
    {
      "epoch": 2.038216560509554,
      "grad_norm": 0.18038131296634674,
      "learning_rate": 7.196261682242991e-06,
      "loss": 0.0107,
      "step": 161
    },
    {
      "epoch": 2.127388535031847,
      "grad_norm": 8.816960334777832,
      "learning_rate": 6.542056074766355e-06,
      "loss": 0.0197,
      "step": 168
    },
    {
      "epoch": 2.21656050955414,
      "grad_norm": 0.17330680787563324,
      "learning_rate": 5.88785046728972e-06,
      "loss": 0.0492,
      "step": 175
    },
    {
      "epoch": 2.305732484076433,
      "grad_norm": 0.10762328654527664,
      "learning_rate": 5.233644859813084e-06,
      "loss": 0.0223,
      "step": 182
    },
    {
      "epoch": 2.394904458598726,
      "grad_norm": 0.06711781769990921,
      "learning_rate": 4.579439252336449e-06,
      "loss": 0.0263,
      "step": 189
    },
    {
      "epoch": 2.484076433121019,
      "grad_norm": 0.7490471601486206,
      "learning_rate": 3.925233644859814e-06,
      "loss": 0.004,
      "step": 196
    },
    {
      "epoch": 2.573248407643312,
      "grad_norm": 0.060274120420217514,
      "learning_rate": 3.2710280373831774e-06,
      "loss": 0.0124,
      "step": 203
    },
    {
      "epoch": 2.662420382165605,
      "grad_norm": 12.112130165100098,
      "learning_rate": 2.616822429906542e-06,
      "loss": 0.013,
      "step": 210
    },
    {
      "epoch": 2.7515923566878984,
      "grad_norm": 0.036105964332818985,
      "learning_rate": 1.962616822429907e-06,
      "loss": 0.0034,
      "step": 217
    },
    {
      "epoch": 2.840764331210191,
      "grad_norm": 0.03519793599843979,
      "learning_rate": 1.308411214953271e-06,
      "loss": 0.0106,
      "step": 224
    },
    {
      "epoch": 2.9299363057324843,
      "grad_norm": 0.6189459562301636,
      "learning_rate": 6.542056074766355e-07,
      "loss": 0.0095,
      "step": 231
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.993553319779666,
      "eval_f1": 0.9935535559662932,
      "eval_loss": 0.0235657449811697,
      "eval_runtime": 2541.1299,
      "eval_samples_per_second": 22.647,
      "eval_steps_per_second": 0.354,
      "step": 237
    }
  ],
  "logging_steps": 7,
  "max_steps": 237,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1973350632960000.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}