{
  "best_metric": 0.7649385333061218,
  "best_model_checkpoint": "xblock-social-screenshots-4/checkpoint-1542",
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 1542,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.05,
      "grad_norm": 12.289417266845703,
      "learning_rate": 8.064516129032258e-06,
      "loss": 1.0277,
      "step": 25
    },
    {
      "epoch": 0.1,
      "grad_norm": 9.727346420288086,
      "learning_rate": 1.5806451612903226e-05,
      "loss": 1.1841,
      "step": 50
    },
    {
      "epoch": 0.15,
      "grad_norm": 14.582966804504395,
      "learning_rate": 2.3870967741935486e-05,
      "loss": 1.1487,
      "step": 75
    },
    {
      "epoch": 0.19,
      "grad_norm": 13.005010604858398,
      "learning_rate": 3.193548387096774e-05,
      "loss": 1.2848,
      "step": 100
    },
    {
      "epoch": 0.24,
      "grad_norm": 7.13414192199707,
      "learning_rate": 4e-05,
      "loss": 1.2483,
      "step": 125
    },
    {
      "epoch": 0.29,
      "grad_norm": 10.042699813842773,
      "learning_rate": 4.806451612903226e-05,
      "loss": 1.0897,
      "step": 150
    },
    {
      "epoch": 0.34,
      "grad_norm": 11.965900421142578,
      "learning_rate": 4.935111751982697e-05,
      "loss": 1.1733,
      "step": 175
    },
    {
      "epoch": 0.39,
      "grad_norm": 6.2085347175598145,
      "learning_rate": 4.844989185291997e-05,
      "loss": 1.2392,
      "step": 200
    },
    {
      "epoch": 0.44,
      "grad_norm": 7.71270751953125,
      "learning_rate": 4.7548666186012983e-05,
      "loss": 1.1152,
      "step": 225
    },
    {
      "epoch": 0.49,
      "grad_norm": 17.0334529876709,
      "learning_rate": 4.6647440519105984e-05,
      "loss": 1.1423,
      "step": 250
    },
    {
      "epoch": 0.54,
      "grad_norm": 9.85234546661377,
      "learning_rate": 4.574621485219899e-05,
      "loss": 1.1417,
      "step": 275
    },
    {
      "epoch": 0.58,
      "grad_norm": 7.187493801116943,
      "learning_rate": 4.4844989185292e-05,
      "loss": 1.1791,
      "step": 300
    },
    {
      "epoch": 0.63,
      "grad_norm": 8.927938461303711,
      "learning_rate": 4.394376351838501e-05,
      "loss": 1.1275,
      "step": 325
    },
    {
      "epoch": 0.68,
      "grad_norm": 10.718963623046875,
      "learning_rate": 4.304253785147801e-05,
      "loss": 1.0876,
      "step": 350
    },
    {
      "epoch": 0.73,
      "grad_norm": 9.936092376708984,
      "learning_rate": 4.214131218457102e-05,
      "loss": 1.0834,
      "step": 375
    },
    {
      "epoch": 0.78,
      "grad_norm": 11.561012268066406,
      "learning_rate": 4.124008651766402e-05,
      "loss": 0.9958,
      "step": 400
    },
    {
      "epoch": 0.83,
      "grad_norm": 8.201454162597656,
      "learning_rate": 4.033886085075703e-05,
      "loss": 1.1471,
      "step": 425
    },
    {
      "epoch": 0.88,
      "grad_norm": 10.632408142089844,
      "learning_rate": 3.943763518385004e-05,
      "loss": 1.1176,
      "step": 450
    },
    {
      "epoch": 0.92,
      "grad_norm": 13.286883354187012,
      "learning_rate": 3.8536409516943046e-05,
      "loss": 1.1054,
      "step": 475
    },
    {
      "epoch": 0.97,
      "grad_norm": 9.36621379852295,
      "learning_rate": 3.763518385003605e-05,
      "loss": 1.1142,
      "step": 500
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7281553398058253,
      "eval_f1_macro": 0.3571554737240677,
      "eval_f1_micro": 0.7281553398058251,
      "eval_f1_weighted": 0.6902205394370374,
      "eval_loss": 0.8712501525878906,
      "eval_precision_macro": 0.4901931986654834,
      "eval_precision_micro": 0.7281553398058253,
      "eval_precision_weighted": 0.698922565321561,
      "eval_recall_macro": 0.33272931310043147,
      "eval_recall_micro": 0.7281553398058253,
      "eval_recall_weighted": 0.7281553398058253,
      "eval_runtime": 41.8624,
      "eval_samples_per_second": 24.604,
      "eval_steps_per_second": 1.553,
      "step": 514
    },
    {
      "epoch": 1.02,
      "grad_norm": 2.906583786010742,
      "learning_rate": 3.673395818312906e-05,
      "loss": 1.0156,
      "step": 525
    },
    {
      "epoch": 1.07,
      "grad_norm": 4.102613925933838,
      "learning_rate": 3.583273251622206e-05,
      "loss": 0.962,
      "step": 550
    },
    {
      "epoch": 1.12,
      "grad_norm": 8.407520294189453,
      "learning_rate": 3.493150684931507e-05,
      "loss": 0.9899,
      "step": 575
    },
    {
      "epoch": 1.17,
      "grad_norm": 7.069764137268066,
      "learning_rate": 3.403028118240808e-05,
      "loss": 1.0278,
      "step": 600
    },
    {
      "epoch": 1.22,
      "grad_norm": 15.905261993408203,
      "learning_rate": 3.3129055515501085e-05,
      "loss": 1.0595,
      "step": 625
    },
    {
      "epoch": 1.26,
      "grad_norm": 6.158019542694092,
      "learning_rate": 3.2227829848594086e-05,
      "loss": 1.1077,
      "step": 650
    },
    {
      "epoch": 1.31,
      "grad_norm": 9.805268287658691,
      "learning_rate": 3.13266041816871e-05,
      "loss": 0.8815,
      "step": 675
    },
    {
      "epoch": 1.36,
      "grad_norm": 15.077588081359863,
      "learning_rate": 3.04253785147801e-05,
      "loss": 0.967,
      "step": 700
    },
    {
      "epoch": 1.41,
      "grad_norm": 9.116120338439941,
      "learning_rate": 2.952415284787311e-05,
      "loss": 0.9285,
      "step": 725
    },
    {
      "epoch": 1.46,
      "grad_norm": 6.24492883682251,
      "learning_rate": 2.8622927180966113e-05,
      "loss": 0.8676,
      "step": 750
    },
    {
      "epoch": 1.51,
      "grad_norm": 8.600728988647461,
      "learning_rate": 2.7721701514059124e-05,
      "loss": 1.0795,
      "step": 775
    },
    {
      "epoch": 1.56,
      "grad_norm": 12.541471481323242,
      "learning_rate": 2.6820475847152128e-05,
      "loss": 0.7429,
      "step": 800
    },
    {
      "epoch": 1.61,
      "grad_norm": 7.928451061248779,
      "learning_rate": 2.5919250180245136e-05,
      "loss": 0.8436,
      "step": 825
    },
    {
      "epoch": 1.65,
      "grad_norm": 12.788630485534668,
      "learning_rate": 2.501802451333814e-05,
      "loss": 1.0718,
      "step": 850
    },
    {
      "epoch": 1.7,
      "grad_norm": 12.828117370605469,
      "learning_rate": 2.4116798846431148e-05,
      "loss": 1.0588,
      "step": 875
    },
    {
      "epoch": 1.75,
      "grad_norm": 8.542132377624512,
      "learning_rate": 2.3215573179524152e-05,
      "loss": 0.9365,
      "step": 900
    },
    {
      "epoch": 1.8,
      "grad_norm": 6.623122692108154,
      "learning_rate": 2.231434751261716e-05,
      "loss": 0.788,
      "step": 925
    },
    {
      "epoch": 1.85,
      "grad_norm": 10.277917861938477,
      "learning_rate": 2.1413121845710167e-05,
      "loss": 0.7988,
      "step": 950
    },
    {
      "epoch": 1.9,
      "grad_norm": 10.783781051635742,
      "learning_rate": 2.051189617880317e-05,
      "loss": 0.9515,
      "step": 975
    },
    {
      "epoch": 1.95,
      "grad_norm": 9.338010787963867,
      "learning_rate": 1.961067051189618e-05,
      "loss": 0.8298,
      "step": 1000
    },
    {
      "epoch": 1.99,
      "grad_norm": 8.552199363708496,
      "learning_rate": 1.8709444844989187e-05,
      "loss": 1.0186,
      "step": 1025
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7495145631067961,
      "eval_f1_macro": 0.4164105966638123,
      "eval_f1_micro": 0.7495145631067962,
      "eval_f1_weighted": 0.7192389597003817,
      "eval_loss": 0.7742837071418762,
      "eval_precision_macro": 0.6024626428820864,
      "eval_precision_micro": 0.7495145631067961,
      "eval_precision_weighted": 0.7393294285471907,
      "eval_recall_macro": 0.374337750947193,
      "eval_recall_micro": 0.7495145631067961,
      "eval_recall_weighted": 0.7495145631067961,
      "eval_runtime": 42.4523,
      "eval_samples_per_second": 24.263,
      "eval_steps_per_second": 1.531,
      "step": 1028
    },
    {
      "epoch": 2.04,
      "grad_norm": 9.136883735656738,
      "learning_rate": 1.780821917808219e-05,
      "loss": 0.779,
      "step": 1050
    },
    {
      "epoch": 2.09,
      "grad_norm": 10.608307838439941,
      "learning_rate": 1.6943042537851476e-05,
      "loss": 0.9101,
      "step": 1075
    },
    {
      "epoch": 2.14,
      "grad_norm": 7.822091579437256,
      "learning_rate": 1.6041816870944484e-05,
      "loss": 0.892,
      "step": 1100
    },
    {
      "epoch": 2.19,
      "grad_norm": 8.77257251739502,
      "learning_rate": 1.5140591204037492e-05,
      "loss": 0.6383,
      "step": 1125
    },
    {
      "epoch": 2.24,
      "grad_norm": 5.078084468841553,
      "learning_rate": 1.4239365537130498e-05,
      "loss": 0.842,
      "step": 1150
    },
    {
      "epoch": 2.29,
      "grad_norm": 6.022009372711182,
      "learning_rate": 1.3338139870223503e-05,
      "loss": 0.6754,
      "step": 1175
    },
    {
      "epoch": 2.33,
      "grad_norm": 9.004426002502441,
      "learning_rate": 1.2436914203316511e-05,
      "loss": 0.9429,
      "step": 1200
    },
    {
      "epoch": 2.38,
      "grad_norm": 8.876343727111816,
      "learning_rate": 1.1535688536409517e-05,
      "loss": 0.801,
      "step": 1225
    },
    {
      "epoch": 2.43,
      "grad_norm": 7.864119052886963,
      "learning_rate": 1.0634462869502523e-05,
      "loss": 0.7809,
      "step": 1250
    },
    {
      "epoch": 2.48,
      "grad_norm": 9.604496955871582,
      "learning_rate": 9.73323720259553e-06,
      "loss": 0.7273,
      "step": 1275
    },
    {
      "epoch": 2.53,
      "grad_norm": 10.21464729309082,
      "learning_rate": 8.832011535688536e-06,
      "loss": 0.8211,
      "step": 1300
    },
    {
      "epoch": 2.58,
      "grad_norm": 10.150245666503906,
      "learning_rate": 7.930785868781542e-06,
      "loss": 0.7916,
      "step": 1325
    },
    {
      "epoch": 2.63,
      "grad_norm": 8.246322631835938,
      "learning_rate": 7.029560201874549e-06,
      "loss": 0.6806,
      "step": 1350
    },
    {
      "epoch": 2.68,
      "grad_norm": 5.490677356719971,
      "learning_rate": 6.128334534967556e-06,
      "loss": 0.6664,
      "step": 1375
    },
    {
      "epoch": 2.72,
      "grad_norm": 9.993110656738281,
      "learning_rate": 5.227108868060563e-06,
      "loss": 0.8779,
      "step": 1400
    },
    {
      "epoch": 2.77,
      "grad_norm": 11.303633689880371,
      "learning_rate": 4.325883201153569e-06,
      "loss": 0.8084,
      "step": 1425
    },
    {
      "epoch": 2.82,
      "grad_norm": 12.19831657409668,
      "learning_rate": 3.4246575342465754e-06,
      "loss": 0.7864,
      "step": 1450
    },
    {
      "epoch": 2.87,
      "grad_norm": 7.894024848937988,
      "learning_rate": 2.523431867339582e-06,
      "loss": 0.8498,
      "step": 1475
    },
    {
      "epoch": 2.92,
      "grad_norm": 7.604086399078369,
      "learning_rate": 1.6222062004325884e-06,
      "loss": 0.7161,
      "step": 1500
    },
    {
      "epoch": 2.97,
      "grad_norm": 9.083874702453613,
      "learning_rate": 7.209805335255948e-07,
      "loss": 0.9195,
      "step": 1525
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7640776699029126,
      "eval_f1_macro": 0.4616966462629057,
      "eval_f1_micro": 0.7640776699029127,
      "eval_f1_weighted": 0.7363203004281954,
      "eval_loss": 0.7649385333061218,
      "eval_precision_macro": 0.6252686879910994,
      "eval_precision_micro": 0.7640776699029126,
      "eval_precision_weighted": 0.7552425547875681,
      "eval_recall_macro": 0.4119139234072332,
      "eval_recall_micro": 0.7640776699029126,
      "eval_recall_weighted": 0.7640776699029126,
      "eval_runtime": 41.5633,
      "eval_samples_per_second": 24.781,
      "eval_steps_per_second": 1.564,
      "step": 1542
    }
  ],
  "logging_steps": 25,
  "max_steps": 1542,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 9.553208099078246e+17,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}