{
  "best_global_step": 875,
  "best_metric": 0.41610679030418396,
  "best_model_checkpoint": "/home/ricoiban/GEMMA/mnlp_chatsplaining/results_model/try_ft/checkpoint-875",
  "epoch": 0.5460448642266824,
  "eval_steps": 25,
  "global_step": 925,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0029515938606847697,
      "grad_norm": 274.0,
      "learning_rate": 4.800000000000001e-06,
      "loss": 10.7873,
      "step": 5
    },
    {
      "epoch": 0.0059031877213695395,
      "grad_norm": 160.0,
      "learning_rate": 1.08e-05,
      "loss": 8.2818,
      "step": 10
    },
    {
      "epoch": 0.00885478158205431,
      "grad_norm": 96.0,
      "learning_rate": 1.6800000000000002e-05,
      "loss": 5.4171,
      "step": 15
    },
    {
      "epoch": 0.011806375442739079,
      "grad_norm": 141.0,
      "learning_rate": 2.2800000000000002e-05,
      "loss": 4.8734,
      "step": 20
    },
    {
      "epoch": 0.01475796930342385,
      "grad_norm": 111.0,
      "learning_rate": 2.88e-05,
      "loss": 4.5143,
      "step": 25
    },
    {
      "epoch": 0.01475796930342385,
      "eval_loss": 4.148748874664307,
      "eval_runtime": 22.5903,
      "eval_samples_per_second": 88.534,
      "eval_steps_per_second": 88.534,
      "step": 25
    },
    {
      "epoch": 0.01770956316410862,
      "grad_norm": 234.0,
      "learning_rate": 3.48e-05,
      "loss": 4.0827,
      "step": 30
    },
    {
      "epoch": 0.02066115702479339,
      "grad_norm": 173.0,
      "learning_rate": 4.08e-05,
      "loss": 3.624,
      "step": 35
    },
    {
      "epoch": 0.023612750885478158,
      "grad_norm": 91.0,
      "learning_rate": 4.6800000000000006e-05,
      "loss": 2.6036,
      "step": 40
    },
    {
      "epoch": 0.026564344746162927,
      "grad_norm": 163.0,
      "learning_rate": 5.28e-05,
      "loss": 1.9856,
      "step": 45
    },
    {
      "epoch": 0.0295159386068477,
      "grad_norm": 65.0,
      "learning_rate": 5.88e-05,
      "loss": 1.1888,
      "step": 50
    },
    {
      "epoch": 0.0295159386068477,
      "eval_loss": 0.8685765862464905,
      "eval_runtime": 22.0353,
      "eval_samples_per_second": 90.764,
      "eval_steps_per_second": 90.764,
      "step": 50
    },
    {
      "epoch": 0.032467532467532464,
      "grad_norm": 31.125,
      "learning_rate": 5.9999123594193744e-05,
      "loss": 0.7088,
      "step": 55
    },
    {
      "epoch": 0.03541912632821724,
      "grad_norm": 25.625,
      "learning_rate": 5.9995563283365586e-05,
      "loss": 0.4428,
      "step": 60
    },
    {
      "epoch": 0.03837072018890201,
      "grad_norm": 38.25,
      "learning_rate": 5.998926461693058e-05,
      "loss": 0.7334,
      "step": 65
    },
    {
      "epoch": 0.04132231404958678,
      "grad_norm": 13.375,
      "learning_rate": 5.9980228169906714e-05,
      "loss": 0.6483,
      "step": 70
    },
    {
      "epoch": 0.04427390791027155,
      "grad_norm": 41.5,
      "learning_rate": 5.9968454767249506e-05,
      "loss": 0.5933,
      "step": 75
    },
    {
      "epoch": 0.04427390791027155,
      "eval_loss": 0.6194470524787903,
      "eval_runtime": 22.0585,
      "eval_samples_per_second": 90.668,
      "eval_steps_per_second": 90.668,
      "step": 75
    },
    {
      "epoch": 0.047225501770956316,
      "grad_norm": 26.0,
      "learning_rate": 5.995394548377669e-05,
      "loss": 0.7882,
      "step": 80
    },
    {
      "epoch": 0.050177095631641085,
      "grad_norm": 17.875,
      "learning_rate": 5.993670164407008e-05,
      "loss": 0.6423,
      "step": 85
    },
    {
      "epoch": 0.053128689492325853,
      "grad_norm": 19.75,
      "learning_rate": 5.991672482235466e-05,
      "loss": 0.6952,
      "step": 90
    },
    {
      "epoch": 0.05608028335301062,
      "grad_norm": 19.875,
      "learning_rate": 5.9894016842354855e-05,
      "loss": 0.6287,
      "step": 95
    },
    {
      "epoch": 0.0590318772136954,
      "grad_norm": 19.75,
      "learning_rate": 5.986857977712809e-05,
      "loss": 0.6606,
      "step": 100
    },
    {
      "epoch": 0.0590318772136954,
      "eval_loss": 0.5243311524391174,
      "eval_runtime": 22.0497,
      "eval_samples_per_second": 90.704,
      "eval_steps_per_second": 90.704,
      "step": 100
    },
    {
      "epoch": 0.06198347107438017,
      "grad_norm": 26.25,
      "learning_rate": 5.9840415948875444e-05,
      "loss": 0.6193,
      "step": 105
    },
    {
      "epoch": 0.06493506493506493,
      "grad_norm": 9.9375,
      "learning_rate": 5.980952792872975e-05,
      "loss": 0.5167,
      "step": 110
    },
    {
      "epoch": 0.0678866587957497,
      "grad_norm": 35.0,
      "learning_rate": 5.9775918536520786e-05,
      "loss": 0.5922,
      "step": 115
    },
    {
      "epoch": 0.07083825265643448,
      "grad_norm": 35.25,
      "learning_rate": 5.973959084051791e-05,
      "loss": 0.7114,
      "step": 120
    },
    {
      "epoch": 0.07378984651711924,
      "grad_norm": 8.5625,
      "learning_rate": 5.970054815714995e-05,
      "loss": 0.6309,
      "step": 125
    },
    {
      "epoch": 0.07378984651711924,
      "eval_loss": 0.644968569278717,
      "eval_runtime": 22.0752,
      "eval_samples_per_second": 90.6,
      "eval_steps_per_second": 90.6,
      "step": 125
    },
    {
      "epoch": 0.07674144037780402,
      "grad_norm": 19.375,
      "learning_rate": 5.965879405070235e-05,
      "loss": 0.622,
      "step": 130
    },
    {
      "epoch": 0.07969303423848878,
      "grad_norm": 15.4375,
      "learning_rate": 5.961433233299193e-05,
      "loss": 0.5902,
      "step": 135
    },
    {
      "epoch": 0.08264462809917356,
      "grad_norm": 19.375,
      "learning_rate": 5.956716706301877e-05,
      "loss": 0.6647,
      "step": 140
    },
    {
      "epoch": 0.08559622195985832,
      "grad_norm": 11.0,
      "learning_rate": 5.951730254659569e-05,
      "loss": 0.7817,
      "step": 145
    },
    {
      "epoch": 0.0885478158205431,
      "grad_norm": 16.875,
      "learning_rate": 5.946474333595521e-05,
      "loss": 0.6154,
      "step": 150
    },
    {
      "epoch": 0.0885478158205431,
      "eval_loss": 0.5631881356239319,
      "eval_runtime": 22.058,
      "eval_samples_per_second": 90.67,
      "eval_steps_per_second": 90.67,
      "step": 150
    },
    {
      "epoch": 0.09149940968122787,
      "grad_norm": 17.625,
      "learning_rate": 5.9409494229333904e-05,
      "loss": 0.6532,
      "step": 155
    },
    {
      "epoch": 0.09445100354191263,
      "grad_norm": 22.25,
      "learning_rate": 5.935156027053442e-05,
      "loss": 0.6099,
      "step": 160
    },
    {
      "epoch": 0.09740259740259741,
      "grad_norm": 25.5,
      "learning_rate": 5.929094674846495e-05,
      "loss": 0.7848,
      "step": 165
    },
    {
      "epoch": 0.10035419126328217,
      "grad_norm": 22.0,
      "learning_rate": 5.922765919665644e-05,
      "loss": 0.5597,
      "step": 170
    },
    {
      "epoch": 0.10330578512396695,
      "grad_norm": 11.9375,
      "learning_rate": 5.916170339275745e-05,
      "loss": 0.6166,
      "step": 175
    },
    {
      "epoch": 0.10330578512396695,
      "eval_loss": 0.5547081828117371,
      "eval_runtime": 22.0339,
      "eval_samples_per_second": 90.769,
      "eval_steps_per_second": 90.769,
      "step": 175
    },
    {
      "epoch": 0.10625737898465171,
      "grad_norm": 12.625,
      "learning_rate": 5.909308535800664e-05,
      "loss": 0.5548,
      "step": 180
    },
    {
      "epoch": 0.10920897284533648,
      "grad_norm": 32.5,
      "learning_rate": 5.90218113566831e-05,
      "loss": 0.5793,
      "step": 185
    },
    {
      "epoch": 0.11216056670602124,
      "grad_norm": 19.0,
      "learning_rate": 5.8947887895534504e-05,
      "loss": 0.8419,
      "step": 190
    },
    {
      "epoch": 0.11511216056670602,
      "grad_norm": 12.5625,
      "learning_rate": 5.8871321723183046e-05,
      "loss": 0.6571,
      "step": 195
    },
    {
      "epoch": 0.1180637544273908,
      "grad_norm": 18.375,
      "learning_rate": 5.879211982950937e-05,
      "loss": 0.6579,
      "step": 200
    },
    {
      "epoch": 0.1180637544273908,
      "eval_loss": 0.5850950479507446,
      "eval_runtime": 22.079,
      "eval_samples_per_second": 90.584,
      "eval_steps_per_second": 90.584,
      "step": 200
    },
    {
      "epoch": 0.12101534828807556,
      "grad_norm": 20.0,
      "learning_rate": 5.871028944501446e-05,
      "loss": 0.5835,
      "step": 205
    },
    {
      "epoch": 0.12396694214876033,
      "grad_norm": 20.25,
      "learning_rate": 5.862583804015953e-05,
      "loss": 0.5418,
      "step": 210
    },
    {
      "epoch": 0.1269185360094451,
      "grad_norm": 27.375,
      "learning_rate": 5.853877332468404e-05,
      "loss": 0.6755,
      "step": 215
    },
    {
      "epoch": 0.12987012987012986,
      "grad_norm": 29.5,
      "learning_rate": 5.844910324690187e-05,
      "loss": 0.6172,
      "step": 220
    },
    {
      "epoch": 0.13282172373081463,
      "grad_norm": 25.375,
      "learning_rate": 5.835683599297568e-05,
      "loss": 0.5844,
      "step": 225
    },
    {
      "epoch": 0.13282172373081463,
      "eval_loss": 0.5579066276550293,
      "eval_runtime": 22.0218,
      "eval_samples_per_second": 90.819,
      "eval_steps_per_second": 90.819,
      "step": 225
    },
    {
      "epoch": 0.1357733175914994,
      "grad_norm": 16.5,
      "learning_rate": 5.8261979986169596e-05,
      "loss": 0.6147,
      "step": 230
    },
    {
      "epoch": 0.13872491145218419,
      "grad_norm": 10.6875,
      "learning_rate": 5.816454388608023e-05,
      "loss": 0.6352,
      "step": 235
    },
    {
      "epoch": 0.14167650531286896,
      "grad_norm": 16.125,
      "learning_rate": 5.8064536587846115e-05,
      "loss": 0.7107,
      "step": 240
    },
    {
      "epoch": 0.1446280991735537,
      "grad_norm": 23.5,
      "learning_rate": 5.7961967221335674e-05,
      "loss": 0.6752,
      "step": 245
    },
    {
      "epoch": 0.14757969303423848,
      "grad_norm": 27.625,
      "learning_rate": 5.7856845150313716e-05,
      "loss": 0.6039,
      "step": 250
    },
    {
      "epoch": 0.14757969303423848,
      "eval_loss": 0.5708025693893433,
      "eval_runtime": 22.0736,
      "eval_samples_per_second": 90.606,
      "eval_steps_per_second": 90.606,
      "step": 250
    },
    {
      "epoch": 0.15053128689492326,
      "grad_norm": 19.375,
      "learning_rate": 5.7749179971586596e-05,
      "loss": 0.6029,
      "step": 255
    },
    {
      "epoch": 0.15348288075560804,
      "grad_norm": 24.75,
      "learning_rate": 5.763898151412613e-05,
      "loss": 0.4986,
      "step": 260
    },
    {
      "epoch": 0.15643447461629278,
      "grad_norm": 36.0,
      "learning_rate": 5.752625983817225e-05,
      "loss": 0.6049,
      "step": 265
    },
    {
      "epoch": 0.15938606847697756,
      "grad_norm": 22.25,
      "learning_rate": 5.7411025234314634e-05,
      "loss": 0.5895,
      "step": 270
    },
    {
      "epoch": 0.16233766233766234,
      "grad_norm": 13.0,
      "learning_rate": 5.729328822255319e-05,
      "loss": 0.6957,
      "step": 275
    },
    {
      "epoch": 0.16233766233766234,
      "eval_loss": 0.5568270087242126,
      "eval_runtime": 22.052,
      "eval_samples_per_second": 90.695,
      "eval_steps_per_second": 90.695,
      "step": 275
    },
    {
      "epoch": 0.1652892561983471,
      "grad_norm": 27.75,
      "learning_rate": 5.717305955133773e-05,
      "loss": 0.7125,
      "step": 280
    },
    {
      "epoch": 0.1682408500590319,
      "grad_norm": 12.8125,
      "learning_rate": 5.7050350196586686e-05,
      "loss": 0.5977,
      "step": 285
    },
    {
      "epoch": 0.17119244391971664,
      "grad_norm": 18.75,
      "learning_rate": 5.692517136068511e-05,
      "loss": 0.5908,
      "step": 290
    },
    {
      "epoch": 0.1741440377804014,
      "grad_norm": 24.25,
      "learning_rate": 5.679753447146195e-05,
      "loss": 0.5334,
      "step": 295
    },
    {
      "epoch": 0.1770956316410862,
      "grad_norm": 11.625,
      "learning_rate": 5.666745118114688e-05,
      "loss": 0.4347,
      "step": 300
    },
    {
      "epoch": 0.1770956316410862,
      "eval_loss": 0.5026609301567078,
      "eval_runtime": 22.0534,
      "eval_samples_per_second": 90.689,
      "eval_steps_per_second": 90.689,
      "step": 300
    },
    {
      "epoch": 0.18004722550177096,
      "grad_norm": 24.75,
      "learning_rate": 5.6534933365306394e-05,
      "loss": 0.5473,
      "step": 305
    },
    {
      "epoch": 0.18299881936245574,
      "grad_norm": 27.75,
      "learning_rate": 5.6399993121759797e-05,
      "loss": 0.5315,
      "step": 310
    },
    {
      "epoch": 0.1859504132231405,
      "grad_norm": 11.3125,
      "learning_rate": 5.626264276947469e-05,
      "loss": 0.4026,
      "step": 315
    },
    {
      "epoch": 0.18890200708382526,
      "grad_norm": 27.125,
      "learning_rate": 5.612289484744238e-05,
      "loss": 0.55,
      "step": 320
    },
    {
      "epoch": 0.19185360094451004,
      "grad_norm": 29.375,
      "learning_rate": 5.5980762113533166e-05,
      "loss": 0.5988,
      "step": 325
    },
    {
      "epoch": 0.19185360094451004,
      "eval_loss": 0.5229803919792175,
      "eval_runtime": 22.0483,
      "eval_samples_per_second": 90.71,
      "eval_steps_per_second": 90.71,
      "step": 325
    },
    {
      "epoch": 0.19480519480519481,
      "grad_norm": 18.0,
      "learning_rate": 5.5836257543331644e-05,
      "loss": 0.5174,
      "step": 330
    },
    {
      "epoch": 0.19775678866587956,
      "grad_norm": 17.75,
      "learning_rate": 5.568939432895213e-05,
      "loss": 0.6662,
      "step": 335
    },
    {
      "epoch": 0.20070838252656434,
      "grad_norm": 13.4375,
      "learning_rate": 5.554018587783435e-05,
      "loss": 0.6594,
      "step": 340
    },
    {
      "epoch": 0.20365997638724911,
      "grad_norm": 20.125,
      "learning_rate": 5.538864581151943e-05,
      "loss": 0.5776,
      "step": 345
    },
    {
      "epoch": 0.2066115702479339,
      "grad_norm": 29.0,
      "learning_rate": 5.523478796440633e-05,
      "loss": 0.6647,
      "step": 350
    },
    {
      "epoch": 0.2066115702479339,
      "eval_loss": 0.5277854204177856,
      "eval_runtime": 22.0408,
      "eval_samples_per_second": 90.741,
      "eval_steps_per_second": 90.741,
      "step": 350
    },
    {
      "epoch": 0.20956316410861867,
      "grad_norm": 17.625,
      "learning_rate": 5.507862638248896e-05,
      "loss": 0.5446,
      "step": 355
    },
    {
      "epoch": 0.21251475796930341,
      "grad_norm": 15.9375,
      "learning_rate": 5.49201753220738e-05,
      "loss": 0.5346,
      "step": 360
    },
    {
      "epoch": 0.2154663518299882,
      "grad_norm": 13.8125,
      "learning_rate": 5.475944924847845e-05,
      "loss": 0.4782,
      "step": 365
    },
    {
      "epoch": 0.21841794569067297,
      "grad_norm": 46.0,
      "learning_rate": 5.459646283471106e-05,
      "loss": 0.6363,
      "step": 370
    },
    {
      "epoch": 0.22136953955135774,
      "grad_norm": 12.625,
      "learning_rate": 5.443123096013083e-05,
      "loss": 0.5603,
      "step": 375
    },
    {
      "epoch": 0.22136953955135774,
      "eval_loss": 0.49818602204322815,
      "eval_runtime": 22.0724,
      "eval_samples_per_second": 90.611,
      "eval_steps_per_second": 90.611,
      "step": 375
    },
    {
      "epoch": 0.2243211334120425,
      "grad_norm": 25.125,
      "learning_rate": 5.426376870908959e-05,
      "loss": 0.6536,
      "step": 380
    },
    {
      "epoch": 0.22727272727272727,
      "grad_norm": 17.25,
      "learning_rate": 5.409409136955476e-05,
      "loss": 0.5464,
      "step": 385
    },
    {
      "epoch": 0.23022432113341204,
      "grad_norm": 23.375,
      "learning_rate": 5.3922214431713654e-05,
      "loss": 0.5587,
      "step": 390
    },
    {
      "epoch": 0.23317591499409682,
      "grad_norm": 10.75,
      "learning_rate": 5.3748153586559385e-05,
      "loss": 0.5231,
      "step": 395
    },
    {
      "epoch": 0.2361275088547816,
      "grad_norm": 19.875,
      "learning_rate": 5.357192472445835e-05,
      "loss": 0.503,
      "step": 400
    },
    {
      "epoch": 0.2361275088547816,
      "eval_loss": 0.4984550178050995,
      "eval_runtime": 22.074,
      "eval_samples_per_second": 90.604,
      "eval_steps_per_second": 90.604,
      "step": 400
    },
    {
      "epoch": 0.23907910271546634,
      "grad_norm": 20.125,
      "learning_rate": 5.339354393369962e-05,
      "loss": 0.4524,
      "step": 405
    },
    {
      "epoch": 0.24203069657615112,
      "grad_norm": 11.375,
      "learning_rate": 5.321302749902615e-05,
      "loss": 0.492,
      "step": 410
    },
    {
      "epoch": 0.2449822904368359,
      "grad_norm": 20.75,
      "learning_rate": 5.303039190014818e-05,
      "loss": 0.4989,
      "step": 415
    },
    {
      "epoch": 0.24793388429752067,
      "grad_norm": 14.0,
      "learning_rate": 5.284565381023873e-05,
      "loss": 0.6195,
      "step": 420
    },
    {
      "epoch": 0.25088547815820544,
      "grad_norm": 25.125,
      "learning_rate": 5.265883009441147e-05,
      "loss": 0.5687,
      "step": 425
    },
    {
      "epoch": 0.25088547815820544,
      "eval_loss": 0.4847257137298584,
      "eval_runtime": 22.029,
      "eval_samples_per_second": 90.789,
      "eval_steps_per_second": 90.789,
      "step": 425
    },
    {
      "epoch": 0.2538370720188902,
      "grad_norm": 16.75,
      "learning_rate": 5.2469937808181055e-05,
      "loss": 0.4048,
      "step": 430
    },
    {
      "epoch": 0.256788665879575,
      "grad_norm": 39.75,
      "learning_rate": 5.227899419590614e-05,
      "loss": 0.5483,
      "step": 435
    },
    {
      "epoch": 0.2597402597402597,
      "grad_norm": 15.125,
      "learning_rate": 5.208601668921508e-05,
      "loss": 0.4843,
      "step": 440
    },
    {
      "epoch": 0.2626918536009445,
      "grad_norm": 15.5,
      "learning_rate": 5.1891022905414546e-05,
      "loss": 0.6146,
      "step": 445
    },
    {
      "epoch": 0.26564344746162927,
      "grad_norm": 22.875,
      "learning_rate": 5.169403064588125e-05,
      "loss": 0.4279,
      "step": 450
    },
    {
      "epoch": 0.26564344746162927,
      "eval_loss": 0.48101305961608887,
      "eval_runtime": 22.0566,
      "eval_samples_per_second": 90.676,
      "eval_steps_per_second": 90.676,
      "step": 450
    },
    {
      "epoch": 0.26859504132231404,
      "grad_norm": 28.625,
      "learning_rate": 5.1495057894436757e-05,
      "loss": 0.5749,
      "step": 455
    },
    {
      "epoch": 0.2715466351829988,
      "grad_norm": 20.5,
      "learning_rate": 5.1294122815705773e-05,
      "loss": 0.4963,
      "step": 460
    },
    {
      "epoch": 0.2744982290436836,
      "grad_norm": 22.25,
      "learning_rate": 5.109124375345781e-05,
      "loss": 0.4213,
      "step": 465
    },
    {
      "epoch": 0.27744982290436837,
      "grad_norm": 18.125,
      "learning_rate": 5.0886439228932576e-05,
      "loss": 0.5002,
      "step": 470
    },
    {
      "epoch": 0.28040141676505315,
      "grad_norm": 34.0,
      "learning_rate": 5.067972793914911e-05,
      "loss": 0.5136,
      "step": 475
    },
    {
      "epoch": 0.28040141676505315,
      "eval_loss": 0.5137303471565247,
      "eval_runtime": 22.0698,
      "eval_samples_per_second": 90.621,
      "eval_steps_per_second": 90.621,
      "step": 475
    },
    {
      "epoch": 0.2833530106257379,
      "grad_norm": 15.0,
      "learning_rate": 5.047112875519892e-05,
      "loss": 0.5162,
      "step": 480
    },
    {
      "epoch": 0.28630460448642264,
      "grad_norm": 38.0,
      "learning_rate": 5.02606607205232e-05,
      "loss": 0.4831,
      "step": 485
    },
    {
      "epoch": 0.2892561983471074,
      "grad_norm": 26.5,
      "learning_rate": 5.004834304917425e-05,
      "loss": 0.5147,
      "step": 490
    },
    {
      "epoch": 0.2922077922077922,
      "grad_norm": 37.0,
      "learning_rate": 4.983419512406151e-05,
      "loss": 0.553,
      "step": 495
    },
    {
      "epoch": 0.29515938606847697,
      "grad_norm": 23.625,
      "learning_rate": 4.9618236495181936e-05,
      "loss": 0.3999,
      "step": 500
    },
    {
      "epoch": 0.29515938606847697,
      "eval_loss": 0.5061969757080078,
      "eval_runtime": 22.0382,
      "eval_samples_per_second": 90.752,
      "eval_steps_per_second": 90.752,
      "step": 500
    },
    {
      "epoch": 0.29811097992916175,
      "grad_norm": 13.75,
      "learning_rate": 4.9400486877835325e-05,
      "loss": 0.4205,
      "step": 505
    },
    {
      "epoch": 0.3010625737898465,
      "grad_norm": 22.5,
      "learning_rate": 4.91809661508244e-05,
      "loss": 0.5904,
      "step": 510
    },
    {
      "epoch": 0.3040141676505313,
      "grad_norm": 16.75,
      "learning_rate": 4.895969435464009e-05,
      "loss": 0.5749,
      "step": 515
    },
    {
      "epoch": 0.3069657615112161,
      "grad_norm": 10.8125,
      "learning_rate": 4.873669168963196e-05,
      "loss": 0.5841,
      "step": 520
    },
    {
      "epoch": 0.30991735537190085,
      "grad_norm": 14.25,
      "learning_rate": 4.851197851416409e-05,
      "loss": 0.5454,
      "step": 525
    },
    {
      "epoch": 0.30991735537190085,
      "eval_loss": 0.6045746207237244,
      "eval_runtime": 22.0653,
      "eval_samples_per_second": 90.64,
      "eval_steps_per_second": 90.64,
      "step": 525
    },
    {
      "epoch": 0.31286894923258557,
      "grad_norm": 12.4375,
      "learning_rate": 4.828557534275651e-05,
      "loss": 0.4654,
      "step": 530
    },
    {
      "epoch": 0.31582054309327035,
      "grad_norm": 28.125,
      "learning_rate": 4.8057502844212406e-05,
      "loss": 0.5669,
      "step": 535
    },
    {
      "epoch": 0.3187721369539551,
      "grad_norm": 17.625,
      "learning_rate": 4.78277818397312e-05,
      "loss": 0.3907,
      "step": 540
    },
    {
      "epoch": 0.3217237308146399,
      "grad_norm": 12.625,
      "learning_rate": 4.7596433301007775e-05,
      "loss": 0.508,
      "step": 545
    },
    {
      "epoch": 0.3246753246753247,
      "grad_norm": 26.75,
      "learning_rate": 4.736347834831789e-05,
      "loss": 0.408,
      "step": 550
    },
    {
      "epoch": 0.3246753246753247,
      "eval_loss": 0.48477041721343994,
      "eval_runtime": 22.0413,
      "eval_samples_per_second": 90.739,
      "eval_steps_per_second": 90.739,
      "step": 550
    },
    {
      "epoch": 0.32762691853600945,
      "grad_norm": 32.25,
      "learning_rate": 4.712893824859008e-05,
      "loss": 0.3498,
      "step": 555
    },
    {
      "epoch": 0.3305785123966942,
      "grad_norm": 21.25,
      "learning_rate": 4.6892834413464163e-05,
      "loss": 0.6738,
      "step": 560
    },
    {
      "epoch": 0.333530106257379,
      "grad_norm": 23.0,
      "learning_rate": 4.6655188397336515e-05,
      "loss": 0.4393,
      "step": 565
    },
    {
      "epoch": 0.3364817001180638,
      "grad_norm": 13.625,
      "learning_rate": 4.641602189539235e-05,
      "loss": 0.5021,
      "step": 570
    },
    {
      "epoch": 0.33943329397874855,
      "grad_norm": 16.5,
      "learning_rate": 4.617535674162509e-05,
      "loss": 0.5409,
      "step": 575
    },
    {
      "epoch": 0.33943329397874855,
      "eval_loss": 0.4802681505680084,
      "eval_runtime": 22.041,
      "eval_samples_per_second": 90.74,
      "eval_steps_per_second": 90.74,
      "step": 575
    },
    {
      "epoch": 0.34238488783943327,
      "grad_norm": 18.0,
      "learning_rate": 4.59332149068431e-05,
      "loss": 0.4619,
      "step": 580
    },
    {
      "epoch": 0.34533648170011805,
      "grad_norm": 19.25,
      "learning_rate": 4.5689618496664e-05,
      "loss": 0.5019,
      "step": 585
    },
    {
      "epoch": 0.3482880755608028,
      "grad_norm": 17.25,
      "learning_rate": 4.544458974949646e-05,
      "loss": 0.3654,
      "step": 590
    },
    {
      "epoch": 0.3512396694214876,
      "grad_norm": 25.625,
      "learning_rate": 4.519815103451012e-05,
      "loss": 0.6236,
      "step": 595
    },
    {
      "epoch": 0.3541912632821724,
      "grad_norm": 30.125,
      "learning_rate": 4.4950324849593455e-05,
      "loss": 0.4846,
      "step": 600
    },
    {
      "epoch": 0.3541912632821724,
      "eval_loss": 0.47454240918159485,
      "eval_runtime": 22.0763,
      "eval_samples_per_second": 90.595,
      "eval_steps_per_second": 90.595,
      "step": 600
    },
    {
      "epoch": 0.35714285714285715,
      "grad_norm": 17.75,
      "learning_rate": 4.470113381929984e-05,
      "loss": 0.5738,
      "step": 605
    },
    {
      "epoch": 0.3600944510035419,
      "grad_norm": 16.0,
      "learning_rate": 4.445060069278218e-05,
      "loss": 0.3647,
      "step": 610
    },
    {
      "epoch": 0.3630460448642267,
      "grad_norm": 12.3125,
      "learning_rate": 4.419874834171601e-05,
      "loss": 0.4772,
      "step": 615
    },
    {
      "epoch": 0.3659976387249115,
      "grad_norm": 12.4375,
      "learning_rate": 4.3945599758211594e-05,
      "loss": 0.4309,
      "step": 620
    },
    {
      "epoch": 0.3689492325855962,
      "grad_norm": 15.4375,
      "learning_rate": 4.369117805271482e-05,
      "loss": 0.3699,
      "step": 625
    },
    {
      "epoch": 0.3689492325855962,
      "eval_loss": 0.5046902298927307,
      "eval_runtime": 22.0518,
      "eval_samples_per_second": 90.696,
      "eval_steps_per_second": 90.696,
      "step": 625
    },
    {
      "epoch": 0.371900826446281,
      "grad_norm": 26.5,
      "learning_rate": 4.343550645189751e-05,
      "loss": 0.4449,
      "step": 630
    },
    {
      "epoch": 0.37485242030696575,
      "grad_norm": 21.125,
      "learning_rate": 4.317860829653692e-05,
      "loss": 0.2902,
      "step": 635
    },
    {
      "epoch": 0.3778040141676505,
      "grad_norm": 28.875,
      "learning_rate": 4.292050703938496e-05,
      "loss": 0.5582,
      "step": 640
    },
    {
      "epoch": 0.3807556080283353,
      "grad_norm": 13.75,
      "learning_rate": 4.266122624302714e-05,
      "loss": 0.5335,
      "step": 645
    },
    {
      "epoch": 0.3837072018890201,
      "grad_norm": 13.5,
      "learning_rate": 4.2400789577731485e-05,
      "loss": 0.3741,
      "step": 650
    },
    {
      "epoch": 0.3837072018890201,
      "eval_loss": 0.4519728720188141,
      "eval_runtime": 22.0605,
      "eval_samples_per_second": 90.66,
      "eval_steps_per_second": 90.66,
      "step": 650
    },
    {
      "epoch": 0.38665879574970485,
      "grad_norm": 19.5,
      "learning_rate": 4.213922081928763e-05,
      "loss": 0.5017,
      "step": 655
    },
    {
      "epoch": 0.38961038961038963,
      "grad_norm": 13.4375,
      "learning_rate": 4.187654384683628e-05,
      "loss": 0.359,
      "step": 660
    },
    {
      "epoch": 0.3925619834710744,
      "grad_norm": 11.1875,
      "learning_rate": 4.161278264068925e-05,
      "loss": 0.2609,
      "step": 665
    },
    {
      "epoch": 0.3955135773317591,
      "grad_norm": 13.0625,
      "learning_rate": 4.134796128014022e-05,
      "loss": 0.4038,
      "step": 670
    },
    {
      "epoch": 0.3984651711924439,
      "grad_norm": 17.625,
      "learning_rate": 4.108210394126652e-05,
      "loss": 0.3166,
      "step": 675
    },
    {
      "epoch": 0.3984651711924439,
      "eval_loss": 0.44906917214393616,
      "eval_runtime": 22.0309,
      "eval_samples_per_second": 90.782,
      "eval_steps_per_second": 90.782,
      "step": 675
    },
    {
      "epoch": 0.4014167650531287,
      "grad_norm": 32.25,
      "learning_rate": 4.0815234894722035e-05,
      "loss": 0.3372,
      "step": 680
    },
    {
      "epoch": 0.40436835891381345,
      "grad_norm": 27.25,
      "learning_rate": 4.05473785035215e-05,
      "loss": 0.3346,
      "step": 685
    },
    {
      "epoch": 0.40731995277449823,
      "grad_norm": 24.75,
      "learning_rate": 4.0278559220816304e-05,
      "loss": 0.3993,
      "step": 690
    },
    {
      "epoch": 0.410271546635183,
      "grad_norm": 20.0,
      "learning_rate": 4.0008801587662194e-05,
      "loss": 0.34,
      "step": 695
    },
    {
      "epoch": 0.4132231404958678,
      "grad_norm": 27.25,
      "learning_rate": 3.9738130230778796e-05,
      "loss": 0.4442,
      "step": 700
    },
    {
      "epoch": 0.4132231404958678,
      "eval_loss": 0.4622591435909271,
      "eval_runtime": 22.0728,
      "eval_samples_per_second": 90.609,
      "eval_steps_per_second": 90.609,
      "step": 700
    },
    {
      "epoch": 0.41617473435655256,
      "grad_norm": 22.625,
      "learning_rate": 3.946656986030142e-05,
      "loss": 0.3391,
      "step": 705
    },
    {
      "epoch": 0.41912632821723733,
      "grad_norm": 26.125,
      "learning_rate": 3.919414526752524e-05,
      "loss": 0.333,
      "step": 710
    },
    {
      "epoch": 0.42207792207792205,
      "grad_norm": 7.5,
      "learning_rate": 3.8920881322642036e-05,
      "loss": 0.3791,
      "step": 715
    },
    {
      "epoch": 0.42502951593860683,
      "grad_norm": 32.0,
      "learning_rate": 3.864680297246972e-05,
      "loss": 0.5102,
      "step": 720
    },
    {
      "epoch": 0.4279811097992916,
      "grad_norm": 15.5,
      "learning_rate": 3.8371935238174924e-05,
      "loss": 0.4723,
      "step": 725
    },
    {
      "epoch": 0.4279811097992916,
      "eval_loss": 0.44943633675575256,
      "eval_runtime": 22.0672,
      "eval_samples_per_second": 90.632,
      "eval_steps_per_second": 90.632,
      "step": 725
    },
    {
      "epoch": 0.4309327036599764,
      "grad_norm": 26.875,
      "learning_rate": 3.809630321298872e-05,
      "loss": 0.4499,
      "step": 730
    },
    {
      "epoch": 0.43388429752066116,
      "grad_norm": 14.3125,
      "learning_rate": 3.78199320599159e-05,
      "loss": 0.2574,
      "step": 735
    },
    {
      "epoch": 0.43683589138134593,
      "grad_norm": 10.0625,
      "learning_rate": 3.754284700943767e-05,
      "loss": 0.3981,
      "step": 740
    },
    {
      "epoch": 0.4397874852420307,
      "grad_norm": 20.5,
      "learning_rate": 3.726507335720842e-05,
      "loss": 0.3897,
      "step": 745
    },
    {
      "epoch": 0.4427390791027155,
      "grad_norm": 12.4375,
      "learning_rate": 3.6986636461746365e-05,
      "loss": 0.3282,
      "step": 750
    },
    {
      "epoch": 0.4427390791027155,
      "eval_loss": 0.46209147572517395,
      "eval_runtime": 22.0979,
      "eval_samples_per_second": 90.506,
      "eval_steps_per_second": 90.506,
      "step": 750
    },
    {
      "epoch": 0.44569067296340026,
      "grad_norm": 17.0,
      "learning_rate": 3.6707561742118546e-05,
      "loss": 0.3905,
      "step": 755
    },
    {
      "epoch": 0.448642266824085,
      "grad_norm": 16.875,
      "learning_rate": 3.642787467562024e-05,
      "loss": 0.2688,
      "step": 760
    },
    {
      "epoch": 0.45159386068476975,
      "grad_norm": 17.625,
      "learning_rate": 3.614760079544913e-05,
      "loss": 0.369,
      "step": 765
    },
    {
      "epoch": 0.45454545454545453,
      "grad_norm": 24.0,
      "learning_rate": 3.5866765688374296e-05,
      "loss": 0.4732,
      "step": 770
    },
    {
      "epoch": 0.4574970484061393,
      "grad_norm": 16.375,
      "learning_rate": 3.558539499240037e-05,
      "loss": 0.3928,
      "step": 775
    },
    {
      "epoch": 0.4574970484061393,
      "eval_loss": 0.4500948488712311,
      "eval_runtime": 22.071,
      "eval_samples_per_second": 90.617,
      "eval_steps_per_second": 90.617,
      "step": 775
    },
    {
      "epoch": 0.4604486422668241,
      "grad_norm": 11.375,
      "learning_rate": 3.530351439442696e-05,
      "loss": 0.498,
      "step": 780
    },
    {
      "epoch": 0.46340023612750886,
      "grad_norm": 9.125,
      "learning_rate": 3.502114962790366e-05,
      "loss": 0.3545,
      "step": 785
    },
    {
      "epoch": 0.46635182998819363,
      "grad_norm": 12.375,
      "learning_rate": 3.473832647048079e-05,
      "loss": 0.5654,
      "step": 790
    },
    {
      "epoch": 0.4693034238488784,
      "grad_norm": 10.9375,
      "learning_rate": 3.445507074165612e-05,
      "loss": 0.3995,
      "step": 795
    },
    {
      "epoch": 0.4722550177095632,
      "grad_norm": 8.375,
      "learning_rate": 3.41714083004177e-05,
      "loss": 0.4174,
      "step": 800
    },
    {
      "epoch": 0.4722550177095632,
      "eval_loss": 0.4428013265132904,
      "eval_runtime": 22.0412,
      "eval_samples_per_second": 90.739,
      "eval_steps_per_second": 90.739,
      "step": 800
    },
    {
      "epoch": 0.47520661157024796,
      "grad_norm": 14.3125,
      "learning_rate": 3.3887365042883226e-05,
      "loss": 0.487,
      "step": 805
    },
    {
      "epoch": 0.4781582054309327,
      "grad_norm": 12.4375,
      "learning_rate": 3.360296689993586e-05,
      "loss": 0.367,
      "step": 810
    },
    {
      "epoch": 0.48110979929161746,
      "grad_norm": 10.375,
      "learning_rate": 3.331823983485695e-05,
      "loss": 0.2972,
      "step": 815
    },
    {
      "epoch": 0.48406139315230223,
      "grad_norm": 14.0,
      "learning_rate": 3.303320984095584e-05,
      "loss": 0.417,
      "step": 820
    },
    {
      "epoch": 0.487012987012987,
      "grad_norm": 25.0,
      "learning_rate": 3.274790293919685e-05,
      "loss": 0.3814,
      "step": 825
    },
    {
      "epoch": 0.487012987012987,
      "eval_loss": 0.45124053955078125,
      "eval_runtime": 22.0625,
      "eval_samples_per_second": 90.652,
      "eval_steps_per_second": 90.652,
      "step": 825
    },
    {
      "epoch": 0.4899645808736718,
      "grad_norm": 19.125,
      "learning_rate": 3.246234517582378e-05,
      "loss": 0.4698,
      "step": 830
    },
    {
      "epoch": 0.49291617473435656,
      "grad_norm": 19.625,
      "learning_rate": 3.217656261998208e-05,
      "loss": 0.4826,
      "step": 835
    },
    {
      "epoch": 0.49586776859504134,
      "grad_norm": 9.25,
      "learning_rate": 3.189058136133898e-05,
      "loss": 0.4592,
      "step": 840
    },
    {
      "epoch": 0.4988193624557261,
      "grad_norm": 7.40625,
      "learning_rate": 3.1604427507701675e-05,
      "loss": 0.551,
      "step": 845
    },
    {
      "epoch": 0.5017709563164109,
      "grad_norm": 22.25,
      "learning_rate": 3.131812718263392e-05,
      "loss": 0.3845,
      "step": 850
    },
    {
      "epoch": 0.5017709563164109,
      "eval_loss": 0.42552411556243896,
      "eval_runtime": 22.0632,
      "eval_samples_per_second": 90.649,
      "eval_steps_per_second": 90.649,
      "step": 850
    },
    {
      "epoch": 0.5047225501770957,
      "grad_norm": 11.875,
      "learning_rate": 3.1031706523071115e-05,
      "loss": 0.5727,
      "step": 855
    },
    {
      "epoch": 0.5076741440377804,
      "grad_norm": 5.21875,
      "learning_rate": 3.0745191676934285e-05,
      "loss": 0.4759,
      "step": 860
    },
    {
      "epoch": 0.5106257378984652,
      "grad_norm": 20.75,
      "learning_rate": 3.0458608800742883e-05,
      "loss": 0.4727,
      "step": 865
    },
    {
      "epoch": 0.51357733175915,
      "grad_norm": 19.125,
      "learning_rate": 3.0171984057227008e-05,
      "loss": 0.4733,
      "step": 870
    },
    {
      "epoch": 0.5165289256198347,
      "grad_norm": 21.125,
      "learning_rate": 2.988534361293888e-05,
      "loss": 0.3789,
      "step": 875
    },
    {
      "epoch": 0.5165289256198347,
      "eval_loss": 0.41610679030418396,
      "eval_runtime": 22.0756,
      "eval_samples_per_second": 90.598,
      "eval_steps_per_second": 90.598,
      "step": 875
    },
    {
      "epoch": 0.5194805194805194,
      "grad_norm": 10.8125,
      "learning_rate": 2.959871363586411e-05,
      "loss": 0.4258,
      "step": 880
    },
    {
      "epoch": 0.5224321133412042,
      "grad_norm": 9.0625,
      "learning_rate": 2.9312120293032703e-05,
      "loss": 0.408,
      "step": 885
    },
    {
      "epoch": 0.525383707201889,
      "grad_norm": 29.375,
      "learning_rate": 2.902558974813026e-05,
      "loss": 0.5145,
      "step": 890
    },
    {
      "epoch": 0.5283353010625738,
      "grad_norm": 14.9375,
      "learning_rate": 2.873914815910944e-05,
      "loss": 0.4281,
      "step": 895
    },
    {
      "epoch": 0.5312868949232585,
      "grad_norm": 19.125,
      "learning_rate": 2.8452821675801944e-05,
      "loss": 0.3189,
      "step": 900
    },
    {
      "epoch": 0.5312868949232585,
      "eval_loss": 0.41652774810791016,
      "eval_runtime": 22.0235,
      "eval_samples_per_second": 90.812,
      "eval_steps_per_second": 90.812,
      "step": 900
    },
    {
      "epoch": 0.5342384887839433,
      "grad_norm": 19.375,
      "learning_rate": 2.81666364375312e-05,
      "loss": 0.4598,
      "step": 905
    },
    {
      "epoch": 0.5371900826446281,
      "grad_norm": 13.4375,
      "learning_rate": 2.7880618570726142e-05,
      "loss": 0.4811,
      "step": 910
    },
    {
      "epoch": 0.5401416765053129,
      "grad_norm": 23.0,
      "learning_rate": 2.7594794186535993e-05,
      "loss": 0.4931,
      "step": 915
    },
    {
      "epoch": 0.5430932703659976,
      "grad_norm": 24.125,
      "learning_rate": 2.7309189378446578e-05,
      "loss": 0.5955,
      "step": 920
    },
    {
      "epoch": 0.5460448642266824,
      "grad_norm": 19.875,
      "learning_rate": 2.702383021989817e-05,
      "loss": 0.3743,
      "step": 925
    },
    {
      "epoch": 0.5460448642266824,
      "eval_loss": 0.41767558455467224,
      "eval_runtime": 22.0719,
      "eval_samples_per_second": 90.613,
      "eval_steps_per_second": 90.613,
      "step": 925
    }
  ],
  "logging_steps": 5,
  "max_steps": 1694,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 25,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.3649613399392256e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}