{
  "best_global_step": 377,
  "best_metric": 0.15596222877502441,
  "best_model_checkpoint": "saves/p-tuning/llama-3-8b-instruct/train_cb_1754652158/checkpoint-377",
  "epoch": 10.0,
  "eval_steps": 29,
  "global_step": 570,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.08771929824561403,
      "grad_norm": 59.72950744628906,
      "learning_rate": 3.5087719298245615e-06,
      "loss": 3.7311,
      "num_input_tokens_seen": 3552,
      "step": 5
    },
    {
      "epoch": 0.17543859649122806,
      "grad_norm": 85.96363067626953,
      "learning_rate": 7.894736842105263e-06,
      "loss": 3.2007,
      "num_input_tokens_seen": 7264,
      "step": 10
    },
    {
      "epoch": 0.2631578947368421,
      "grad_norm": 155.43385314941406,
      "learning_rate": 1.2280701754385964e-05,
      "loss": 2.1438,
      "num_input_tokens_seen": 10528,
      "step": 15
    },
    {
      "epoch": 0.3508771929824561,
      "grad_norm": 61.96742630004883,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 0.8229,
      "num_input_tokens_seen": 14720,
      "step": 20
    },
    {
      "epoch": 0.43859649122807015,
      "grad_norm": 62.340431213378906,
      "learning_rate": 2.105263157894737e-05,
      "loss": 0.7984,
      "num_input_tokens_seen": 18016,
      "step": 25
    },
    {
      "epoch": 0.5087719298245614,
      "eval_loss": 0.7324875593185425,
      "eval_runtime": 0.8441,
      "eval_samples_per_second": 29.617,
      "eval_steps_per_second": 8.293,
      "num_input_tokens_seen": 20064,
      "step": 29
    },
    {
      "epoch": 0.5263157894736842,
      "grad_norm": 128.35720825195312,
      "learning_rate": 2.5438596491228074e-05,
      "loss": 1.0221,
      "num_input_tokens_seen": 20640,
      "step": 30
    },
    {
      "epoch": 0.6140350877192983,
      "grad_norm": 25.795623779296875,
      "learning_rate": 2.9824561403508772e-05,
      "loss": 0.4402,
      "num_input_tokens_seen": 24800,
      "step": 35
    },
    {
      "epoch": 0.7017543859649122,
      "grad_norm": 11.380049705505371,
      "learning_rate": 3.421052631578947e-05,
      "loss": 0.3711,
      "num_input_tokens_seen": 28064,
      "step": 40
    },
    {
      "epoch": 0.7894736842105263,
      "grad_norm": 31.992555618286133,
      "learning_rate": 3.859649122807018e-05,
      "loss": 0.3899,
      "num_input_tokens_seen": 30944,
      "step": 45
    },
    {
      "epoch": 0.8771929824561403,
      "grad_norm": 26.026948928833008,
      "learning_rate": 4.298245614035088e-05,
      "loss": 0.5316,
      "num_input_tokens_seen": 33664,
      "step": 50
    },
    {
      "epoch": 0.9649122807017544,
      "grad_norm": 322.9638977050781,
      "learning_rate": 4.736842105263158e-05,
      "loss": 1.5006,
      "num_input_tokens_seen": 36320,
      "step": 55
    },
    {
      "epoch": 1.0175438596491229,
      "eval_loss": 0.5227728486061096,
      "eval_runtime": 0.8526,
      "eval_samples_per_second": 29.324,
      "eval_steps_per_second": 8.211,
      "num_input_tokens_seen": 37832,
      "step": 58
    },
    {
      "epoch": 1.0526315789473684,
      "grad_norm": 67.29098510742188,
      "learning_rate": 4.999812487773597e-05,
      "loss": 0.349,
      "num_input_tokens_seen": 39080,
      "step": 60
    },
    {
      "epoch": 1.1403508771929824,
      "grad_norm": 40.01487731933594,
      "learning_rate": 4.997703298253406e-05,
      "loss": 0.9668,
      "num_input_tokens_seen": 42536,
      "step": 65
    },
    {
      "epoch": 1.2280701754385965,
      "grad_norm": 18.30699920654297,
      "learning_rate": 4.993252512887069e-05,
      "loss": 0.7583,
      "num_input_tokens_seen": 45608,
      "step": 70
    },
    {
      "epoch": 1.3157894736842106,
      "grad_norm": 163.50320434570312,
      "learning_rate": 4.986464304284091e-05,
      "loss": 1.4897,
      "num_input_tokens_seen": 49352,
      "step": 75
    },
    {
      "epoch": 1.4035087719298245,
      "grad_norm": 44.9167594909668,
      "learning_rate": 4.977345036387331e-05,
      "loss": 0.4684,
      "num_input_tokens_seen": 52328,
      "step": 80
    },
    {
      "epoch": 1.4912280701754386,
      "grad_norm": 3.2939820289611816,
      "learning_rate": 4.965903258506806e-05,
      "loss": 0.3837,
      "num_input_tokens_seen": 56328,
      "step": 85
    },
    {
      "epoch": 1.526315789473684,
      "eval_loss": 0.5208507776260376,
      "eval_runtime": 0.853,
      "eval_samples_per_second": 29.308,
      "eval_steps_per_second": 8.206,
      "num_input_tokens_seen": 57288,
      "step": 87
    },
    {
      "epoch": 1.5789473684210527,
      "grad_norm": 15.485786437988281,
      "learning_rate": 4.952149697304716e-05,
      "loss": 0.634,
      "num_input_tokens_seen": 59048,
      "step": 90
    },
    {
      "epoch": 1.6666666666666665,
      "grad_norm": 4.415175437927246,
      "learning_rate": 4.9360972467392056e-05,
      "loss": 0.2586,
      "num_input_tokens_seen": 62504,
      "step": 95
    },
    {
      "epoch": 1.7543859649122808,
      "grad_norm": 12.429476737976074,
      "learning_rate": 4.917760955976277e-05,
      "loss": 0.2382,
      "num_input_tokens_seen": 65832,
      "step": 100
    },
    {
      "epoch": 1.8421052631578947,
      "grad_norm": 9.031868934631348,
      "learning_rate": 4.897158015281209e-05,
      "loss": 0.2652,
      "num_input_tokens_seen": 68808,
      "step": 105
    },
    {
      "epoch": 1.9298245614035088,
      "grad_norm": 6.544041633605957,
      "learning_rate": 4.874307739902689e-05,
      "loss": 0.233,
      "num_input_tokens_seen": 71848,
      "step": 110
    },
    {
      "epoch": 2.017543859649123,
      "grad_norm": 2.9446492195129395,
      "learning_rate": 4.849231551964771e-05,
      "loss": 0.2138,
      "num_input_tokens_seen": 74040,
      "step": 115
    },
    {
      "epoch": 2.0350877192982457,
      "eval_loss": 0.23195452988147736,
      "eval_runtime": 0.8523,
      "eval_samples_per_second": 29.332,
      "eval_steps_per_second": 8.213,
      "num_input_tokens_seen": 74520,
      "step": 116
    },
    {
      "epoch": 2.1052631578947367,
      "grad_norm": 20.17977523803711,
      "learning_rate": 4.821952960383649e-05,
      "loss": 0.1899,
      "num_input_tokens_seen": 77272,
      "step": 120
    },
    {
      "epoch": 2.192982456140351,
      "grad_norm": 9.810575485229492,
      "learning_rate": 4.7924975388280524e-05,
      "loss": 0.2362,
      "num_input_tokens_seen": 80280,
      "step": 125
    },
    {
      "epoch": 2.280701754385965,
      "grad_norm": 12.129171371459961,
      "learning_rate": 4.760892901743944e-05,
      "loss": 0.3055,
      "num_input_tokens_seen": 83480,
      "step": 130
    },
    {
      "epoch": 2.3684210526315788,
      "grad_norm": 9.584630966186523,
      "learning_rate": 4.727168678465988e-05,
      "loss": 0.4557,
      "num_input_tokens_seen": 86232,
      "step": 135
    },
    {
      "epoch": 2.456140350877193,
      "grad_norm": 5.297253131866455,
      "learning_rate": 4.6913564854400595e-05,
      "loss": 0.2885,
      "num_input_tokens_seen": 89656,
      "step": 140
    },
    {
      "epoch": 2.543859649122807,
      "grad_norm": 5.190830707550049,
      "learning_rate": 4.6534898965828405e-05,
      "loss": 0.342,
      "num_input_tokens_seen": 93080,
      "step": 145
    },
    {
      "epoch": 2.543859649122807,
      "eval_loss": 0.24903666973114014,
      "eval_runtime": 0.8518,
      "eval_samples_per_second": 29.349,
      "eval_steps_per_second": 8.218,
      "num_input_tokens_seen": 93080,
      "step": 145
    },
    {
      "epoch": 2.6315789473684212,
      "grad_norm": 6.145389080047607,
      "learning_rate": 4.613604411806285e-05,
      "loss": 0.2368,
      "num_input_tokens_seen": 96440,
      "step": 150
    },
    {
      "epoch": 2.719298245614035,
      "grad_norm": 14.407633781433105,
      "learning_rate": 4.5717374237364665e-05,
      "loss": 0.2185,
      "num_input_tokens_seen": 100280,
      "step": 155
    },
    {
      "epoch": 2.807017543859649,
      "grad_norm": 13.796713829040527,
      "learning_rate": 4.5279281826580056e-05,
      "loss": 0.4247,
      "num_input_tokens_seen": 103512,
      "step": 160
    },
    {
      "epoch": 2.8947368421052633,
      "grad_norm": 5.334498405456543,
      "learning_rate": 4.482217759716946e-05,
      "loss": 0.3187,
      "num_input_tokens_seen": 106168,
      "step": 165
    },
    {
      "epoch": 2.982456140350877,
      "grad_norm": 2.5159807205200195,
      "learning_rate": 4.434649008416565e-05,
      "loss": 0.231,
      "num_input_tokens_seen": 109624,
      "step": 170
    },
    {
      "epoch": 3.0526315789473686,
      "eval_loss": 0.16980019211769104,
      "eval_runtime": 0.8481,
      "eval_samples_per_second": 29.479,
      "eval_steps_per_second": 8.254,
      "num_input_tokens_seen": 111928,
      "step": 174
    },
    {
      "epoch": 3.0701754385964914,
      "grad_norm": 9.546319007873535,
      "learning_rate": 4.385266524442241e-05,
      "loss": 0.1977,
      "num_input_tokens_seen": 112472,
      "step": 175
    },
    {
      "epoch": 3.1578947368421053,
      "grad_norm": 6.8020195960998535,
      "learning_rate": 4.334116603853007e-05,
      "loss": 0.31,
      "num_input_tokens_seen": 115576,
      "step": 180
    },
    {
      "epoch": 3.245614035087719,
      "grad_norm": 167.5514678955078,
      "learning_rate": 4.2812471996790206e-05,
      "loss": 0.3518,
      "num_input_tokens_seen": 119000,
      "step": 185
    },
    {
      "epoch": 3.3333333333333335,
      "grad_norm": 6.39103364944458,
      "learning_rate": 4.226707876965611e-05,
      "loss": 0.2729,
      "num_input_tokens_seen": 122360,
      "step": 190
    },
    {
      "epoch": 3.4210526315789473,
      "grad_norm": 7.074615955352783,
      "learning_rate": 4.1705497663060767e-05,
      "loss": 0.2196,
      "num_input_tokens_seen": 125496,
      "step": 195
    },
    {
      "epoch": 3.5087719298245617,
      "grad_norm": 5.8528361320495605,
      "learning_rate": 4.1128255159067665e-05,
      "loss": 0.3391,
      "num_input_tokens_seen": 128760,
      "step": 200
    },
    {
      "epoch": 3.56140350877193,
      "eval_loss": 0.2744066119194031,
      "eval_runtime": 0.8498,
      "eval_samples_per_second": 29.417,
      "eval_steps_per_second": 8.237,
      "num_input_tokens_seen": 131160,
      "step": 203
    },
    {
      "epoch": 3.5964912280701755,
      "grad_norm": 15.246909141540527,
      "learning_rate": 4.053589242229412e-05,
      "loss": 0.2077,
      "num_input_tokens_seen": 132248,
      "step": 205
    },
    {
      "epoch": 3.6842105263157894,
      "grad_norm": 3.061079263687134,
      "learning_rate": 3.9928964792569655e-05,
      "loss": 0.2695,
      "num_input_tokens_seen": 135160,
      "step": 210
    },
    {
      "epoch": 3.7719298245614032,
      "grad_norm": 3.2342686653137207,
      "learning_rate": 3.930804126430513e-05,
      "loss": 0.2965,
      "num_input_tokens_seen": 138200,
      "step": 215
    },
    {
      "epoch": 3.8596491228070176,
      "grad_norm": 5.967337608337402,
      "learning_rate": 3.867370395306068e-05,
      "loss": 0.1714,
      "num_input_tokens_seen": 141752,
      "step": 220
    },
    {
      "epoch": 3.9473684210526314,
      "grad_norm": 11.130901336669922,
      "learning_rate": 3.8026547549812665e-05,
      "loss": 0.3448,
      "num_input_tokens_seen": 146008,
      "step": 225
    },
    {
      "epoch": 4.035087719298246,
      "grad_norm": 3.796250104904175,
      "learning_rate": 3.736717876343106e-05,
      "loss": 0.2601,
      "num_input_tokens_seen": 148648,
      "step": 230
    },
    {
      "epoch": 4.0701754385964914,
      "eval_loss": 0.16180922091007233,
      "eval_runtime": 0.8511,
      "eval_samples_per_second": 29.374,
      "eval_steps_per_second": 8.225,
      "num_input_tokens_seen": 150056,
      "step": 232
    },
    {
      "epoch": 4.12280701754386,
      "grad_norm": 5.916079521179199,
      "learning_rate": 3.66962157518902e-05,
      "loss": 0.2069,
      "num_input_tokens_seen": 151656,
      "step": 235
    },
    {
      "epoch": 4.2105263157894735,
      "grad_norm": 3.373704671859741,
      "learning_rate": 3.601428754274584e-05,
      "loss": 0.1738,
      "num_input_tokens_seen": 154280,
      "step": 240
    },
    {
      "epoch": 4.298245614035087,
      "grad_norm": 7.352924346923828,
      "learning_rate": 3.532203344342212e-05,
      "loss": 0.2579,
      "num_input_tokens_seen": 157160,
      "step": 245
    },
    {
      "epoch": 4.385964912280702,
      "grad_norm": 5.744953632354736,
      "learning_rate": 3.4620102441861143e-05,
      "loss": 0.227,
      "num_input_tokens_seen": 160072,
      "step": 250
    },
    {
      "epoch": 4.473684210526316,
      "grad_norm": 8.015254020690918,
      "learning_rate": 3.390915259809696e-05,
      "loss": 0.1874,
      "num_input_tokens_seen": 163752,
      "step": 255
    },
    {
      "epoch": 4.56140350877193,
      "grad_norm": 7.88545560836792,
      "learning_rate": 3.318985042732461e-05,
      "loss": 0.1828,
      "num_input_tokens_seen": 166600,
      "step": 260
    },
    {
      "epoch": 4.578947368421053,
      "eval_loss": 0.25386178493499756,
      "eval_runtime": 0.8502,
      "eval_samples_per_second": 29.403,
      "eval_steps_per_second": 8.233,
      "num_input_tokens_seen": 167208,
      "step": 261
    },
    {
      "epoch": 4.649122807017544,
      "grad_norm": 4.994500160217285,
      "learning_rate": 3.246287027504237e-05,
      "loss": 0.2042,
      "num_input_tokens_seen": 169704,
      "step": 265
    },
    {
      "epoch": 4.7368421052631575,
      "grad_norm": 5.937520503997803,
      "learning_rate": 3.172889368485311e-05,
      "loss": 0.1841,
      "num_input_tokens_seen": 173000,
      "step": 270
    },
    {
      "epoch": 4.824561403508772,
      "grad_norm": 10.352025032043457,
      "learning_rate": 3.0988608759517475e-05,
      "loss": 0.2559,
      "num_input_tokens_seen": 177128,
      "step": 275
    },
    {
      "epoch": 4.912280701754386,
      "grad_norm": 16.774810791015625,
      "learning_rate": 3.0242709515857758e-05,
      "loss": 0.2189,
      "num_input_tokens_seen": 180680,
      "step": 280
    },
    {
      "epoch": 5.0,
      "grad_norm": 19.19439697265625,
      "learning_rate": 2.949189523411747e-05,
      "loss": 0.1958,
      "num_input_tokens_seen": 183280,
      "step": 285
    },
    {
      "epoch": 5.087719298245614,
      "grad_norm": 9.190163612365723,
      "learning_rate": 2.8736869802386364e-05,
      "loss": 0.3148,
      "num_input_tokens_seen": 186160,
      "step": 290
    },
    {
      "epoch": 5.087719298245614,
      "eval_loss": 0.25446251034736633,
      "eval_runtime": 0.8549,
      "eval_samples_per_second": 29.243,
      "eval_steps_per_second": 8.188,
      "num_input_tokens_seen": 186160,
      "step": 290
    },
    {
      "epoch": 5.175438596491228,
      "grad_norm": 6.987154960632324,
      "learning_rate": 2.797834105670559e-05,
      "loss": 0.203,
      "num_input_tokens_seen": 189584,
      "step": 295
    },
    {
      "epoch": 5.2631578947368425,
      "grad_norm": 7.0422539710998535,
      "learning_rate": 2.7217020117471793e-05,
      "loss": 0.2338,
      "num_input_tokens_seen": 193296,
      "step": 300
    },
    {
      "epoch": 5.350877192982456,
      "grad_norm": 6.827476501464844,
      "learning_rate": 2.6453620722761896e-05,
      "loss": 0.2074,
      "num_input_tokens_seen": 197328,
      "step": 305
    },
    {
      "epoch": 5.43859649122807,
      "grad_norm": 2.4718358516693115,
      "learning_rate": 2.5688858559204053e-05,
      "loss": 0.1303,
      "num_input_tokens_seen": 200560,
      "step": 310
    },
    {
      "epoch": 5.526315789473684,
      "grad_norm": 12.62780475616455,
      "learning_rate": 2.492345059102164e-05,
      "loss": 0.1062,
      "num_input_tokens_seen": 203632,
      "step": 315
    },
    {
      "epoch": 5.5964912280701755,
      "eval_loss": 0.21558518707752228,
      "eval_runtime": 0.8533,
      "eval_samples_per_second": 29.299,
      "eval_steps_per_second": 8.204,
      "num_input_tokens_seen": 206000,
      "step": 319
    },
    {
      "epoch": 5.614035087719298,
      "grad_norm": 5.2489752769470215,
      "learning_rate": 2.4158114387879616e-05,
      "loss": 0.1738,
      "num_input_tokens_seen": 206480,
      "step": 320
    },
    {
      "epoch": 5.701754385964913,
      "grad_norm": 9.622069358825684,
      "learning_rate": 2.3393567452163252e-05,
      "loss": 0.2403,
      "num_input_tokens_seen": 209232,
      "step": 325
    },
    {
      "epoch": 5.7894736842105265,
      "grad_norm": 2.9533114433288574,
      "learning_rate": 2.2630526546319914e-05,
      "loss": 0.3923,
      "num_input_tokens_seen": 213168,
      "step": 330
    },
    {
      "epoch": 5.87719298245614,
      "grad_norm": 6.1617937088012695,
      "learning_rate": 2.186970702089457e-05,
      "loss": 0.159,
      "num_input_tokens_seen": 216752,
      "step": 335
    },
    {
      "epoch": 5.964912280701754,
      "grad_norm": 6.799952507019043,
      "learning_rate": 2.111182214388893e-05,
      "loss": 0.1847,
      "num_input_tokens_seen": 219536,
      "step": 340
    },
    {
      "epoch": 6.052631578947368,
      "grad_norm": 6.728596210479736,
      "learning_rate": 2.0357582432072957e-05,
      "loss": 0.0974,
      "num_input_tokens_seen": 222272,
      "step": 345
    },
    {
      "epoch": 6.105263157894737,
      "eval_loss": 0.2205836921930313,
      "eval_runtime": 0.8503,
      "eval_samples_per_second": 29.402,
      "eval_steps_per_second": 8.233,
      "num_input_tokens_seen": 224064,
      "step": 348
    },
    {
      "epoch": 6.140350877192983,
      "grad_norm": 3.4464352130889893,
      "learning_rate": 1.9607694984875754e-05,
      "loss": 0.3393,
      "num_input_tokens_seen": 225664,
      "step": 350
    },
    {
      "epoch": 6.228070175438597,
      "grad_norm": 4.825738906860352,
      "learning_rate": 1.8862862821480025e-05,
      "loss": 0.1941,
      "num_input_tokens_seen": 229408,
      "step": 355
    },
    {
      "epoch": 6.315789473684211,
      "grad_norm": 4.794721603393555,
      "learning_rate": 1.8123784221741964e-05,
      "loss": 0.3113,
      "num_input_tokens_seen": 232832,
      "step": 360
    },
    {
      "epoch": 6.4035087719298245,
      "grad_norm": 2.929058313369751,
      "learning_rate": 1.73911520715541e-05,
      "loss": 0.0671,
      "num_input_tokens_seen": 236288,
      "step": 365
    },
    {
      "epoch": 6.491228070175438,
      "grad_norm": 0.9531017541885376,
      "learning_rate": 1.666565321326512e-05,
      "loss": 0.1522,
      "num_input_tokens_seen": 239296,
      "step": 370
    },
    {
      "epoch": 6.578947368421053,
      "grad_norm": 10.87993049621582,
      "learning_rate": 1.5947967801765345e-05,
      "loss": 0.1758,
      "num_input_tokens_seen": 242848,
      "step": 375
    },
    {
      "epoch": 6.614035087719298,
      "eval_loss": 0.15596222877502441,
      "eval_runtime": 0.8511,
      "eval_samples_per_second": 29.374,
      "eval_steps_per_second": 8.225,
      "num_input_tokens_seen": 243840,
      "step": 377
    },
    {
      "epoch": 6.666666666666667,
      "grad_norm": 6.758040904998779,
      "learning_rate": 1.5238768666841907e-05,
      "loss": 0.1513,
      "num_input_tokens_seen": 245344,
      "step": 380
    },
    {
      "epoch": 6.754385964912281,
      "grad_norm": 4.473215103149414,
      "learning_rate": 1.4538720682400969e-05,
      "loss": 0.2004,
      "num_input_tokens_seen": 248256,
      "step": 385
    },
    {
      "epoch": 6.842105263157895,
      "grad_norm": 5.833415985107422,
      "learning_rate": 1.3848480143148839e-05,
      "loss": 0.2411,
      "num_input_tokens_seen": 250976,
      "step": 390
    },
    {
      "epoch": 6.9298245614035086,
      "grad_norm": 4.30411958694458,
      "learning_rate": 1.3168694149315796e-05,
      "loss": 0.075,
      "num_input_tokens_seen": 254944,
      "step": 395
    },
    {
      "epoch": 7.017543859649122,
      "grad_norm": 2.6023566722869873,
      "learning_rate": 1.2500000000000006e-05,
      "loss": 0.0674,
      "num_input_tokens_seen": 257888,
      "step": 400
    },
    {
      "epoch": 7.105263157894737,
      "grad_norm": 1.8268622159957886,
      "learning_rate": 1.1843024595699805e-05,
      "loss": 0.0921,
      "num_input_tokens_seen": 261088,
      "step": 405
    },
    {
      "epoch": 7.12280701754386,
      "eval_loss": 0.19613634049892426,
      "eval_runtime": 0.849,
      "eval_samples_per_second": 29.446,
      "eval_steps_per_second": 8.245,
      "num_input_tokens_seen": 261504,
      "step": 406
    },
    {
      "epoch": 7.192982456140351,
      "grad_norm": 10.325926780700684,
      "learning_rate": 1.1198383850594758e-05,
      "loss": 0.0531,
      "num_input_tokens_seen": 263840,
      "step": 410
    },
    {
      "epoch": 7.280701754385965,
      "grad_norm": 12.008359909057617,
      "learning_rate": 1.0566682115126344e-05,
      "loss": 0.2284,
      "num_input_tokens_seen": 267424,
      "step": 415
    },
    {
      "epoch": 7.368421052631579,
      "grad_norm": 9.924817085266113,
      "learning_rate": 9.948511609419675e-06,
      "loss": 0.3437,
      "num_input_tokens_seen": 270720,
      "step": 420
    },
    {
      "epoch": 7.456140350877193,
      "grad_norm": 2.3891053199768066,
      "learning_rate": 9.344451868077353e-06,
      "loss": 0.0494,
      "num_input_tokens_seen": 273504,
      "step": 425
    },
    {
      "epoch": 7.543859649122807,
      "grad_norm": 6.745932102203369,
      "learning_rate": 8.755069196866014e-06,
      "loss": 0.1256,
      "num_input_tokens_seen": 276608,
      "step": 430
    },
    {
      "epoch": 7.631578947368421,
      "grad_norm": 11.279396057128906,
      "learning_rate": 8.180916141804906e-06,
      "loss": 0.2705,
      "num_input_tokens_seen": 280352,
      "step": 435
    },
    {
      "epoch": 7.631578947368421,
      "eval_loss": 0.23887619376182556,
      "eval_runtime": 0.8494,
      "eval_samples_per_second": 29.432,
      "eval_steps_per_second": 8.241,
      "num_input_tokens_seen": 280352,
      "step": 435
    },
    {
      "epoch": 7.719298245614035,
      "grad_norm": 3.9152004718780518,
      "learning_rate": 7.622530971154199e-06,
      "loss": 0.2732,
      "num_input_tokens_seen": 283808,
      "step": 440
    },
    {
      "epoch": 7.807017543859649,
      "grad_norm": 8.49035930633545,
      "learning_rate": 7.080437170788723e-06,
      "loss": 0.1001,
      "num_input_tokens_seen": 286688,
      "step": 445
    },
    {
      "epoch": 7.894736842105263,
      "grad_norm": 5.865164279937744,
      "learning_rate": 6.555142953430158e-06,
      "loss": 0.1136,
      "num_input_tokens_seen": 290240,
      "step": 450
    },
    {
      "epoch": 7.982456140350877,
      "grad_norm": 0.9978966116905212,
      "learning_rate": 6.0471407821978135e-06,
      "loss": 0.1173,
      "num_input_tokens_seen": 293568,
      "step": 455
    },
    {
      "epoch": 8.070175438596491,
      "grad_norm": 5.051326751708984,
      "learning_rate": 5.556906908924655e-06,
      "loss": 0.0502,
      "num_input_tokens_seen": 296720,
      "step": 460
    },
    {
      "epoch": 8.140350877192983,
      "eval_loss": 0.16461919248104095,
      "eval_runtime": 0.8517,
      "eval_samples_per_second": 29.353,
      "eval_steps_per_second": 8.219,
      "num_input_tokens_seen": 299344,
      "step": 464
    },
    {
      "epoch": 8.157894736842104,
      "grad_norm": 1.7852911949157715,
      "learning_rate": 5.084900927671393e-06,
      "loss": 0.3081,
      "num_input_tokens_seen": 300112,
      "step": 465
    },
    {
      "epoch": 8.24561403508772,
      "grad_norm": 1.0807344913482666,
      "learning_rate": 4.631565343857239e-06,
      "loss": 0.0428,
      "num_input_tokens_seen": 303664,
      "step": 470
    },
    {
      "epoch": 8.333333333333334,
      "grad_norm": 5.201423645019531,
      "learning_rate": 4.19732515941125e-06,
      "loss": 0.0843,
      "num_input_tokens_seen": 306480,
      "step": 475
    },
    {
      "epoch": 8.421052631578947,
      "grad_norm": 19.143348693847656,
      "learning_rate": 3.7825874743331907e-06,
      "loss": 0.1734,
      "num_input_tokens_seen": 309744,
      "step": 480
    },
    {
      "epoch": 8.508771929824562,
      "grad_norm": 1.025292992591858,
      "learning_rate": 3.3877411050374424e-06,
      "loss": 0.1019,
      "num_input_tokens_seen": 313360,
      "step": 485
    },
    {
      "epoch": 8.596491228070175,
      "grad_norm": 5.185237884521484,
      "learning_rate": 3.013156219837776e-06,
      "loss": 0.1646,
      "num_input_tokens_seen": 316784,
      "step": 490
    },
    {
      "epoch": 8.649122807017545,
      "eval_loss": 0.2208017259836197,
      "eval_runtime": 0.8544,
      "eval_samples_per_second": 29.261,
      "eval_steps_per_second": 8.193,
      "num_input_tokens_seen": 318672,
      "step": 493
    },
    {
      "epoch": 8.68421052631579,
      "grad_norm": 0.6018063426017761,
      "learning_rate": 2.659183991914696e-06,
      "loss": 0.0134,
      "num_input_tokens_seen": 319760,
      "step": 495
    },
    {
      "epoch": 8.771929824561404,
      "grad_norm": 1.9732301235198975,
      "learning_rate": 2.326156270090735e-06,
      "loss": 0.0183,
      "num_input_tokens_seen": 322512,
      "step": 500
    },
    {
      "epoch": 8.859649122807017,
      "grad_norm": 0.3454095423221588,
      "learning_rate": 2.0143852677223075e-06,
      "loss": 0.163,
      "num_input_tokens_seen": 325936,
      "step": 505
    },
    {
      "epoch": 8.947368421052632,
      "grad_norm": 6.823419570922852,
      "learning_rate": 1.7241632699998123e-06,
      "loss": 0.234,
      "num_input_tokens_seen": 329360,
      "step": 510
    },
    {
      "epoch": 9.035087719298245,
      "grad_norm": 0.5328010320663452,
      "learning_rate": 1.4557623599303903e-06,
      "loss": 0.1047,
      "num_input_tokens_seen": 332168,
      "step": 515
    },
    {
      "epoch": 9.12280701754386,
      "grad_norm": 8.00861644744873,
      "learning_rate": 1.2094341632602064e-06,
      "loss": 0.1549,
      "num_input_tokens_seen": 336296,
      "step": 520
    },
    {
      "epoch": 9.157894736842104,
      "eval_loss": 0.19831164181232452,
      "eval_runtime": 0.8581,
      "eval_samples_per_second": 29.136,
      "eval_steps_per_second": 8.158,
      "num_input_tokens_seen": 337480,
      "step": 522
    },
    {
      "epoch": 9.210526315789474,
      "grad_norm": 10.439408302307129,
      "learning_rate": 9.85409612575411e-07,
      "loss": 0.1365,
      "num_input_tokens_seen": 339080,
      "step": 525
    },
    {
      "epoch": 9.298245614035087,
      "grad_norm": 7.713301181793213,
      "learning_rate": 7.838987308029427e-07,
      "loss": 0.0859,
      "num_input_tokens_seen": 342568,
      "step": 530
    },
    {
      "epoch": 9.385964912280702,
      "grad_norm": 0.42152801156044006,
      "learning_rate": 6.050904343141095e-07,
      "loss": 0.067,
      "num_input_tokens_seen": 345576,
      "step": 535
    },
    {
      "epoch": 9.473684210526315,
      "grad_norm": 18.922243118286133,
      "learning_rate": 4.491523558155714e-07,
      "loss": 0.1141,
      "num_input_tokens_seen": 348392,
      "step": 540
    },
    {
      "epoch": 9.56140350877193,
      "grad_norm": 6.063596725463867,
      "learning_rate": 3.162306871937387e-07,
      "loss": 0.0688,
      "num_input_tokens_seen": 351912,
      "step": 545
    },
    {
      "epoch": 9.649122807017545,
      "grad_norm": 0.5159357786178589,
      "learning_rate": 2.064500424599436e-07,
      "loss": 0.072,
      "num_input_tokens_seen": 355432,
      "step": 550
    },
    {
      "epoch": 9.666666666666666,
      "eval_loss": 0.2247486412525177,
      "eval_runtime": 0.8497,
      "eval_samples_per_second": 29.422,
      "eval_steps_per_second": 8.238,
      "num_input_tokens_seen": 356456,
      "step": 551
    },
    {
      "epoch": 9.736842105263158,
      "grad_norm": 1.148868203163147,
      "learning_rate": 1.1991334092484318e-07,
      "loss": 0.0824,
      "num_input_tokens_seen": 358760,
      "step": 555
    },
    {
      "epoch": 9.824561403508772,
      "grad_norm": 1.0751643180847168,
      "learning_rate": 5.6701710711626334e-08,
      "loss": 0.141,
      "num_input_tokens_seen": 362088,
      "step": 560
    },
    {
      "epoch": 9.912280701754385,
      "grad_norm": 4.292356491088867,
      "learning_rate": 1.6874412698408836e-08,
      "loss": 0.0604,
      "num_input_tokens_seen": 365544,
      "step": 565
    },
    {
      "epoch": 10.0,
      "grad_norm": 0.1350361406803131,
      "learning_rate": 4.687849611939576e-10,
      "loss": 0.3128,
      "num_input_tokens_seen": 367864,
      "step": 570
    },
    {
      "epoch": 10.0,
      "num_input_tokens_seen": 367864,
      "step": 570,
      "total_flos": 1.6564749657243648e+16,
      "train_loss": 0.34152856785477254,
      "train_runtime": 183.4948,
      "train_samples_per_second": 12.262,
      "train_steps_per_second": 3.106
    }
  ],
  "logging_steps": 5,
  "max_steps": 570,
  "num_input_tokens_seen": 367864,
  "num_train_epochs": 10,
  "save_steps": 29,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.6564749657243648e+16,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}