{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 28.0,
  "eval_steps": 500,
  "global_step": 14812,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.19,
      "learning_rate": 5e-05,
      "loss": 1.6355,
      "step": 100
    },
    {
      "epoch": 0.38,
      "learning_rate": 5e-05,
      "loss": 1.4037,
      "step": 200
    },
    {
      "epoch": 0.57,
      "learning_rate": 5e-05,
      "loss": 1.4183,
      "step": 300
    },
    {
      "epoch": 0.76,
      "learning_rate": 5e-05,
      "loss": 1.3881,
      "step": 400
    },
    {
      "epoch": 0.95,
      "learning_rate": 5e-05,
      "loss": 1.3891,
      "step": 500
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6137948717948718,
      "eval_loss": 1.3015451431274414,
      "eval_runtime": 4.1821,
      "eval_samples_per_second": 119.557,
      "eval_steps_per_second": 15.064,
      "step": 529
    },
    {
      "epoch": 1.0,
      "eval_exact_match": 23.2,
      "eval_f1": 35.17014759120022,
      "step": 529
    },
    {
      "epoch": 1.13,
      "learning_rate": 5e-05,
      "loss": 1.3808,
      "step": 600
    },
    {
      "epoch": 1.32,
      "learning_rate": 5e-05,
      "loss": 1.348,
      "step": 700
    },
    {
      "epoch": 1.51,
      "learning_rate": 5e-05,
      "loss": 1.3627,
      "step": 800
    },
    {
      "epoch": 1.7,
      "learning_rate": 5e-05,
      "loss": 1.3473,
      "step": 900
    },
    {
      "epoch": 1.89,
      "learning_rate": 5e-05,
      "loss": 1.3633,
      "step": 1000
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6165641025641025,
      "eval_loss": 1.2854759693145752,
      "eval_runtime": 4.1064,
      "eval_samples_per_second": 121.761,
      "eval_steps_per_second": 15.342,
      "step": 1058
    },
    {
      "epoch": 2.0,
      "eval_exact_match": 25.8,
      "eval_f1": 37.509047619047635,
      "step": 1058
    },
    {
      "epoch": 2.08,
      "learning_rate": 5e-05,
      "loss": 1.3358,
      "step": 1100
    },
    {
      "epoch": 2.27,
      "learning_rate": 5e-05,
      "loss": 1.2917,
      "step": 1200
    },
    {
      "epoch": 2.46,
      "learning_rate": 5e-05,
      "loss": 1.2948,
      "step": 1300
    },
    {
      "epoch": 2.65,
      "learning_rate": 5e-05,
      "loss": 1.3144,
      "step": 1400
    },
    {
      "epoch": 2.84,
      "learning_rate": 5e-05,
      "loss": 1.2929,
      "step": 1500
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6176923076923077,
      "eval_loss": 1.2953603267669678,
      "eval_runtime": 4.0992,
      "eval_samples_per_second": 121.974,
      "eval_steps_per_second": 15.369,
      "step": 1587
    },
    {
      "epoch": 3.0,
      "eval_exact_match": 25.0,
      "eval_f1": 37.08253968253969,
      "step": 1587
    },
    {
      "epoch": 3.02,
      "learning_rate": 5e-05,
      "loss": 1.2906,
      "step": 1600
    },
    {
      "epoch": 3.21,
      "learning_rate": 5e-05,
      "loss": 1.2216,
      "step": 1700
    },
    {
      "epoch": 3.4,
      "learning_rate": 5e-05,
      "loss": 1.2459,
      "step": 1800
    },
    {
      "epoch": 3.59,
      "learning_rate": 5e-05,
      "loss": 1.2474,
      "step": 1900
    },
    {
      "epoch": 3.78,
      "learning_rate": 5e-05,
      "loss": 1.2518,
      "step": 2000
    },
    {
      "epoch": 3.97,
      "learning_rate": 5e-05,
      "loss": 1.2361,
      "step": 2100
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6045128205128205,
      "eval_loss": 1.3488503694534302,
      "eval_runtime": 4.2015,
      "eval_samples_per_second": 119.004,
      "eval_steps_per_second": 14.995,
      "step": 2116
    },
    {
      "epoch": 4.0,
      "eval_exact_match": 25.0,
      "eval_f1": 36.82126984126986,
      "step": 2116
    },
    {
      "epoch": 4.16,
      "learning_rate": 5e-05,
      "loss": 1.1856,
      "step": 2200
    },
    {
      "epoch": 4.35,
      "learning_rate": 5e-05,
      "loss": 1.1556,
      "step": 2300
    },
    {
      "epoch": 4.54,
      "learning_rate": 5e-05,
      "loss": 1.1828,
      "step": 2400
    },
    {
      "epoch": 4.73,
      "learning_rate": 5e-05,
      "loss": 1.1687,
      "step": 2500
    },
    {
      "epoch": 4.91,
      "learning_rate": 5e-05,
      "loss": 1.1856,
      "step": 2600
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6125128205128205,
      "eval_loss": 1.3968186378479004,
      "eval_runtime": 4.3081,
      "eval_samples_per_second": 116.06,
      "eval_steps_per_second": 14.624,
      "step": 2645
    },
    {
      "epoch": 5.0,
      "eval_exact_match": 24.6,
      "eval_f1": 37.079523809523806,
      "step": 2645
    },
    {
      "epoch": 5.1,
      "learning_rate": 5e-05,
      "loss": 1.1376,
      "step": 2700
    },
    {
      "epoch": 5.29,
      "learning_rate": 5e-05,
      "loss": 1.1079,
      "step": 2800
    },
    {
      "epoch": 5.48,
      "learning_rate": 5e-05,
      "loss": 1.0954,
      "step": 2900
    },
    {
      "epoch": 5.67,
      "learning_rate": 5e-05,
      "loss": 1.1317,
      "step": 3000
    },
    {
      "epoch": 5.86,
      "learning_rate": 5e-05,
      "loss": 1.1098,
      "step": 3100
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.6114871794871795,
      "eval_loss": 1.4720605611801147,
      "eval_runtime": 4.5,
      "eval_samples_per_second": 111.11,
      "eval_steps_per_second": 14.0,
      "step": 3174
    },
    {
      "epoch": 6.0,
      "eval_exact_match": 24.4,
      "eval_f1": 35.78015873015873,
      "step": 3174
    },
    {
      "epoch": 6.05,
      "learning_rate": 5e-05,
      "loss": 1.0945,
      "step": 3200
    },
    {
      "epoch": 6.24,
      "learning_rate": 5e-05,
      "loss": 1.0309,
      "step": 3300
    },
    {
      "epoch": 6.43,
      "learning_rate": 5e-05,
      "loss": 1.0389,
      "step": 3400
    },
    {
      "epoch": 6.62,
      "learning_rate": 5e-05,
      "loss": 1.0547,
      "step": 3500
    },
    {
      "epoch": 6.81,
      "learning_rate": 5e-05,
      "loss": 1.0618,
      "step": 3600
    },
    {
      "epoch": 6.99,
      "learning_rate": 5e-05,
      "loss": 1.0753,
      "step": 3700
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.6076410256410256,
      "eval_loss": 1.5797967910766602,
      "eval_runtime": 4.8128,
      "eval_samples_per_second": 103.891,
      "eval_steps_per_second": 13.09,
      "step": 3703
    },
    {
      "epoch": 7.0,
      "eval_exact_match": 21.8,
      "eval_f1": 33.15111111111111,
      "step": 3703
    },
    {
      "epoch": 7.18,
      "learning_rate": 5e-05,
      "loss": 0.9727,
      "step": 3800
    },
    {
      "epoch": 7.37,
      "learning_rate": 5e-05,
      "loss": 0.9815,
      "step": 3900
    },
    {
      "epoch": 7.56,
      "learning_rate": 5e-05,
      "loss": 0.9777,
      "step": 4000
    },
    {
      "epoch": 7.75,
      "learning_rate": 5e-05,
      "loss": 0.9984,
      "step": 4100
    },
    {
      "epoch": 7.94,
      "learning_rate": 5e-05,
      "loss": 1.0048,
      "step": 4200
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.6084102564102564,
      "eval_loss": 1.6041829586029053,
      "eval_runtime": 4.4087,
      "eval_samples_per_second": 113.411,
      "eval_steps_per_second": 14.29,
      "step": 4232
    },
    {
      "epoch": 8.0,
      "eval_exact_match": 22.4,
      "eval_f1": 35.09904761904763,
      "step": 4232
    },
    {
      "epoch": 8.13,
      "learning_rate": 5e-05,
      "loss": 0.9568,
      "step": 4300
    },
    {
      "epoch": 8.32,
      "learning_rate": 5e-05,
      "loss": 0.9088,
      "step": 4400
    },
    {
      "epoch": 8.51,
      "learning_rate": 5e-05,
      "loss": 0.9341,
      "step": 4500
    },
    {
      "epoch": 8.7,
      "learning_rate": 5e-05,
      "loss": 0.9285,
      "step": 4600
    },
    {
      "epoch": 8.88,
      "learning_rate": 5e-05,
      "loss": 0.9456,
      "step": 4700
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.5976923076923077,
      "eval_loss": 1.684310793876648,
      "eval_runtime": 4.9249,
      "eval_samples_per_second": 101.524,
      "eval_steps_per_second": 12.792,
      "step": 4761
    },
    {
      "epoch": 9.0,
      "eval_exact_match": 24.6,
      "eval_f1": 36.32493506493508,
      "step": 4761
    },
    {
      "epoch": 9.07,
      "learning_rate": 5e-05,
      "loss": 0.912,
      "step": 4800
    },
    {
      "epoch": 9.26,
      "learning_rate": 5e-05,
      "loss": 0.8443,
      "step": 4900
    },
    {
      "epoch": 9.45,
      "learning_rate": 5e-05,
      "loss": 0.8789,
      "step": 5000
    },
    {
      "epoch": 9.64,
      "learning_rate": 5e-05,
      "loss": 0.8865,
      "step": 5100
    },
    {
      "epoch": 9.83,
      "learning_rate": 5e-05,
      "loss": 0.8766,
      "step": 5200
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.6051282051282051,
      "eval_loss": 1.7828805446624756,
      "eval_runtime": 4.1983,
      "eval_samples_per_second": 119.096,
      "eval_steps_per_second": 15.006,
      "step": 5290
    },
    {
      "epoch": 10.0,
      "eval_exact_match": 23.0,
      "eval_f1": 34.336190476190495,
      "step": 5290
    },
    {
      "epoch": 10.02,
      "learning_rate": 5e-05,
      "loss": 0.8835,
      "step": 5300
    },
    {
      "epoch": 10.21,
      "learning_rate": 5e-05,
      "loss": 0.8095,
      "step": 5400
    },
    {
      "epoch": 10.4,
      "learning_rate": 5e-05,
      "loss": 0.8008,
      "step": 5500
    },
    {
      "epoch": 10.59,
      "learning_rate": 5e-05,
      "loss": 0.8208,
      "step": 5600
    },
    {
      "epoch": 10.78,
      "learning_rate": 5e-05,
      "loss": 0.8282,
      "step": 5700
    },
    {
      "epoch": 10.96,
      "learning_rate": 5e-05,
      "loss": 0.8273,
      "step": 5800
    },
    {
      "epoch": 11.0,
      "eval_accuracy": 0.6043076923076923,
      "eval_loss": 1.8059720993041992,
      "eval_runtime": 4.2264,
      "eval_samples_per_second": 118.304,
      "eval_steps_per_second": 14.906,
      "step": 5819
    },
    {
      "epoch": 11.0,
      "eval_exact_match": 21.6,
      "eval_f1": 32.628095238095234,
      "step": 5819
    },
    {
      "epoch": 11.15,
      "learning_rate": 5e-05,
      "loss": 0.7637,
      "step": 5900
    },
    {
      "epoch": 11.34,
      "learning_rate": 5e-05,
      "loss": 0.7581,
      "step": 6000
    },
    {
      "epoch": 11.53,
      "learning_rate": 5e-05,
      "loss": 0.7571,
      "step": 6100
    },
    {
      "epoch": 11.72,
      "learning_rate": 5e-05,
      "loss": 0.7637,
      "step": 6200
    },
    {
      "epoch": 11.91,
      "learning_rate": 5e-05,
      "loss": 0.7755,
      "step": 6300
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.6018974358974359,
      "eval_loss": 1.8729300498962402,
      "eval_runtime": 4.2001,
      "eval_samples_per_second": 119.045,
      "eval_steps_per_second": 15.0,
      "step": 6348
    },
    {
      "epoch": 12.0,
      "eval_exact_match": 16.8,
      "eval_f1": 27.317351339157547,
      "step": 6348
    },
    {
      "epoch": 12.1,
      "learning_rate": 5e-05,
      "loss": 0.7386,
      "step": 6400
    },
    {
      "epoch": 12.29,
      "learning_rate": 5e-05,
      "loss": 0.6977,
      "step": 6500
    },
    {
      "epoch": 12.48,
      "learning_rate": 5e-05,
      "loss": 0.7185,
      "step": 6600
    },
    {
      "epoch": 12.67,
      "learning_rate": 5e-05,
      "loss": 0.7166,
      "step": 6700
    },
    {
      "epoch": 12.85,
      "learning_rate": 5e-05,
      "loss": 0.715,
      "step": 6800
    },
    {
      "epoch": 13.0,
      "eval_accuracy": 0.6016923076923076,
      "eval_loss": 1.9620097875595093,
      "eval_runtime": 4.4055,
      "eval_samples_per_second": 113.495,
      "eval_steps_per_second": 14.3,
      "step": 6877
    },
    {
      "epoch": 13.0,
      "eval_exact_match": 4.8,
      "eval_f1": 9.356456790950029,
      "step": 6877
    },
    {
      "epoch": 13.04,
      "learning_rate": 5e-05,
      "loss": 0.7032,
      "step": 6900
    },
    {
      "epoch": 13.23,
      "learning_rate": 5e-05,
      "loss": 0.6494,
      "step": 7000
    },
    {
      "epoch": 13.42,
      "learning_rate": 5e-05,
      "loss": 0.6584,
      "step": 7100
    },
    {
      "epoch": 13.61,
      "learning_rate": 5e-05,
      "loss": 0.6681,
      "step": 7200
    },
    {
      "epoch": 13.8,
      "learning_rate": 5e-05,
      "loss": 0.6556,
      "step": 7300
    },
    {
      "epoch": 13.99,
      "learning_rate": 5e-05,
      "loss": 0.6804,
      "step": 7400
    },
    {
      "epoch": 14.0,
      "eval_accuracy": 0.6009230769230769,
      "eval_loss": 2.00297474861145,
      "eval_runtime": 4.2009,
      "eval_samples_per_second": 119.021,
      "eval_steps_per_second": 14.997,
      "step": 7406
    },
    {
      "epoch": 14.0,
      "eval_exact_match": 5.8,
      "eval_f1": 13.996400415302853,
      "step": 7406
    },
    {
      "epoch": 14.18,
      "learning_rate": 5e-05,
      "loss": 0.5983,
      "step": 7500
    },
    {
      "epoch": 14.37,
      "learning_rate": 5e-05,
      "loss": 0.6056,
      "step": 7600
    },
    {
      "epoch": 14.56,
      "learning_rate": 5e-05,
      "loss": 0.6048,
      "step": 7700
    },
    {
      "epoch": 14.74,
      "learning_rate": 5e-05,
      "loss": 0.6322,
      "step": 7800
    },
    {
      "epoch": 14.93,
      "learning_rate": 5e-05,
      "loss": 0.6277,
      "step": 7900
    },
    {
      "epoch": 15.0,
      "eval_accuracy": 0.5998461538461538,
      "eval_loss": 2.0527751445770264,
      "eval_runtime": 4.4287,
      "eval_samples_per_second": 112.9,
      "eval_steps_per_second": 14.225,
      "step": 7935
    },
    {
      "epoch": 15.0,
      "eval_exact_match": 3.2,
      "eval_f1": 8.78350649350649,
      "step": 7935
    },
    {
      "epoch": 15.12,
      "learning_rate": 5e-05,
      "loss": 0.5776,
      "step": 8000
    },
    {
      "epoch": 15.31,
      "learning_rate": 5e-05,
      "loss": 0.5522,
      "step": 8100
    },
    {
      "epoch": 15.5,
      "learning_rate": 5e-05,
      "loss": 0.5735,
      "step": 8200
    },
    {
      "epoch": 15.69,
      "learning_rate": 5e-05,
      "loss": 0.5637,
      "step": 8300
    },
    {
      "epoch": 15.88,
      "learning_rate": 5e-05,
      "loss": 0.5733,
      "step": 8400
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.6012307692307692,
      "eval_loss": 2.047515392303467,
      "eval_runtime": 4.1989,
      "eval_samples_per_second": 119.079,
      "eval_steps_per_second": 15.004,
      "step": 8464
    },
    {
      "epoch": 16.0,
      "eval_exact_match": 0.6,
      "eval_f1": 1.077443805758639,
      "step": 8464
    },
    {
      "epoch": 16.07,
      "learning_rate": 5e-05,
      "loss": 0.5601,
      "step": 8500
    },
    {
      "epoch": 16.26,
      "learning_rate": 5e-05,
      "loss": 0.5126,
      "step": 8600
    },
    {
      "epoch": 16.45,
      "learning_rate": 5e-05,
      "loss": 0.5217,
      "step": 8700
    },
    {
      "epoch": 16.64,
      "learning_rate": 5e-05,
      "loss": 0.5291,
      "step": 8800
    },
    {
      "epoch": 16.82,
      "learning_rate": 5e-05,
      "loss": 0.5409,
      "step": 8900
    },
    {
      "epoch": 17.0,
      "eval_accuracy": 0.5748717948717949,
      "eval_loss": 2.092020273208618,
      "eval_runtime": 4.1044,
      "eval_samples_per_second": 121.822,
      "eval_steps_per_second": 15.35,
      "step": 8993
    },
    {
      "epoch": 17.0,
      "eval_exact_match": 0.4,
      "eval_f1": 2.398489535596529,
      "step": 8993
    },
    {
      "epoch": 17.01,
      "learning_rate": 5e-05,
      "loss": 0.5354,
      "step": 9000
    },
    {
      "epoch": 17.2,
      "learning_rate": 5e-05,
      "loss": 0.4629,
      "step": 9100
    },
    {
      "epoch": 17.39,
      "learning_rate": 5e-05,
      "loss": 0.4795,
      "step": 9200
    },
    {
      "epoch": 17.58,
      "learning_rate": 5e-05,
      "loss": 0.4921,
      "step": 9300
    },
    {
      "epoch": 17.77,
      "learning_rate": 5e-05,
      "loss": 0.4928,
      "step": 9400
    },
    {
      "epoch": 17.96,
      "learning_rate": 5e-05,
      "loss": 0.5024,
      "step": 9500
    },
    {
      "epoch": 18.0,
      "eval_accuracy": 0.5986153846153847,
      "eval_loss": 2.120666980743408,
      "eval_runtime": 4.3114,
      "eval_samples_per_second": 115.973,
      "eval_steps_per_second": 14.613,
      "step": 9522
    },
    {
      "epoch": 18.0,
      "eval_exact_match": 0.2,
      "eval_f1": 0.6343183616587786,
      "step": 9522
    },
    {
      "epoch": 18.15,
      "learning_rate": 5e-05,
      "loss": 0.4539,
      "step": 9600
    },
    {
      "epoch": 18.34,
      "learning_rate": 5e-05,
      "loss": 0.4345,
      "step": 9700
    },
    {
      "epoch": 18.53,
      "learning_rate": 5e-05,
      "loss": 0.4399,
      "step": 9800
    },
    {
      "epoch": 18.71,
      "learning_rate": 5e-05,
      "loss": 0.4581,
      "step": 9900
    },
    {
      "epoch": 18.9,
      "learning_rate": 5e-05,
      "loss": 0.4699,
      "step": 10000
    },
    {
      "epoch": 19.0,
      "eval_accuracy": 0.5992820512820513,
      "eval_loss": 2.110786199569702,
      "eval_runtime": 4.2008,
      "eval_samples_per_second": 119.024,
      "eval_steps_per_second": 14.997,
      "step": 10051
    },
    {
      "epoch": 19.0,
      "eval_exact_match": 0.0,
      "eval_f1": 0.006153846153846154,
      "step": 10051
    },
    {
      "epoch": 19.09,
      "learning_rate": 5e-05,
      "loss": 0.4363,
      "step": 10100
    },
    {
      "epoch": 19.28,
      "learning_rate": 5e-05,
      "loss": 0.4021,
      "step": 10200
    },
    {
      "epoch": 19.47,
      "learning_rate": 5e-05,
      "loss": 0.4161,
      "step": 10300
    },
    {
      "epoch": 19.66,
      "learning_rate": 5e-05,
      "loss": 0.425,
      "step": 10400
    },
    {
      "epoch": 19.85,
      "learning_rate": 5e-05,
      "loss": 0.4367,
      "step": 10500
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.6004615384615385,
      "eval_loss": 2.1088976860046387,
      "eval_runtime": 4.0958,
      "eval_samples_per_second": 122.078,
      "eval_steps_per_second": 15.382,
      "step": 10580
    },
    {
      "epoch": 20.0,
      "eval_exact_match": 0.0,
      "eval_f1": 0.04062256613300257,
      "step": 10580
    },
    {
      "epoch": 20.04,
      "learning_rate": 5e-05,
      "loss": 0.4264,
      "step": 10600
    },
    {
      "epoch": 20.23,
      "learning_rate": 5e-05,
      "loss": 0.3746,
      "step": 10700
    },
    {
      "epoch": 20.42,
      "learning_rate": 5e-05,
      "loss": 0.3834,
      "step": 10800
    },
    {
      "epoch": 20.6,
      "learning_rate": 5e-05,
      "loss": 0.4015,
      "step": 10900
    },
    {
      "epoch": 20.79,
      "learning_rate": 5e-05,
      "loss": 0.4284,
      "step": 11000
    },
    {
      "epoch": 20.98,
      "learning_rate": 5e-05,
      "loss": 0.857,
      "step": 11100
    },
    {
      "epoch": 21.0,
      "eval_accuracy": 0.5983076923076923,
      "eval_loss": 2.0215413570404053,
      "eval_runtime": 4.0942,
      "eval_samples_per_second": 122.125,
      "eval_steps_per_second": 15.388,
      "step": 11109
    },
    {
      "epoch": 21.0,
      "eval_exact_match": 0.0,
      "eval_f1": 0.0,
      "step": 11109
    },
    {
      "epoch": 21.17,
      "learning_rate": 5e-05,
      "loss": 0.8884,
      "step": 11200
    },
    {
      "epoch": 21.36,
      "learning_rate": 5e-05,
      "loss": 1.4123,
      "step": 11300
    },
    {
      "epoch": 21.55,
      "learning_rate": 5e-05,
      "loss": 2.3033,
      "step": 11400
    },
    {
      "epoch": 21.74,
      "learning_rate": 5e-05,
      "loss": 2.1144,
      "step": 11500
    },
    {
      "epoch": 21.93,
      "learning_rate": 5e-05,
      "loss": 3.7434,
      "step": 11600
    },
    {
      "epoch": 22.0,
      "eval_accuracy": 0.22333333333333333,
      "eval_loss": 10.118596076965332,
      "eval_runtime": 4.3085,
      "eval_samples_per_second": 116.049,
      "eval_steps_per_second": 14.622,
      "step": 11638
    },
    {
      "epoch": 22.0,
      "eval_exact_match": 0.0,
      "eval_f1": 0.0,
      "step": 11638
    },
    {
      "epoch": 22.12,
      "learning_rate": 5e-05,
      "loss": 9.1896,
      "step": 11700
    },
    {
      "epoch": 22.31,
      "learning_rate": 5e-05,
      "loss": 9.6027,
      "step": 11800
    },
    {
      "epoch": 22.5,
      "learning_rate": 5e-05,
      "loss": 11.4681,
      "step": 11900
    },
    {
      "epoch": 22.68,
      "learning_rate": 5e-05,
      "loss": 8.4061,
      "step": 12000
    },
    {
      "epoch": 22.87,
      "learning_rate": 5e-05,
      "loss": 7.7259,
      "step": 12100
    },
    {
      "epoch": 23.0,
      "eval_accuracy": 0.19856410256410256,
      "eval_loss": 7.537852764129639,
      "eval_runtime": 4.1062,
      "eval_samples_per_second": 121.768,
      "eval_steps_per_second": 15.343,
      "step": 12167
    },
    {
      "epoch": 23.0,
      "eval_exact_match": 0.0,
      "eval_f1": 0.08333333333333333,
      "step": 12167
    },
    {
      "epoch": 23.06,
      "learning_rate": 5e-05,
      "loss": 7.3869,
      "step": 12200
    },
    {
      "epoch": 23.25,
      "learning_rate": 5e-05,
      "loss": 7.3438,
      "step": 12300
    },
    {
      "epoch": 23.44,
      "learning_rate": 5e-05,
      "loss": 7.1672,
      "step": 12400
    },
    {
      "epoch": 23.63,
      "learning_rate": 5e-05,
      "loss": 7.5678,
      "step": 12500
    },
    {
      "epoch": 23.82,
      "learning_rate": 5e-05,
      "loss": 4.2204,
      "step": 12600
    },
    {
      "epoch": 24.0,
      "eval_accuracy": 0.5344615384615384,
      "eval_loss": 2.1567773818969727,
      "eval_runtime": 4.8187,
      "eval_samples_per_second": 103.763,
      "eval_steps_per_second": 13.074,
      "step": 12696
    },
    {
      "epoch": 24.0,
      "eval_exact_match": 0.0,
      "eval_f1": 0.1556396274796546,
      "step": 12696
    },
    {
      "epoch": 24.01,
      "learning_rate": 5e-05,
      "loss": 2.243,
      "step": 12700
    },
    {
      "epoch": 24.2,
      "learning_rate": 5e-05,
      "loss": 1.4067,
      "step": 12800
    },
    {
      "epoch": 24.39,
      "learning_rate": 5e-05,
      "loss": 1.0065,
      "step": 12900
    },
    {
      "epoch": 24.57,
      "learning_rate": 5e-05,
      "loss": 0.8179,
      "step": 13000
    },
    {
      "epoch": 24.76,
      "learning_rate": 5e-05,
      "loss": 0.7917,
      "step": 13100
    },
    {
      "epoch": 24.95,
      "learning_rate": 5e-05,
      "loss": 0.7385,
      "step": 13200
    },
    {
      "epoch": 25.0,
      "eval_accuracy": 0.5963076923076923,
      "eval_loss": 1.8228886127471924,
      "eval_runtime": 4.8179,
      "eval_samples_per_second": 103.781,
      "eval_steps_per_second": 13.076,
      "step": 13225
    },
    {
      "epoch": 25.0,
      "eval_exact_match": 5.2,
      "eval_f1": 8.718379522680063,
      "step": 13225
    },
    {
      "epoch": 25.14,
      "learning_rate": 5e-05,
      "loss": 0.8862,
      "step": 13300
    },
    {
      "epoch": 25.33,
      "learning_rate": 5e-05,
      "loss": 0.5963,
      "step": 13400
    },
    {
      "epoch": 25.52,
      "learning_rate": 5e-05,
      "loss": 2.7783,
      "step": 13500
    },
    {
      "epoch": 25.71,
      "learning_rate": 5e-05,
      "loss": 1.4347,
      "step": 13600
    },
    {
      "epoch": 25.9,
      "learning_rate": 5e-05,
      "loss": 1.1473,
      "step": 13700
    },
    {
      "epoch": 26.0,
      "eval_accuracy": 0.5788205128205128,
      "eval_loss": 1.7569822072982788,
      "eval_runtime": 4.1999,
      "eval_samples_per_second": 119.049,
      "eval_steps_per_second": 15.0,
      "step": 13754
    },
    {
      "epoch": 26.0,
      "eval_exact_match": 0.0,
      "eval_f1": 0.0,
      "step": 13754
    },
    {
      "epoch": 26.09,
      "learning_rate": 5e-05,
      "loss": 1.2078,
      "step": 13800
    },
    {
      "epoch": 26.28,
      "learning_rate": 5e-05,
      "loss": 1.2602,
      "step": 13900
    },
    {
      "epoch": 26.47,
      "learning_rate": 5e-05,
      "loss": 2.3188,
      "step": 14000
    },
    {
      "epoch": 26.65,
      "learning_rate": 5e-05,
      "loss": 2.2177,
      "step": 14100
    },
    {
      "epoch": 26.84,
      "learning_rate": 5e-05,
      "loss": 2.0182,
      "step": 14200
    },
    {
      "epoch": 27.0,
      "eval_accuracy": 0.5573333333333333,
      "eval_loss": 1.729324221611023,
      "eval_runtime": 4.5012,
      "eval_samples_per_second": 111.082,
      "eval_steps_per_second": 13.996,
      "step": 14283
    },
    {
      "epoch": 27.0,
      "eval_exact_match": 0.0,
      "eval_f1": 0.03798603689725877,
      "step": 14283
    },
    {
      "epoch": 27.03,
      "learning_rate": 5e-05,
      "loss": 1.4892,
      "step": 14300
    },
    {
      "epoch": 27.22,
      "learning_rate": 5e-05,
      "loss": 1.6886,
      "step": 14400
    },
    {
      "epoch": 27.41,
      "learning_rate": 5e-05,
      "loss": 2.2033,
      "step": 14500
    },
    {
      "epoch": 27.6,
      "learning_rate": 5e-05,
      "loss": 2.5533,
      "step": 14600
    },
    {
      "epoch": 27.79,
      "learning_rate": 5e-05,
      "loss": 2.5524,
      "step": 14700
    },
    {
      "epoch": 27.98,
      "learning_rate": 5e-05,
      "loss": 2.2707,
      "step": 14800
    },
    {
      "epoch": 28.0,
      "eval_accuracy": 0.4955897435897436,
      "eval_loss": 2.7017431259155273,
      "eval_runtime": 4.2078,
      "eval_samples_per_second": 118.828,
      "eval_steps_per_second": 14.972,
      "step": 14812
    },
    {
      "epoch": 28.0,
      "eval_exact_match": 0.0,
      "eval_f1": 0.0059701492537313425,
      "step": 14812
    }
  ],
  "logging_steps": 100,
  "max_steps": 26450,
  "num_train_epochs": 50,
  "save_steps": 500,
  "total_flos": 2.5017805261640827e+18,
  "trial_name": null,
  "trial_params": null
}