{
  "best_global_step": 2500,
  "best_metric": 3.4325599670410156,
  "best_model_checkpoint": "/kaggle/working/checkpoints/checkpoint-2500",
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 2688,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.05585032113934655,
      "grad_norm": 0.708092451095581,
      "learning_rate": 3.0246913580246916e-05,
      "loss": 20.53129150390625,
      "step": 50
    },
    {
      "epoch": 0.1117006422786931,
      "grad_norm": 0.6190989017486572,
      "learning_rate": 4.9654775604142695e-05,
      "loss": 20.3445361328125,
      "step": 100
    },
    {
      "epoch": 0.16755096341803966,
      "grad_norm": 0.5028762221336365,
      "learning_rate": 4.8695818948983505e-05,
      "loss": 20.208651123046874,
      "step": 150
    },
    {
      "epoch": 0.2234012845573862,
      "grad_norm": 0.5830720663070679,
      "learning_rate": 4.773686229382432e-05,
      "loss": 20.0191845703125,
      "step": 200
    },
    {
      "epoch": 0.27925160569673274,
      "grad_norm": 1.2075963020324707,
      "learning_rate": 4.677790563866514e-05,
      "loss": 19.857196044921874,
      "step": 250
    },
    {
      "epoch": 0.3351019268360793,
      "grad_norm": 0.7960541844367981,
      "learning_rate": 4.581894898350595e-05,
      "loss": 19.7943798828125,
      "step": 300
    },
    {
      "epoch": 0.3909522479754259,
      "grad_norm": 0.5681961178779602,
      "learning_rate": 4.4859992328346764e-05,
      "loss": 19.644644775390624,
      "step": 350
    },
    {
      "epoch": 0.4468025691147724,
      "grad_norm": 0.5845203995704651,
      "learning_rate": 4.3901035673187574e-05,
      "loss": 19.550875244140624,
      "step": 400
    },
    {
      "epoch": 0.502652890254119,
      "grad_norm": 0.579823911190033,
      "learning_rate": 4.2942079018028384e-05,
      "loss": 19.49732666015625,
      "step": 450
    },
    {
      "epoch": 0.5585032113934655,
      "grad_norm": 0.5753065943717957,
      "learning_rate": 4.19831223628692e-05,
      "loss": 19.393812255859373,
      "step": 500
    },
    {
      "epoch": 0.5585032113934655,
      "eval_loss": 3.5303454399108887,
      "eval_runtime": 9.0561,
      "eval_samples_per_second": 63.935,
      "eval_steps_per_second": 8.061,
      "step": 500
    },
    {
      "epoch": 0.6143535325328121,
      "grad_norm": 0.6646850109100342,
      "learning_rate": 4.102416570771002e-05,
      "loss": 19.3396826171875,
      "step": 550
    },
    {
      "epoch": 0.6702038536721586,
      "grad_norm": 0.5389931797981262,
      "learning_rate": 4.006520905255083e-05,
      "loss": 19.26292724609375,
      "step": 600
    },
    {
      "epoch": 0.7260541748115051,
      "grad_norm": 18.08650779724121,
      "learning_rate": 3.9106252397391644e-05,
      "loss": 19.2409912109375,
      "step": 650
    },
    {
      "epoch": 0.7819044959508518,
      "grad_norm": 0.5375329852104187,
      "learning_rate": 3.8147295742232454e-05,
      "loss": 19.192125244140627,
      "step": 700
    },
    {
      "epoch": 0.8377548170901983,
      "grad_norm": 0.5194340348243713,
      "learning_rate": 3.7188339087073263e-05,
      "loss": 19.237432861328124,
      "step": 750
    },
    {
      "epoch": 0.8936051382295448,
      "grad_norm": 0.5628377199172974,
      "learning_rate": 3.622938243191408e-05,
      "loss": 19.087825927734375,
      "step": 800
    },
    {
      "epoch": 0.9494554593688914,
      "grad_norm": 0.5108382105827332,
      "learning_rate": 3.52704257767549e-05,
      "loss": 19.06609375,
      "step": 850
    },
    {
      "epoch": 1.0044680256911478,
      "grad_norm": 0.5543744564056396,
      "learning_rate": 3.4311469121595707e-05,
      "loss": 18.77072021484375,
      "step": 900
    },
    {
      "epoch": 1.0603183468304942,
      "grad_norm": 0.6037471890449524,
      "learning_rate": 3.3352512466436516e-05,
      "loss": 18.96982421875,
      "step": 950
    },
    {
      "epoch": 1.1161686679698408,
      "grad_norm": 0.5822243094444275,
      "learning_rate": 3.239355581127733e-05,
      "loss": 18.948385009765627,
      "step": 1000
    },
    {
      "epoch": 1.1161686679698408,
      "eval_loss": 3.4744184017181396,
      "eval_runtime": 9.0468,
      "eval_samples_per_second": 64.001,
      "eval_steps_per_second": 8.069,
      "step": 1000
    },
    {
      "epoch": 1.1720189891091874,
      "grad_norm": 0.6110577583312988,
      "learning_rate": 3.143459915611814e-05,
      "loss": 18.887904052734374,
      "step": 1050
    },
    {
      "epoch": 1.2278693102485339,
      "grad_norm": 0.6810954213142395,
      "learning_rate": 3.0475642500958956e-05,
      "loss": 18.870596923828124,
      "step": 1100
    },
    {
      "epoch": 1.2837196313878805,
      "grad_norm": 0.5141286849975586,
      "learning_rate": 2.951668584579977e-05,
      "loss": 18.86670654296875,
      "step": 1150
    },
    {
      "epoch": 1.339569952527227,
      "grad_norm": 0.5824834704399109,
      "learning_rate": 2.8557729190640586e-05,
      "loss": 18.940731201171875,
      "step": 1200
    },
    {
      "epoch": 1.3954202736665735,
      "grad_norm": 0.5844902992248535,
      "learning_rate": 2.75987725354814e-05,
      "loss": 18.92754150390625,
      "step": 1250
    },
    {
      "epoch": 1.4512705948059201,
      "grad_norm": 0.5832068920135498,
      "learning_rate": 2.6639815880322212e-05,
      "loss": 18.806781005859374,
      "step": 1300
    },
    {
      "epoch": 1.5071209159452668,
      "grad_norm": 0.5681942701339722,
      "learning_rate": 2.5680859225163022e-05,
      "loss": 18.814879150390624,
      "step": 1350
    },
    {
      "epoch": 1.5629712370846134,
      "grad_norm": 0.5460590720176697,
      "learning_rate": 2.472190257000384e-05,
      "loss": 18.78223388671875,
      "step": 1400
    },
    {
      "epoch": 1.6188215582239598,
      "grad_norm": 0.6541451811790466,
      "learning_rate": 2.376294591484465e-05,
      "loss": 18.7625,
      "step": 1450
    },
    {
      "epoch": 1.6746718793633062,
      "grad_norm": 0.5325785279273987,
      "learning_rate": 2.2803989259685462e-05,
      "loss": 18.798499755859375,
      "step": 1500
    },
    {
      "epoch": 1.6746718793633062,
      "eval_loss": 3.448370933532715,
      "eval_runtime": 9.0498,
      "eval_samples_per_second": 63.98,
      "eval_steps_per_second": 8.067,
      "step": 1500
    },
    {
      "epoch": 1.7305222005026528,
      "grad_norm": 1.295803189277649,
      "learning_rate": 2.184503260452628e-05,
      "loss": 18.752318115234374,
      "step": 1550
    },
    {
      "epoch": 1.7863725216419994,
      "grad_norm": 0.5719322562217712,
      "learning_rate": 2.088607594936709e-05,
      "loss": 18.77613037109375,
      "step": 1600
    },
    {
      "epoch": 1.842222842781346,
      "grad_norm": 0.5277766585350037,
      "learning_rate": 1.9927119294207902e-05,
      "loss": 18.774757080078125,
      "step": 1650
    },
    {
      "epoch": 1.8980731639206927,
      "grad_norm": 0.7951223254203796,
      "learning_rate": 1.896816263904872e-05,
      "loss": 18.718056640625,
      "step": 1700
    },
    {
      "epoch": 1.953923485060039,
      "grad_norm": 0.6089980602264404,
      "learning_rate": 1.8009205983889528e-05,
      "loss": 18.657005615234375,
      "step": 1750
    },
    {
      "epoch": 2.0089360513822956,
      "grad_norm": 0.689607560634613,
      "learning_rate": 1.705024932873034e-05,
      "loss": 18.4244384765625,
      "step": 1800
    },
    {
      "epoch": 2.064786372521642,
      "grad_norm": 0.5486071705818176,
      "learning_rate": 1.6091292673571158e-05,
      "loss": 18.660447998046877,
      "step": 1850
    },
    {
      "epoch": 2.1206366936609884,
      "grad_norm": 0.5985398888587952,
      "learning_rate": 1.513233601841197e-05,
      "loss": 18.749791259765626,
      "step": 1900
    },
    {
      "epoch": 2.176487014800335,
      "grad_norm": 0.6870599985122681,
      "learning_rate": 1.4173379363252781e-05,
      "loss": 18.71692138671875,
      "step": 1950
    },
    {
      "epoch": 2.2323373359396816,
      "grad_norm": 0.5832541584968567,
      "learning_rate": 1.3214422708093596e-05,
      "loss": 18.6642626953125,
      "step": 2000
    },
    {
      "epoch": 2.2323373359396816,
      "eval_loss": 3.434812545776367,
      "eval_runtime": 9.0755,
      "eval_samples_per_second": 63.798,
      "eval_steps_per_second": 8.044,
      "step": 2000
    },
    {
      "epoch": 2.2881876570790283,
      "grad_norm": 0.5366337895393372,
      "learning_rate": 1.2255466052934408e-05,
      "loss": 18.6031298828125,
      "step": 2050
    },
    {
      "epoch": 2.344037978218375,
      "grad_norm": 0.5727832317352295,
      "learning_rate": 1.1296509397775221e-05,
      "loss": 18.664876708984377,
      "step": 2100
    },
    {
      "epoch": 2.3998882993577215,
      "grad_norm": 0.5798032283782959,
      "learning_rate": 1.0337552742616034e-05,
      "loss": 18.55861572265625,
      "step": 2150
    },
    {
      "epoch": 2.4557386204970677,
      "grad_norm": 0.6107646226882935,
      "learning_rate": 9.378596087456847e-06,
      "loss": 18.614580078125,
      "step": 2200
    },
    {
      "epoch": 2.5115889416364143,
      "grad_norm": 0.564208447933197,
      "learning_rate": 8.41963943229766e-06,
      "loss": 18.59590087890625,
      "step": 2250
    },
    {
      "epoch": 2.567439262775761,
      "grad_norm": 0.5309619903564453,
      "learning_rate": 7.460682777138474e-06,
      "loss": 18.567191162109374,
      "step": 2300
    },
    {
      "epoch": 2.6232895839151076,
      "grad_norm": 0.7527127265930176,
      "learning_rate": 6.501726121979287e-06,
      "loss": 18.62651611328125,
      "step": 2350
    },
    {
      "epoch": 2.679139905054454,
      "grad_norm": 0.5464082956314087,
      "learning_rate": 5.5427694668200995e-06,
      "loss": 18.56793212890625,
      "step": 2400
    },
    {
      "epoch": 2.7349902261938004,
      "grad_norm": 0.5442795753479004,
      "learning_rate": 4.583812811660914e-06,
      "loss": 18.675174560546875,
      "step": 2450
    },
    {
      "epoch": 2.790840547333147,
      "grad_norm": 0.6047973036766052,
      "learning_rate": 3.6248561565017264e-06,
      "loss": 18.68692626953125,
      "step": 2500
    },
    {
      "epoch": 2.790840547333147,
      "eval_loss": 3.4325599670410156,
      "eval_runtime": 9.0716,
      "eval_samples_per_second": 63.826,
      "eval_steps_per_second": 8.047,
      "step": 2500
    },
    {
      "epoch": 2.8466908684724936,
      "grad_norm": 0.5565047264099121,
      "learning_rate": 2.6658995013425397e-06,
      "loss": 18.537215576171874,
      "step": 2550
    },
    {
      "epoch": 2.9025411896118403,
      "grad_norm": 1.1597777605056763,
      "learning_rate": 1.7069428461833525e-06,
      "loss": 18.571363525390623,
      "step": 2600
    },
    {
      "epoch": 2.958391510751187,
      "grad_norm": 0.6304506659507751,
      "learning_rate": 7.479861910241657e-07,
      "loss": 18.627213134765626,
      "step": 2650
    },
    {
      "epoch": 3.0,
      "step": 2688,
      "total_flos": 2.281119745572864e+16,
      "train_loss": 18.99221926643735,
      "train_runtime": 5922.3639,
      "train_samples_per_second": 29.023,
      "train_steps_per_second": 0.454
    }
  ],
  "logging_steps": 50,
  "max_steps": 2688,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "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": 2.281119745572864e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}