{
  "best_metric": 0.6138996138996139,
  "best_model_checkpoint": "./Validated_cracks_raw_dataset_359_relabeled2_outputs/checkpoint-432",
  "epoch": 30.0,
  "eval_steps": 500,
  "global_step": 3240,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.46296296296296297,
      "grad_norm": 11.11386489868164,
      "learning_rate": 2.7777777777777783e-06,
      "loss": 0.9464,
      "step": 50
    },
    {
      "epoch": 0.9259259259259259,
      "grad_norm": 22.558269500732422,
      "learning_rate": 5.864197530864199e-06,
      "loss": 0.9468,
      "step": 100
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6023166023166023,
      "eval_loss": 0.9880222678184509,
      "eval_runtime": 1.7676,
      "eval_samples_per_second": 146.524,
      "eval_steps_per_second": 18.669,
      "step": 108
    },
    {
      "epoch": 1.3888888888888888,
      "grad_norm": 23.05500030517578,
      "learning_rate": 8.950617283950618e-06,
      "loss": 0.9061,
      "step": 150
    },
    {
      "epoch": 1.8518518518518519,
      "grad_norm": 20.498937606811523,
      "learning_rate": 9.997164110237345e-06,
      "loss": 0.9275,
      "step": 200
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.583011583011583,
      "eval_loss": 1.0344593524932861,
      "eval_runtime": 1.7521,
      "eval_samples_per_second": 147.825,
      "eval_steps_per_second": 18.835,
      "step": 216
    },
    {
      "epoch": 2.314814814814815,
      "grad_norm": 10.24534797668457,
      "learning_rate": 9.98206922906478e-06,
      "loss": 0.9077,
      "step": 250
    },
    {
      "epoch": 2.7777777777777777,
      "grad_norm": 62.0069580078125,
      "learning_rate": 9.954002016824226e-06,
      "loss": 0.9377,
      "step": 300
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.5675675675675675,
      "eval_loss": 1.023147463798523,
      "eval_runtime": 1.7707,
      "eval_samples_per_second": 146.27,
      "eval_steps_per_second": 18.637,
      "step": 324
    },
    {
      "epoch": 3.240740740740741,
      "grad_norm": 19.093095779418945,
      "learning_rate": 9.91398066213228e-06,
      "loss": 0.8912,
      "step": 350
    },
    {
      "epoch": 3.7037037037037037,
      "grad_norm": 14.229964256286621,
      "learning_rate": 9.860476209695013e-06,
      "loss": 0.8461,
      "step": 400
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6138996138996139,
      "eval_loss": 0.9997061491012573,
      "eval_runtime": 1.7439,
      "eval_samples_per_second": 148.519,
      "eval_steps_per_second": 18.923,
      "step": 432
    },
    {
      "epoch": 4.166666666666667,
      "grad_norm": 25.24408721923828,
      "learning_rate": 9.794316030562418e-06,
      "loss": 0.8663,
      "step": 450
    },
    {
      "epoch": 4.62962962962963,
      "grad_norm": 27.978118896484375,
      "learning_rate": 9.715672392864017e-06,
      "loss": 0.8429,
      "step": 500
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.5868725868725869,
      "eval_loss": 0.9872773885726929,
      "eval_runtime": 1.7504,
      "eval_samples_per_second": 147.962,
      "eval_steps_per_second": 18.852,
      "step": 540
    },
    {
      "epoch": 5.092592592592593,
      "grad_norm": 22.3253173828125,
      "learning_rate": 9.62475006920687e-06,
      "loss": 0.8246,
      "step": 550
    },
    {
      "epoch": 5.555555555555555,
      "grad_norm": 29.907615661621094,
      "learning_rate": 9.521785803487888e-06,
      "loss": 0.781,
      "step": 600
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.6061776061776062,
      "eval_loss": 0.9970757365226746,
      "eval_runtime": 1.7424,
      "eval_samples_per_second": 148.643,
      "eval_steps_per_second": 18.939,
      "step": 648
    },
    {
      "epoch": 6.018518518518518,
      "grad_norm": 23.293292999267578,
      "learning_rate": 9.407047694459149e-06,
      "loss": 0.8409,
      "step": 650
    },
    {
      "epoch": 6.481481481481482,
      "grad_norm": 21.52410888671875,
      "learning_rate": 9.280834497651334e-06,
      "loss": 0.8137,
      "step": 700
    },
    {
      "epoch": 6.944444444444445,
      "grad_norm": 34.57286071777344,
      "learning_rate": 9.143474847472932e-06,
      "loss": 0.7621,
      "step": 750
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.583011583011583,
      "eval_loss": 1.0092536211013794,
      "eval_runtime": 1.7382,
      "eval_samples_per_second": 149.004,
      "eval_steps_per_second": 18.985,
      "step": 756
    },
    {
      "epoch": 7.407407407407407,
      "grad_norm": 15.154962539672852,
      "learning_rate": 8.9953264015107e-06,
      "loss": 0.7191,
      "step": 800
    },
    {
      "epoch": 7.87037037037037,
      "grad_norm": 40.10356140136719,
      "learning_rate": 8.836774909259442e-06,
      "loss": 0.7649,
      "step": 850
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.5675675675675675,
      "eval_loss": 1.0584689378738403,
      "eval_runtime": 1.7294,
      "eval_samples_per_second": 149.767,
      "eval_steps_per_second": 19.082,
      "step": 864
    },
    {
      "epoch": 8.333333333333334,
      "grad_norm": 65.0152587890625,
      "learning_rate": 8.668233207706014e-06,
      "loss": 0.7822,
      "step": 900
    },
    {
      "epoch": 8.796296296296296,
      "grad_norm": 27.71040916442871,
      "learning_rate": 8.490140146382767e-06,
      "loss": 0.7912,
      "step": 950
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.5752895752895753,
      "eval_loss": 1.0433698892593384,
      "eval_runtime": 1.7507,
      "eval_samples_per_second": 147.94,
      "eval_steps_per_second": 18.85,
      "step": 972
    },
    {
      "epoch": 9.25925925925926,
      "grad_norm": 35.2958984375,
      "learning_rate": 8.302959444689442e-06,
      "loss": 0.7821,
      "step": 1000
    },
    {
      "epoch": 9.722222222222221,
      "grad_norm": 47.18446731567383,
      "learning_rate": 8.107178484458825e-06,
      "loss": 0.7529,
      "step": 1050
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.61003861003861,
      "eval_loss": 1.022578477859497,
      "eval_runtime": 1.747,
      "eval_samples_per_second": 148.256,
      "eval_steps_per_second": 18.89,
      "step": 1080
    },
    {
      "epoch": 10.185185185185185,
      "grad_norm": 18.027130126953125,
      "learning_rate": 7.903307040910035e-06,
      "loss": 0.7478,
      "step": 1100
    },
    {
      "epoch": 10.648148148148149,
      "grad_norm": 17.87278175354004,
      "learning_rate": 7.691875955293851e-06,
      "loss": 0.6924,
      "step": 1150
    },
    {
      "epoch": 11.0,
      "eval_accuracy": 0.5868725868725869,
      "eval_loss": 1.0567034482955933,
      "eval_runtime": 1.7505,
      "eval_samples_per_second": 147.961,
      "eval_steps_per_second": 18.852,
      "step": 1188
    },
    {
      "epoch": 11.11111111111111,
      "grad_norm": 30.35451316833496,
      "learning_rate": 7.473435752686213e-06,
      "loss": 0.7193,
      "step": 1200
    },
    {
      "epoch": 11.574074074074074,
      "grad_norm": 27.861392974853516,
      "learning_rate": 7.248555208528913e-06,
      "loss": 0.7015,
      "step": 1250
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.583011583011583,
      "eval_loss": 1.0600991249084473,
      "eval_runtime": 1.7432,
      "eval_samples_per_second": 148.579,
      "eval_steps_per_second": 18.931,
      "step": 1296
    },
    {
      "epoch": 12.037037037037036,
      "grad_norm": 16.875736236572266,
      "learning_rate": 7.017819867649898e-06,
      "loss": 0.7261,
      "step": 1300
    },
    {
      "epoch": 12.5,
      "grad_norm": 29.472204208374023,
      "learning_rate": 6.781830519619392e-06,
      "loss": 0.7286,
      "step": 1350
    },
    {
      "epoch": 12.962962962962964,
      "grad_norm": 79.9528579711914,
      "learning_rate": 6.541201634411704e-06,
      "loss": 0.7094,
      "step": 1400
    },
    {
      "epoch": 13.0,
      "eval_accuracy": 0.5637065637065637,
      "eval_loss": 1.0830413103103638,
      "eval_runtime": 1.7485,
      "eval_samples_per_second": 148.126,
      "eval_steps_per_second": 18.873,
      "step": 1404
    },
    {
      "epoch": 13.425925925925926,
      "grad_norm": 32.787437438964844,
      "learning_rate": 6.296559762445928e-06,
      "loss": 0.6587,
      "step": 1450
    },
    {
      "epoch": 13.88888888888889,
      "grad_norm": 21.805397033691406,
      "learning_rate": 6.048541903171552e-06,
      "loss": 0.749,
      "step": 1500
    },
    {
      "epoch": 14.0,
      "eval_accuracy": 0.5907335907335908,
      "eval_loss": 1.0383802652359009,
      "eval_runtime": 1.7396,
      "eval_samples_per_second": 148.885,
      "eval_steps_per_second": 18.97,
      "step": 1512
    },
    {
      "epoch": 14.351851851851851,
      "grad_norm": 14.21861457824707,
      "learning_rate": 5.79779384644684e-06,
      "loss": 0.6565,
      "step": 1550
    },
    {
      "epoch": 14.814814814814815,
      "grad_norm": 111.92208862304688,
      "learning_rate": 5.550041104970398e-06,
      "loss": 0.7433,
      "step": 1600
    },
    {
      "epoch": 15.0,
      "eval_accuracy": 0.5675675675675675,
      "eval_loss": 1.0580743551254272,
      "eval_runtime": 1.737,
      "eval_samples_per_second": 149.108,
      "eval_steps_per_second": 18.998,
      "step": 1620
    },
    {
      "epoch": 15.277777777777779,
      "grad_norm": 27.13836669921875,
      "learning_rate": 5.2958186601645375e-06,
      "loss": 0.7729,
      "step": 1650
    },
    {
      "epoch": 15.74074074074074,
      "grad_norm": 41.231163024902344,
      "learning_rate": 5.040825961590755e-06,
      "loss": 0.7595,
      "step": 1700
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.5714285714285714,
      "eval_loss": 1.0641090869903564,
      "eval_runtime": 1.7544,
      "eval_samples_per_second": 147.629,
      "eval_steps_per_second": 18.81,
      "step": 1728
    },
    {
      "epoch": 16.203703703703702,
      "grad_norm": 18.411924362182617,
      "learning_rate": 4.785726960220769e-06,
      "loss": 0.7803,
      "step": 1750
    },
    {
      "epoch": 16.666666666666668,
      "grad_norm": 34.5395393371582,
      "learning_rate": 4.531185883817934e-06,
      "loss": 0.8132,
      "step": 1800
    },
    {
      "epoch": 17.0,
      "eval_accuracy": 0.5714285714285714,
      "eval_loss": 1.0863285064697266,
      "eval_runtime": 1.7323,
      "eval_samples_per_second": 149.51,
      "eval_steps_per_second": 19.05,
      "step": 1836
    },
    {
      "epoch": 17.12962962962963,
      "grad_norm": 30.300302505493164,
      "learning_rate": 4.277865507418416e-06,
      "loss": 0.8507,
      "step": 1850
    },
    {
      "epoch": 17.59259259259259,
      "grad_norm": 19.417051315307617,
      "learning_rate": 4.026425427595017e-06,
      "loss": 0.7951,
      "step": 1900
    },
    {
      "epoch": 18.0,
      "eval_accuracy": 0.5868725868725869,
      "eval_loss": 1.0517489910125732,
      "eval_runtime": 1.7405,
      "eval_samples_per_second": 148.809,
      "eval_steps_per_second": 18.96,
      "step": 1944
    },
    {
      "epoch": 18.055555555555557,
      "grad_norm": 20.044008255004883,
      "learning_rate": 3.777520344997052e-06,
      "loss": 0.8131,
      "step": 1950
    },
    {
      "epoch": 18.51851851851852,
      "grad_norm": 29.19565200805664,
      "learning_rate": 3.5317983596382976e-06,
      "loss": 0.7714,
      "step": 2000
    },
    {
      "epoch": 18.98148148148148,
      "grad_norm": 32.99634552001953,
      "learning_rate": 3.289899283371657e-06,
      "loss": 0.7861,
      "step": 2050
    },
    {
      "epoch": 19.0,
      "eval_accuracy": 0.5907335907335908,
      "eval_loss": 1.0631417036056519,
      "eval_runtime": 1.739,
      "eval_samples_per_second": 148.94,
      "eval_steps_per_second": 18.977,
      "step": 2052
    },
    {
      "epoch": 19.444444444444443,
      "grad_norm": 20.90323829650879,
      "learning_rate": 3.052452973944603e-06,
      "loss": 0.7129,
      "step": 2100
    },
    {
      "epoch": 19.90740740740741,
      "grad_norm": 40.24602508544922,
      "learning_rate": 2.8200776949731435e-06,
      "loss": 0.7334,
      "step": 2150
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.5907335907335908,
      "eval_loss": 1.0507447719573975,
      "eval_runtime": 1.893,
      "eval_samples_per_second": 136.818,
      "eval_steps_per_second": 17.432,
      "step": 2160
    },
    {
      "epoch": 20.37037037037037,
      "grad_norm": 19.65435028076172,
      "learning_rate": 2.5933785061046333e-06,
      "loss": 0.7563,
      "step": 2200
    },
    {
      "epoch": 20.833333333333332,
      "grad_norm": 96.42213439941406,
      "learning_rate": 2.372945687561132e-06,
      "loss": 0.6807,
      "step": 2250
    },
    {
      "epoch": 21.0,
      "eval_accuracy": 0.5791505791505791,
      "eval_loss": 1.0659432411193848,
      "eval_runtime": 1.7425,
      "eval_samples_per_second": 148.637,
      "eval_steps_per_second": 18.938,
      "step": 2268
    },
    {
      "epoch": 21.296296296296298,
      "grad_norm": 30.792505264282227,
      "learning_rate": 2.159353203165484e-06,
      "loss": 0.7706,
      "step": 2300
    },
    {
      "epoch": 21.75925925925926,
      "grad_norm": 62.397857666015625,
      "learning_rate": 1.9531572058521074e-06,
      "loss": 0.6905,
      "step": 2350
    },
    {
      "epoch": 22.0,
      "eval_accuracy": 0.5752895752895753,
      "eval_loss": 1.0680582523345947,
      "eval_runtime": 1.7542,
      "eval_samples_per_second": 147.645,
      "eval_steps_per_second": 18.812,
      "step": 2376
    },
    {
      "epoch": 22.22222222222222,
      "grad_norm": 17.79286766052246,
      "learning_rate": 1.754894589553836e-06,
      "loss": 0.763,
      "step": 2400
    },
    {
      "epoch": 22.685185185185187,
      "grad_norm": 20.265037536621094,
      "learning_rate": 1.5650815912354285e-06,
      "loss": 0.7242,
      "step": 2450
    },
    {
      "epoch": 23.0,
      "eval_accuracy": 0.5907335907335908,
      "eval_loss": 1.0529667139053345,
      "eval_runtime": 1.7502,
      "eval_samples_per_second": 147.984,
      "eval_steps_per_second": 18.855,
      "step": 2484
    },
    {
      "epoch": 23.14814814814815,
      "grad_norm": 43.985389709472656,
      "learning_rate": 1.3842124467137664e-06,
      "loss": 0.7003,
      "step": 2500
    },
    {
      "epoch": 23.61111111111111,
      "grad_norm": 58.851200103759766,
      "learning_rate": 1.212758103764733e-06,
      "loss": 0.7342,
      "step": 2550
    },
    {
      "epoch": 24.0,
      "eval_accuracy": 0.5637065637065637,
      "eval_loss": 1.0659518241882324,
      "eval_runtime": 1.7466,
      "eval_samples_per_second": 148.291,
      "eval_steps_per_second": 18.894,
      "step": 2592
    },
    {
      "epoch": 24.074074074074073,
      "grad_norm": 34.4920654296875,
      "learning_rate": 1.0511649958675745e-06,
      "loss": 0.7262,
      "step": 2600
    },
    {
      "epoch": 24.537037037037038,
      "grad_norm": 29.049121856689453,
      "learning_rate": 8.998538797796879e-07,
      "loss": 0.6924,
      "step": 2650
    },
    {
      "epoch": 25.0,
      "grad_norm": 54.39902877807617,
      "learning_rate": 7.592187399685524e-07,
      "loss": 0.7736,
      "step": 2700
    },
    {
      "epoch": 25.0,
      "eval_accuracy": 0.5791505791505791,
      "eval_loss": 1.070913314819336,
      "eval_runtime": 1.7438,
      "eval_samples_per_second": 148.528,
      "eval_steps_per_second": 18.924,
      "step": 2700
    },
    {
      "epoch": 25.462962962962962,
      "grad_norm": 41.184539794921875,
      "learning_rate": 6.296257627534552e-07,
      "loss": 0.6561,
      "step": 2750
    },
    {
      "epoch": 25.925925925925927,
      "grad_norm": 57.05064010620117,
      "learning_rate": 5.114123828281398e-07,
      "loss": 0.7798,
      "step": 2800
    },
    {
      "epoch": 26.0,
      "eval_accuracy": 0.5714285714285714,
      "eval_loss": 1.0627210140228271,
      "eval_runtime": 1.737,
      "eval_samples_per_second": 149.108,
      "eval_steps_per_second": 18.998,
      "step": 2808
    },
    {
      "epoch": 26.38888888888889,
      "grad_norm": 29.81525993347168,
      "learning_rate": 4.048864046470502e-07,
      "loss": 0.7423,
      "step": 2850
    },
    {
      "epoch": 26.85185185185185,
      "grad_norm": 36.73592758178711,
      "learning_rate": 3.1032520096290817e-07,
      "loss": 0.71,
      "step": 2900
    },
    {
      "epoch": 27.0,
      "eval_accuracy": 0.5868725868725869,
      "eval_loss": 1.060496211051941,
      "eval_runtime": 1.7352,
      "eval_samples_per_second": 149.263,
      "eval_steps_per_second": 19.018,
      "step": 2916
    },
    {
      "epoch": 27.314814814814813,
      "grad_norm": 63.34288024902344,
      "learning_rate": 2.2797499060246253e-07,
      "loss": 0.6777,
      "step": 2950
    },
    {
      "epoch": 27.77777777777778,
      "grad_norm": 25.37461280822754,
      "learning_rate": 1.5805019736097105e-07,
      "loss": 0.706,
      "step": 3000
    },
    {
      "epoch": 28.0,
      "eval_accuracy": 0.5791505791505791,
      "eval_loss": 1.0605989694595337,
      "eval_runtime": 1.7369,
      "eval_samples_per_second": 149.116,
      "eval_steps_per_second": 18.999,
      "step": 3024
    },
    {
      "epoch": 28.24074074074074,
      "grad_norm": 42.660560607910156,
      "learning_rate": 1.007328916846817e-07,
      "loss": 0.7275,
      "step": 3050
    },
    {
      "epoch": 28.703703703703702,
      "grad_norm": 31.769351959228516,
      "learning_rate": 5.617231659511446e-08,
      "loss": 0.7421,
      "step": 3100
    },
    {
      "epoch": 29.0,
      "eval_accuracy": 0.5752895752895753,
      "eval_loss": 1.0606402158737183,
      "eval_runtime": 1.7325,
      "eval_samples_per_second": 149.494,
      "eval_steps_per_second": 19.048,
      "step": 3132
    },
    {
      "epoch": 29.166666666666668,
      "grad_norm": 23.639081954956055,
      "learning_rate": 2.448449908949879e-08,
      "loss": 0.6976,
      "step": 3150
    },
    {
      "epoch": 29.62962962962963,
      "grad_norm": 30.702539443969727,
      "learning_rate": 5.751948029234511e-09,
      "loss": 0.654,
      "step": 3200
    },
    {
      "epoch": 30.0,
      "eval_accuracy": 0.5752895752895753,
      "eval_loss": 1.060823917388916,
      "eval_runtime": 3.0177,
      "eval_samples_per_second": 85.826,
      "eval_steps_per_second": 10.935,
      "step": 3240
    },
    {
      "epoch": 30.0,
      "step": 3240,
      "total_flos": 2.0405460235862016e+18,
      "train_loss": 0.7678916913491708,
      "train_runtime": 680.1267,
      "train_samples_per_second": 37.934,
      "train_steps_per_second": 4.764
    }
  ],
  "logging_steps": 50,
  "max_steps": 3240,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 30,
  "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": 2.0405460235862016e+18,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}