{
  "best_global_step": 625,
  "best_metric": 0.6392784657268284,
  "best_model_checkpoint": "/home/rngo/code/liquidai-hackathon/train/checkpoints/v2_no_coords/checkpoint-625",
  "epoch": 5.0,
  "eval_steps": 100,
  "global_step": 625,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.08032128514056225,
      "grad_norm": 7.125,
      "learning_rate": 5.625e-06,
      "loss": 0.23156580924987794,
      "step": 10
    },
    {
      "epoch": 0.1606425702811245,
      "grad_norm": 2.671875,
      "learning_rate": 1.1875e-05,
      "loss": 0.10550506114959717,
      "step": 20
    },
    {
      "epoch": 0.24096385542168675,
      "grad_norm": 6.03125,
      "learning_rate": 1.8125e-05,
      "loss": 0.07342656850814819,
      "step": 30
    },
    {
      "epoch": 0.321285140562249,
      "grad_norm": 4.5,
      "learning_rate": 1.9993124462265045e-05,
      "loss": 0.05777645707130432,
      "step": 40
    },
    {
      "epoch": 0.40160642570281124,
      "grad_norm": 7.96875,
      "learning_rate": 1.995947111744728e-05,
      "loss": 0.05417297482490539,
      "step": 50
    },
    {
      "epoch": 0.4819277108433735,
      "grad_norm": 1.71875,
      "learning_rate": 1.9897871425050598e-05,
      "loss": 0.05731180906295776,
      "step": 60
    },
    {
      "epoch": 0.5622489959839357,
      "grad_norm": 2.46875,
      "learning_rate": 1.980849823425486e-05,
      "loss": 0.04996185302734375,
      "step": 70
    },
    {
      "epoch": 0.642570281124498,
      "grad_norm": 1.2109375,
      "learning_rate": 1.969160232687616e-05,
      "loss": 0.059390270709991456,
      "step": 80
    },
    {
      "epoch": 0.7228915662650602,
      "grad_norm": 6.6875,
      "learning_rate": 1.9547511713671264e-05,
      "loss": 0.04534949958324432,
      "step": 90
    },
    {
      "epoch": 0.8032128514056225,
      "grad_norm": 3.8125,
      "learning_rate": 1.9376630713937043e-05,
      "loss": 0.05621292591094971,
      "step": 100
    },
    {
      "epoch": 0.8835341365461847,
      "grad_norm": 1.96875,
      "learning_rate": 1.9179438820987645e-05,
      "loss": 0.05151752233505249,
      "step": 110
    },
    {
      "epoch": 0.963855421686747,
      "grad_norm": 2.28125,
      "learning_rate": 1.895648935669278e-05,
      "loss": 0.04378549158573151,
      "step": 120
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6263157894736842,
      "eval_loss": 0.04093950241804123,
      "eval_macro_f1": 0.48621561390776613,
      "eval_parse_fail_rate": 0.3736842105263158,
      "eval_runtime": 7.4217,
      "eval_samples_per_second": 102.402,
      "eval_steps_per_second": 6.467,
      "step": 125
    },
    {
      "epoch": 1.0401606425702812,
      "grad_norm": 5.21875,
      "learning_rate": 1.8708407918852608e-05,
      "loss": 0.04595450460910797,
      "step": 130
    },
    {
      "epoch": 1.1204819277108433,
      "grad_norm": 2.5625,
      "learning_rate": 1.8435890625765776e-05,
      "loss": 0.03442436754703522,
      "step": 140
    },
    {
      "epoch": 1.2008032128514057,
      "grad_norm": 3.65625,
      "learning_rate": 1.8139702162916485e-05,
      "loss": 0.0374337911605835,
      "step": 150
    },
    {
      "epoch": 1.2811244979919678,
      "grad_norm": 4.3125,
      "learning_rate": 1.782067363726153e-05,
      "loss": 0.034351897239685056,
      "step": 160
    },
    {
      "epoch": 1.3614457831325302,
      "grad_norm": 1.9609375,
      "learning_rate": 1.7479700245138184e-05,
      "loss": 0.030325621366500854,
      "step": 170
    },
    {
      "epoch": 1.4417670682730923,
      "grad_norm": 3.234375,
      "learning_rate": 1.7117738760336846e-05,
      "loss": 0.03940878212451935,
      "step": 180
    },
    {
      "epoch": 1.5220883534136547,
      "grad_norm": 2.84375,
      "learning_rate": 1.6735804849386914e-05,
      "loss": 0.031922188401222226,
      "step": 190
    },
    {
      "epoch": 1.6024096385542168,
      "grad_norm": 1.5078125,
      "learning_rate": 1.6334970221589182e-05,
      "loss": 0.0363187164068222,
      "step": 200
    },
    {
      "epoch": 1.6827309236947792,
      "grad_norm": 2.734375,
      "learning_rate": 1.5916359621791847e-05,
      "loss": 0.02404191195964813,
      "step": 210
    },
    {
      "epoch": 1.7630522088353415,
      "grad_norm": 4.53125,
      "learning_rate": 1.5481147674348366e-05,
      "loss": 0.03141278922557831,
      "step": 220
    },
    {
      "epoch": 1.8433734939759037,
      "grad_norm": 2.265625,
      "learning_rate": 1.5030555587113091e-05,
      "loss": 0.03363974094390869,
      "step": 230
    },
    {
      "epoch": 1.9236947791164658,
      "grad_norm": 1.2734375,
      "learning_rate": 1.4565847724723225e-05,
      "loss": 0.02886805832386017,
      "step": 240
    },
    {
      "epoch": 2.0,
      "grad_norm": 2.234375,
      "learning_rate": 1.4088328060782573e-05,
      "loss": 0.031261670589447024,
      "step": 250
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7657894736842106,
      "eval_loss": 0.026152363047003746,
      "eval_macro_f1": 0.6361528064663962,
      "eval_parse_fail_rate": 0.23421052631578948,
      "eval_runtime": 7.4084,
      "eval_samples_per_second": 102.587,
      "eval_steps_per_second": 6.479,
      "step": 250
    },
    {
      "epoch": 2.0803212851405624,
      "grad_norm": 7.375,
      "learning_rate": 1.3599336518902228e-05,
      "loss": 0.027535757422447203,
      "step": 260
    },
    {
      "epoch": 2.1606425702811247,
      "grad_norm": 1.7578125,
      "learning_rate": 1.3100245212865279e-05,
      "loss": 0.017315882444381713,
      "step": 270
    },
    {
      "epoch": 2.2409638554216866,
      "grad_norm": 2.140625,
      "learning_rate": 1.259245459646567e-05,
      "loss": 0.02335720956325531,
      "step": 280
    },
    {
      "epoch": 2.321285140562249,
      "grad_norm": 2.265625,
      "learning_rate": 1.2077389533824789e-05,
      "loss": 0.016771413385868073,
      "step": 290
    },
    {
      "epoch": 2.4016064257028114,
      "grad_norm": 2.46875,
      "learning_rate": 1.1556495301212485e-05,
      "loss": 0.016400144994258882,
      "step": 300
    },
    {
      "epoch": 2.4819277108433733,
      "grad_norm": 1.7734375,
      "learning_rate": 1.1031233531591471e-05,
      "loss": 0.019619688391685486,
      "step": 310
    },
    {
      "epoch": 2.5622489959839356,
      "grad_norm": 0.89453125,
      "learning_rate": 1.0503078113264715e-05,
      "loss": 0.016282124817371367,
      "step": 320
    },
    {
      "epoch": 2.642570281124498,
      "grad_norm": 9.625,
      "learning_rate": 9.973511054134259e-06,
      "loss": 0.023008912801742554,
      "step": 330
    },
    {
      "epoch": 2.7228915662650603,
      "grad_norm": 1.6796875,
      "learning_rate": 9.444018323176399e-06,
      "loss": 0.019158574938774108,
      "step": 340
    },
    {
      "epoch": 2.8032128514056227,
      "grad_norm": 4.15625,
      "learning_rate": 8.916085680802038e-06,
      "loss": 0.01898370832204819,
      "step": 350
    },
    {
      "epoch": 2.8835341365461846,
      "grad_norm": 2.328125,
      "learning_rate": 8.391194509802294e-06,
      "loss": 0.021644172072410584,
      "step": 360
    },
    {
      "epoch": 2.963855421686747,
      "grad_norm": 3.03125,
      "learning_rate": 7.870817658577743e-06,
      "loss": 0.02495715022087097,
      "step": 370
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8355263157894737,
      "eval_loss": 0.02834087610244751,
      "eval_macro_f1": 0.6384880909700965,
      "eval_parse_fail_rate": 0.16447368421052633,
      "eval_runtime": 7.4445,
      "eval_samples_per_second": 102.089,
      "eval_steps_per_second": 6.448,
      "step": 375
    },
    {
      "epoch": 3.040160642570281,
      "grad_norm": 5.15625,
      "learning_rate": 7.356415308315201e-06,
      "loss": 0.010920841991901398,
      "step": 380
    },
    {
      "epoch": 3.1204819277108435,
      "grad_norm": 5.5625,
      "learning_rate": 6.849430875708818e-06,
      "loss": 0.0173117995262146,
      "step": 390
    },
    {
      "epoch": 3.2008032128514055,
      "grad_norm": 0.92578125,
      "learning_rate": 6.3512869627224535e-06,
      "loss": 0.011250025033950806,
      "step": 400
    },
    {
      "epoch": 3.281124497991968,
      "grad_norm": 2.859375,
      "learning_rate": 5.8633813647583505e-06,
      "loss": 0.008569899946451187,
      "step": 410
    },
    {
      "epoch": 3.36144578313253,
      "grad_norm": 11.1875,
      "learning_rate": 5.38708314843315e-06,
      "loss": 0.016421397030353547,
      "step": 420
    },
    {
      "epoch": 3.4417670682730925,
      "grad_norm": 5.5,
      "learning_rate": 4.923728809967156e-06,
      "loss": 0.02297772616147995,
      "step": 430
    },
    {
      "epoch": 3.522088353413655,
      "grad_norm": 4.3125,
      "learning_rate": 4.474618524966313e-06,
      "loss": 0.008117502182722091,
      "step": 440
    },
    {
      "epoch": 3.602409638554217,
      "grad_norm": 5.0625,
      "learning_rate": 4.04101250012009e-06,
      "loss": 0.015187962353229523,
      "step": 450
    },
    {
      "epoch": 3.682730923694779,
      "grad_norm": 0.74609375,
      "learning_rate": 3.624127437052484e-06,
      "loss": 0.010340742766857147,
      "step": 460
    },
    {
      "epoch": 3.7630522088353415,
      "grad_norm": 6.0625,
      "learning_rate": 3.2251331182484868e-06,
      "loss": 0.019126126170158388,
      "step": 470
    },
    {
      "epoch": 3.8433734939759034,
      "grad_norm": 5.8125,
      "learning_rate": 2.845149124636014e-06,
      "loss": 0.012674199044704437,
      "step": 480
    },
    {
      "epoch": 3.923694779116466,
      "grad_norm": 2.78125,
      "learning_rate": 2.485241694033793e-06,
      "loss": 0.011451072990894318,
      "step": 490
    },
    {
      "epoch": 4.0,
      "grad_norm": 3.28125,
      "learning_rate": 2.1464207292803696e-06,
      "loss": 0.007020576298236847,
      "step": 500
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8355263157894737,
      "eval_loss": 0.03696506470441818,
      "eval_macro_f1": 0.633622550289217,
      "eval_parse_fail_rate": 0.16447368421052633,
      "eval_runtime": 7.4674,
      "eval_samples_per_second": 101.776,
      "eval_steps_per_second": 6.428,
      "step": 500
    },
    {
      "epoch": 4.080321285140562,
      "grad_norm": 3.890625,
      "learning_rate": 1.8296369644394562e-06,
      "loss": 0.009500738978385926,
      "step": 510
    },
    {
      "epoch": 4.160642570281125,
      "grad_norm": 5.03125,
      "learning_rate": 1.535779297033344e-06,
      "loss": 0.004759049043059349,
      "step": 520
    },
    {
      "epoch": 4.240963855421687,
      "grad_norm": 4.40625,
      "learning_rate": 1.2656722937900534e-06,
      "loss": 0.005586640909314156,
      "step": 530
    },
    {
      "epoch": 4.321285140562249,
      "grad_norm": 5.28125,
      "learning_rate": 1.020073876903147e-06,
      "loss": 0.006852471828460693,
      "step": 540
    },
    {
      "epoch": 4.401606425702811,
      "grad_norm": 3.765625,
      "learning_rate": 7.996731972966287e-07,
      "loss": 0.006881513446569443,
      "step": 550
    },
    {
      "epoch": 4.481927710843373,
      "grad_norm": 0.921875,
      "learning_rate": 6.050887008624817e-07,
      "loss": 0.012200018018484115,
      "step": 560
    },
    {
      "epoch": 4.562248995983936,
      "grad_norm": 1.359375,
      "learning_rate": 4.3686639309705183e-07,
      "loss": 0.011706339567899704,
      "step": 570
    },
    {
      "epoch": 4.642570281124498,
      "grad_norm": 3.328125,
      "learning_rate": 2.9547830700569545e-07,
      "loss": 0.00998671054840088,
      "step": 580
    },
    {
      "epoch": 4.72289156626506,
      "grad_norm": 1.125,
      "learning_rate": 1.8132117857477593e-07,
      "loss": 0.01855793744325638,
      "step": 590
    },
    {
      "epoch": 4.803212851405623,
      "grad_norm": 4.875,
      "learning_rate": 9.471533352762962e-08,
      "loss": 0.006045354157686233,
      "step": 600
    },
    {
      "epoch": 4.883534136546185,
      "grad_norm": 7.78125,
      "learning_rate": 3.59037884882818e-08,
      "loss": 0.013950885832309723,
      "step": 610
    },
    {
      "epoch": 4.9638554216867465,
      "grad_norm": 4.09375,
      "learning_rate": 5.051569075056328e-09,
      "loss": 0.006919116526842117,
      "step": 620
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.85,
      "eval_loss": 0.03865349292755127,
      "eval_macro_f1": 0.6392784657268284,
      "eval_parse_fail_rate": 0.15,
      "eval_runtime": 7.6875,
      "eval_samples_per_second": 98.862,
      "eval_steps_per_second": 6.244,
      "step": 625
    },
    {
      "epoch": 5.0,
      "step": 625,
      "total_flos": 1.2910681058328576e+16,
      "train_loss": 0.02993413337469101,
      "train_runtime": 499.7178,
      "train_samples_per_second": 39.752,
      "train_steps_per_second": 1.251
    }
  ],
  "logging_steps": 10,
  "max_steps": 625,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "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": 1.2910681058328576e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}