{
  "best_metric": 1.3951435089111328,
  "best_model_checkpoint": "./results/checkpoint-18616",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 27924,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.10743446497636441,
      "grad_norm": 4.838443756103516,
      "learning_rate": 4.9462827675118175e-05,
      "loss": 3.2819,
      "step": 500
    },
    {
      "epoch": 0.21486892995272883,
      "grad_norm": 3.5501515865325928,
      "learning_rate": 4.892565535023636e-05,
      "loss": 2.3948,
      "step": 1000
    },
    {
      "epoch": 0.32230339492909327,
      "grad_norm": 3.361682653427124,
      "learning_rate": 4.8388483025354535e-05,
      "loss": 2.1166,
      "step": 1500
    },
    {
      "epoch": 0.42973785990545765,
      "grad_norm": 3.207756996154785,
      "learning_rate": 4.7851310700472715e-05,
      "loss": 1.983,
      "step": 2000
    },
    {
      "epoch": 0.5371723248818221,
      "grad_norm": 3.38034725189209,
      "learning_rate": 4.7314138375590894e-05,
      "loss": 1.8785,
      "step": 2500
    },
    {
      "epoch": 0.6446067898581865,
      "grad_norm": 3.3562798500061035,
      "learning_rate": 4.677696605070907e-05,
      "loss": 1.7957,
      "step": 3000
    },
    {
      "epoch": 0.752041254834551,
      "grad_norm": 2.897372007369995,
      "learning_rate": 4.623979372582725e-05,
      "loss": 1.7387,
      "step": 3500
    },
    {
      "epoch": 0.8594757198109153,
      "grad_norm": 2.9306259155273438,
      "learning_rate": 4.570262140094543e-05,
      "loss": 1.7103,
      "step": 4000
    },
    {
      "epoch": 0.9669101847872797,
      "grad_norm": 2.7418951988220215,
      "learning_rate": 4.5165449076063606e-05,
      "loss": 1.6637,
      "step": 4500
    },
    {
      "epoch": 1.0,
      "eval_loss": 1.5266185998916626,
      "eval_rouge2_fmeasure": 0.4101,
      "eval_rouge2_precision": 0.4101,
      "eval_rouge2_recall": 0.4101,
      "eval_runtime": 72.9922,
      "eval_samples_per_second": 65.582,
      "eval_steps_per_second": 1.028,
      "step": 4654
    },
    {
      "epoch": 1.0743446497636442,
      "grad_norm": 3.1563010215759277,
      "learning_rate": 4.462827675118178e-05,
      "loss": 1.4927,
      "step": 5000
    },
    {
      "epoch": 1.1817791147400085,
      "grad_norm": 2.845944404602051,
      "learning_rate": 4.409110442629996e-05,
      "loss": 1.4188,
      "step": 5500
    },
    {
      "epoch": 1.289213579716373,
      "grad_norm": 2.7506182193756104,
      "learning_rate": 4.355393210141814e-05,
      "loss": 1.418,
      "step": 6000
    },
    {
      "epoch": 1.3966480446927374,
      "grad_norm": 2.623769760131836,
      "learning_rate": 4.301675977653631e-05,
      "loss": 1.406,
      "step": 6500
    },
    {
      "epoch": 1.504082509669102,
      "grad_norm": 2.671093225479126,
      "learning_rate": 4.247958745165449e-05,
      "loss": 1.4021,
      "step": 7000
    },
    {
      "epoch": 1.6115169746454663,
      "grad_norm": 2.96425724029541,
      "learning_rate": 4.194241512677267e-05,
      "loss": 1.3883,
      "step": 7500
    },
    {
      "epoch": 1.7189514396218306,
      "grad_norm": 2.6702911853790283,
      "learning_rate": 4.140524280189085e-05,
      "loss": 1.3883,
      "step": 8000
    },
    {
      "epoch": 1.8263859045981952,
      "grad_norm": 2.6645655632019043,
      "learning_rate": 4.0868070477009024e-05,
      "loss": 1.3761,
      "step": 8500
    },
    {
      "epoch": 1.9338203695745595,
      "grad_norm": 3.023545265197754,
      "learning_rate": 4.0330898152127204e-05,
      "loss": 1.376,
      "step": 9000
    },
    {
      "epoch": 2.0,
      "eval_loss": 1.410889744758606,
      "eval_rouge2_fmeasure": 0.4238,
      "eval_rouge2_precision": 0.4238,
      "eval_rouge2_recall": 0.4238,
      "eval_runtime": 72.9371,
      "eval_samples_per_second": 65.632,
      "eval_steps_per_second": 1.028,
      "step": 9308
    },
    {
      "epoch": 2.041254834550924,
      "grad_norm": 2.6128861904144287,
      "learning_rate": 3.979372582724538e-05,
      "loss": 1.2778,
      "step": 9500
    },
    {
      "epoch": 2.1486892995272884,
      "grad_norm": 2.602187395095825,
      "learning_rate": 3.9256553502363556e-05,
      "loss": 1.1422,
      "step": 10000
    },
    {
      "epoch": 2.256123764503653,
      "grad_norm": 2.725907802581787,
      "learning_rate": 3.8719381177481736e-05,
      "loss": 1.1449,
      "step": 10500
    },
    {
      "epoch": 2.363558229480017,
      "grad_norm": 2.6348464488983154,
      "learning_rate": 3.8182208852599916e-05,
      "loss": 1.1529,
      "step": 11000
    },
    {
      "epoch": 2.4709926944563816,
      "grad_norm": 2.8584792613983154,
      "learning_rate": 3.764503652771809e-05,
      "loss": 1.1552,
      "step": 11500
    },
    {
      "epoch": 2.578427159432746,
      "grad_norm": 2.59346342086792,
      "learning_rate": 3.7107864202836275e-05,
      "loss": 1.1666,
      "step": 12000
    },
    {
      "epoch": 2.6858616244091102,
      "grad_norm": 2.655982255935669,
      "learning_rate": 3.657069187795445e-05,
      "loss": 1.1658,
      "step": 12500
    },
    {
      "epoch": 2.793296089385475,
      "grad_norm": 2.5818564891815186,
      "learning_rate": 3.603351955307263e-05,
      "loss": 1.1674,
      "step": 13000
    },
    {
      "epoch": 2.9007305543618394,
      "grad_norm": 2.8041298389434814,
      "learning_rate": 3.549634722819081e-05,
      "loss": 1.1684,
      "step": 13500
    },
    {
      "epoch": 3.0,
      "eval_loss": 1.3994433879852295,
      "eval_rouge2_fmeasure": 0.4311,
      "eval_rouge2_precision": 0.4311,
      "eval_rouge2_recall": 0.4311,
      "eval_runtime": 72.9642,
      "eval_samples_per_second": 65.607,
      "eval_steps_per_second": 1.028,
      "step": 13962
    },
    {
      "epoch": 3.008165019338204,
      "grad_norm": 2.646003246307373,
      "learning_rate": 3.495917490330898e-05,
      "loss": 1.1485,
      "step": 14000
    },
    {
      "epoch": 3.115599484314568,
      "grad_norm": 2.8257687091827393,
      "learning_rate": 3.442200257842716e-05,
      "loss": 0.9456,
      "step": 14500
    },
    {
      "epoch": 3.2230339492909326,
      "grad_norm": 2.9319422245025635,
      "learning_rate": 3.388483025354534e-05,
      "loss": 0.961,
      "step": 15000
    },
    {
      "epoch": 3.330468414267297,
      "grad_norm": 2.7773501873016357,
      "learning_rate": 3.334765792866352e-05,
      "loss": 0.9748,
      "step": 15500
    },
    {
      "epoch": 3.4379028792436612,
      "grad_norm": 2.673140048980713,
      "learning_rate": 3.281048560378169e-05,
      "loss": 0.9822,
      "step": 16000
    },
    {
      "epoch": 3.5453373442200258,
      "grad_norm": 2.7191991806030273,
      "learning_rate": 3.227331327889987e-05,
      "loss": 0.9808,
      "step": 16500
    },
    {
      "epoch": 3.6527718091963903,
      "grad_norm": 2.717005491256714,
      "learning_rate": 3.173614095401805e-05,
      "loss": 0.9832,
      "step": 17000
    },
    {
      "epoch": 3.760206274172755,
      "grad_norm": 2.696438789367676,
      "learning_rate": 3.1198968629136225e-05,
      "loss": 0.9954,
      "step": 17500
    },
    {
      "epoch": 3.867640739149119,
      "grad_norm": 2.5392231941223145,
      "learning_rate": 3.066179630425441e-05,
      "loss": 0.9978,
      "step": 18000
    },
    {
      "epoch": 3.9750752041254835,
      "grad_norm": 3.3882222175598145,
      "learning_rate": 3.0124623979372585e-05,
      "loss": 0.9957,
      "step": 18500
    },
    {
      "epoch": 4.0,
      "eval_loss": 1.3951435089111328,
      "eval_rouge2_fmeasure": 0.4323,
      "eval_rouge2_precision": 0.4323,
      "eval_rouge2_recall": 0.4323,
      "eval_runtime": 72.9673,
      "eval_samples_per_second": 65.605,
      "eval_steps_per_second": 1.028,
      "step": 18616
    },
    {
      "epoch": 4.082509669101848,
      "grad_norm": 2.655376672744751,
      "learning_rate": 2.958745165449076e-05,
      "loss": 0.8449,
      "step": 19000
    },
    {
      "epoch": 4.189944134078212,
      "grad_norm": 2.5333900451660156,
      "learning_rate": 2.9050279329608944e-05,
      "loss": 0.8125,
      "step": 19500
    },
    {
      "epoch": 4.297378599054577,
      "grad_norm": 2.4350857734680176,
      "learning_rate": 2.8513107004727117e-05,
      "loss": 0.8184,
      "step": 20000
    },
    {
      "epoch": 4.404813064030941,
      "grad_norm": 2.595292329788208,
      "learning_rate": 2.7975934679845293e-05,
      "loss": 0.8317,
      "step": 20500
    },
    {
      "epoch": 4.512247529007306,
      "grad_norm": 2.4971327781677246,
      "learning_rate": 2.743876235496347e-05,
      "loss": 0.833,
      "step": 21000
    },
    {
      "epoch": 4.61968199398367,
      "grad_norm": 2.7136857509613037,
      "learning_rate": 2.6901590030081653e-05,
      "loss": 0.8361,
      "step": 21500
    },
    {
      "epoch": 4.727116458960034,
      "grad_norm": 2.8456897735595703,
      "learning_rate": 2.636441770519983e-05,
      "loss": 0.8458,
      "step": 22000
    },
    {
      "epoch": 4.834550923936399,
      "grad_norm": 2.601877212524414,
      "learning_rate": 2.5827245380318005e-05,
      "loss": 0.846,
      "step": 22500
    },
    {
      "epoch": 4.941985388912763,
      "grad_norm": 2.6655149459838867,
      "learning_rate": 2.5290073055436185e-05,
      "loss": 0.848,
      "step": 23000
    },
    {
      "epoch": 5.0,
      "eval_loss": 1.42265784740448,
      "eval_rouge2_fmeasure": 0.4307,
      "eval_rouge2_precision": 0.4307,
      "eval_rouge2_recall": 0.4307,
      "eval_runtime": 73.0163,
      "eval_samples_per_second": 65.561,
      "eval_steps_per_second": 1.027,
      "step": 23270
    },
    {
      "epoch": 5.049419853889128,
      "grad_norm": 2.525017023086548,
      "learning_rate": 2.475290073055436e-05,
      "loss": 0.7735,
      "step": 23500
    },
    {
      "epoch": 5.156854318865492,
      "grad_norm": 2.2720000743865967,
      "learning_rate": 2.421572840567254e-05,
      "loss": 0.6797,
      "step": 24000
    },
    {
      "epoch": 5.264288783841857,
      "grad_norm": 2.633282423019409,
      "learning_rate": 2.3678556080790718e-05,
      "loss": 0.6955,
      "step": 24500
    },
    {
      "epoch": 5.3717232488182205,
      "grad_norm": 2.6699864864349365,
      "learning_rate": 2.3141383755908897e-05,
      "loss": 0.7026,
      "step": 25000
    },
    {
      "epoch": 5.479157713794585,
      "grad_norm": 2.765111207962036,
      "learning_rate": 2.2604211431027074e-05,
      "loss": 0.7045,
      "step": 25500
    },
    {
      "epoch": 5.58659217877095,
      "grad_norm": 2.8248813152313232,
      "learning_rate": 2.206703910614525e-05,
      "loss": 0.7148,
      "step": 26000
    },
    {
      "epoch": 5.694026643747314,
      "grad_norm": 2.7719056606292725,
      "learning_rate": 2.152986678126343e-05,
      "loss": 0.7218,
      "step": 26500
    },
    {
      "epoch": 5.801461108723679,
      "grad_norm": 2.8628671169281006,
      "learning_rate": 2.099269445638161e-05,
      "loss": 0.717,
      "step": 27000
    },
    {
      "epoch": 5.908895573700043,
      "grad_norm": 2.479224443435669,
      "learning_rate": 2.0455522131499786e-05,
      "loss": 0.7247,
      "step": 27500
    },
    {
      "epoch": 6.0,
      "eval_loss": 1.479453682899475,
      "eval_rouge2_fmeasure": 0.43,
      "eval_rouge2_precision": 0.43,
      "eval_rouge2_recall": 0.43,
      "eval_runtime": 73.0025,
      "eval_samples_per_second": 65.573,
      "eval_steps_per_second": 1.027,
      "step": 27924
    }
  ],
  "logging_steps": 500,
  "max_steps": 46540,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 3,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 2
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.9760389250501837e+17,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}