{
  "best_global_step": 12200,
  "best_metric": 1.0173370838165283,
  "best_model_checkpoint": "saves/lntuning/llama-3-8b-instruct/train_cb_1745950316/checkpoint-12200",
  "epoch": 701.7610619469026,
  "eval_steps": 200,
  "global_step": 40000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.08849557522123894,
      "grad_norm": 9.3125,
      "learning_rate": 4.999999876629946e-05,
      "loss": 1.2638,
      "num_input_tokens_seen": 2944,
      "step": 5
    },
    {
      "epoch": 0.17699115044247787,
      "grad_norm": 7.53125,
      "learning_rate": 4.999999375439123e-05,
      "loss": 1.4007,
      "num_input_tokens_seen": 6048,
      "step": 10
    },
    {
      "epoch": 0.26548672566371684,
      "grad_norm": 7.8125,
      "learning_rate": 4.9999984887169785e-05,
      "loss": 1.2583,
      "num_input_tokens_seen": 8944,
      "step": 15
    },
    {
      "epoch": 0.35398230088495575,
      "grad_norm": 7.21875,
      "learning_rate": 4.9999972164636506e-05,
      "loss": 1.1509,
      "num_input_tokens_seen": 12416,
      "step": 20
    },
    {
      "epoch": 0.4424778761061947,
      "grad_norm": 8.875,
      "learning_rate": 4.999995558679334e-05,
      "loss": 1.3122,
      "num_input_tokens_seen": 15504,
      "step": 25
    },
    {
      "epoch": 0.5309734513274337,
      "grad_norm": 7.3125,
      "learning_rate": 4.999993515364287e-05,
      "loss": 1.3804,
      "num_input_tokens_seen": 17904,
      "step": 30
    },
    {
      "epoch": 0.6194690265486725,
      "grad_norm": 8.5,
      "learning_rate": 4.999991086518822e-05,
      "loss": 1.2557,
      "num_input_tokens_seen": 21216,
      "step": 35
    },
    {
      "epoch": 0.7079646017699115,
      "grad_norm": 8.6875,
      "learning_rate": 4.999988272143315e-05,
      "loss": 1.2571,
      "num_input_tokens_seen": 24016,
      "step": 40
    },
    {
      "epoch": 0.7964601769911505,
      "grad_norm": 7.125,
      "learning_rate": 4.999985072238199e-05,
      "loss": 1.4083,
      "num_input_tokens_seen": 26528,
      "step": 45
    },
    {
      "epoch": 0.8849557522123894,
      "grad_norm": 10.0625,
      "learning_rate": 4.999981486803969e-05,
      "loss": 1.1586,
      "num_input_tokens_seen": 29024,
      "step": 50
    },
    {
      "epoch": 0.9734513274336283,
      "grad_norm": 7.90625,
      "learning_rate": 4.999977515841176e-05,
      "loss": 1.2745,
      "num_input_tokens_seen": 31312,
      "step": 55
    },
    {
      "epoch": 1.0530973451327434,
      "grad_norm": 8.5,
      "learning_rate": 4.9999731593504344e-05,
      "loss": 1.0267,
      "num_input_tokens_seen": 33816,
      "step": 60
    },
    {
      "epoch": 1.1415929203539823,
      "grad_norm": 9.8125,
      "learning_rate": 4.999968417332415e-05,
      "loss": 1.146,
      "num_input_tokens_seen": 36728,
      "step": 65
    },
    {
      "epoch": 1.2300884955752212,
      "grad_norm": 7.1875,
      "learning_rate": 4.999963289787848e-05,
      "loss": 1.1922,
      "num_input_tokens_seen": 39304,
      "step": 70
    },
    {
      "epoch": 1.3185840707964602,
      "grad_norm": 8.375,
      "learning_rate": 4.999957776717526e-05,
      "loss": 1.2018,
      "num_input_tokens_seen": 42456,
      "step": 75
    },
    {
      "epoch": 1.407079646017699,
      "grad_norm": 12.9375,
      "learning_rate": 4.9999518781222984e-05,
      "loss": 1.3446,
      "num_input_tokens_seen": 45144,
      "step": 80
    },
    {
      "epoch": 1.495575221238938,
      "grad_norm": 9.9375,
      "learning_rate": 4.9999455940030746e-05,
      "loss": 1.1291,
      "num_input_tokens_seen": 48232,
      "step": 85
    },
    {
      "epoch": 1.584070796460177,
      "grad_norm": 9.8125,
      "learning_rate": 4.999938924360824e-05,
      "loss": 1.4123,
      "num_input_tokens_seen": 50680,
      "step": 90
    },
    {
      "epoch": 1.672566371681416,
      "grad_norm": 9.0625,
      "learning_rate": 4.999931869196575e-05,
      "loss": 1.2099,
      "num_input_tokens_seen": 53704,
      "step": 95
    },
    {
      "epoch": 1.7610619469026547,
      "grad_norm": 8.5,
      "learning_rate": 4.999924428511416e-05,
      "loss": 1.2482,
      "num_input_tokens_seen": 56440,
      "step": 100
    },
    {
      "epoch": 1.8495575221238938,
      "grad_norm": 9.0,
      "learning_rate": 4.999916602306494e-05,
      "loss": 1.3824,
      "num_input_tokens_seen": 59240,
      "step": 105
    },
    {
      "epoch": 1.9380530973451329,
      "grad_norm": 8.0625,
      "learning_rate": 4.999908390583016e-05,
      "loss": 1.1508,
      "num_input_tokens_seen": 61960,
      "step": 110
    },
    {
      "epoch": 2.017699115044248,
      "grad_norm": 9.5625,
      "learning_rate": 4.999899793342247e-05,
      "loss": 1.3396,
      "num_input_tokens_seen": 64040,
      "step": 115
    },
    {
      "epoch": 2.106194690265487,
      "grad_norm": 8.625,
      "learning_rate": 4.999890810585516e-05,
      "loss": 1.1536,
      "num_input_tokens_seen": 66968,
      "step": 120
    },
    {
      "epoch": 2.1946902654867255,
      "grad_norm": 9.3125,
      "learning_rate": 4.999881442314206e-05,
      "loss": 0.9778,
      "num_input_tokens_seen": 69544,
      "step": 125
    },
    {
      "epoch": 2.2831858407079646,
      "grad_norm": 7.125,
      "learning_rate": 4.9998716885297617e-05,
      "loss": 1.1172,
      "num_input_tokens_seen": 72488,
      "step": 130
    },
    {
      "epoch": 2.3716814159292037,
      "grad_norm": 7.5625,
      "learning_rate": 4.999861549233688e-05,
      "loss": 1.1433,
      "num_input_tokens_seen": 74968,
      "step": 135
    },
    {
      "epoch": 2.4601769911504423,
      "grad_norm": 7.6875,
      "learning_rate": 4.999851024427548e-05,
      "loss": 1.1161,
      "num_input_tokens_seen": 77816,
      "step": 140
    },
    {
      "epoch": 2.5486725663716814,
      "grad_norm": 10.6875,
      "learning_rate": 4.999840114112965e-05,
      "loss": 1.4204,
      "num_input_tokens_seen": 80728,
      "step": 145
    },
    {
      "epoch": 2.6371681415929205,
      "grad_norm": 8.5,
      "learning_rate": 4.999828818291621e-05,
      "loss": 1.1917,
      "num_input_tokens_seen": 83480,
      "step": 150
    },
    {
      "epoch": 2.725663716814159,
      "grad_norm": 7.46875,
      "learning_rate": 4.999817136965259e-05,
      "loss": 1.2222,
      "num_input_tokens_seen": 87000,
      "step": 155
    },
    {
      "epoch": 2.814159292035398,
      "grad_norm": 7.84375,
      "learning_rate": 4.9998050701356794e-05,
      "loss": 1.2101,
      "num_input_tokens_seen": 89976,
      "step": 160
    },
    {
      "epoch": 2.9026548672566372,
      "grad_norm": 9.0,
      "learning_rate": 4.999792617804744e-05,
      "loss": 1.2933,
      "num_input_tokens_seen": 92344,
      "step": 165
    },
    {
      "epoch": 2.991150442477876,
      "grad_norm": 8.75,
      "learning_rate": 4.9997797799743724e-05,
      "loss": 1.3692,
      "num_input_tokens_seen": 95160,
      "step": 170
    },
    {
      "epoch": 3.0707964601769913,
      "grad_norm": 9.8125,
      "learning_rate": 4.999766556646545e-05,
      "loss": 1.3054,
      "num_input_tokens_seen": 97544,
      "step": 175
    },
    {
      "epoch": 3.15929203539823,
      "grad_norm": 7.96875,
      "learning_rate": 4.9997529478232996e-05,
      "loss": 0.9286,
      "num_input_tokens_seen": 100296,
      "step": 180
    },
    {
      "epoch": 3.247787610619469,
      "grad_norm": 9.8125,
      "learning_rate": 4.9997389535067365e-05,
      "loss": 1.1188,
      "num_input_tokens_seen": 103016,
      "step": 185
    },
    {
      "epoch": 3.336283185840708,
      "grad_norm": 7.0,
      "learning_rate": 4.999724573699012e-05,
      "loss": 1.0613,
      "num_input_tokens_seen": 106056,
      "step": 190
    },
    {
      "epoch": 3.4247787610619467,
      "grad_norm": 7.53125,
      "learning_rate": 4.9997098084023457e-05,
      "loss": 1.2455,
      "num_input_tokens_seen": 108920,
      "step": 195
    },
    {
      "epoch": 3.5132743362831858,
      "grad_norm": 9.5625,
      "learning_rate": 4.999694657619013e-05,
      "loss": 1.3945,
      "num_input_tokens_seen": 111736,
      "step": 200
    },
    {
      "epoch": 3.5132743362831858,
      "eval_loss": 1.1587244272232056,
      "eval_runtime": 0.9555,
      "eval_samples_per_second": 26.163,
      "eval_steps_per_second": 13.605,
      "num_input_tokens_seen": 111736,
      "step": 200
    },
    {
      "epoch": 3.601769911504425,
      "grad_norm": 7.875,
      "learning_rate": 4.999679121351352e-05,
      "loss": 1.1668,
      "num_input_tokens_seen": 114776,
      "step": 205
    },
    {
      "epoch": 3.6902654867256635,
      "grad_norm": 9.1875,
      "learning_rate": 4.9996631996017565e-05,
      "loss": 1.194,
      "num_input_tokens_seen": 117320,
      "step": 210
    },
    {
      "epoch": 3.7787610619469025,
      "grad_norm": 7.0625,
      "learning_rate": 4.9996468923726835e-05,
      "loss": 1.4169,
      "num_input_tokens_seen": 119928,
      "step": 215
    },
    {
      "epoch": 3.8672566371681416,
      "grad_norm": 6.90625,
      "learning_rate": 4.999630199666647e-05,
      "loss": 1.1083,
      "num_input_tokens_seen": 122824,
      "step": 220
    },
    {
      "epoch": 3.9557522123893807,
      "grad_norm": 8.75,
      "learning_rate": 4.999613121486222e-05,
      "loss": 1.0882,
      "num_input_tokens_seen": 126392,
      "step": 225
    },
    {
      "epoch": 4.035398230088496,
      "grad_norm": 7.28125,
      "learning_rate": 4.999595657834041e-05,
      "loss": 1.1273,
      "num_input_tokens_seen": 128712,
      "step": 230
    },
    {
      "epoch": 4.123893805309734,
      "grad_norm": 8.625,
      "learning_rate": 4.999577808712798e-05,
      "loss": 1.1679,
      "num_input_tokens_seen": 131384,
      "step": 235
    },
    {
      "epoch": 4.212389380530974,
      "grad_norm": 10.0,
      "learning_rate": 4.999559574125244e-05,
      "loss": 1.259,
      "num_input_tokens_seen": 133816,
      "step": 240
    },
    {
      "epoch": 4.300884955752212,
      "grad_norm": 8.5625,
      "learning_rate": 4.9995409540741934e-05,
      "loss": 1.3171,
      "num_input_tokens_seen": 136472,
      "step": 245
    },
    {
      "epoch": 4.389380530973451,
      "grad_norm": 9.4375,
      "learning_rate": 4.999521948562516e-05,
      "loss": 1.1356,
      "num_input_tokens_seen": 139016,
      "step": 250
    },
    {
      "epoch": 4.477876106194691,
      "grad_norm": 8.3125,
      "learning_rate": 4.999502557593143e-05,
      "loss": 0.9955,
      "num_input_tokens_seen": 142120,
      "step": 255
    },
    {
      "epoch": 4.566371681415929,
      "grad_norm": 11.5,
      "learning_rate": 4.999482781169066e-05,
      "loss": 1.1223,
      "num_input_tokens_seen": 144696,
      "step": 260
    },
    {
      "epoch": 4.654867256637168,
      "grad_norm": 6.5625,
      "learning_rate": 4.9994626192933324e-05,
      "loss": 1.2022,
      "num_input_tokens_seen": 147304,
      "step": 265
    },
    {
      "epoch": 4.743362831858407,
      "grad_norm": 9.75,
      "learning_rate": 4.999442071969054e-05,
      "loss": 1.1006,
      "num_input_tokens_seen": 150104,
      "step": 270
    },
    {
      "epoch": 4.831858407079646,
      "grad_norm": 8.5,
      "learning_rate": 4.999421139199397e-05,
      "loss": 1.0905,
      "num_input_tokens_seen": 153464,
      "step": 275
    },
    {
      "epoch": 4.920353982300885,
      "grad_norm": 7.84375,
      "learning_rate": 4.999399820987592e-05,
      "loss": 1.1411,
      "num_input_tokens_seen": 156440,
      "step": 280
    },
    {
      "epoch": 5.0,
      "grad_norm": 7.09375,
      "learning_rate": 4.999378117336924e-05,
      "loss": 1.0342,
      "num_input_tokens_seen": 158768,
      "step": 285
    },
    {
      "epoch": 5.088495575221239,
      "grad_norm": 9.1875,
      "learning_rate": 4.9993560282507415e-05,
      "loss": 1.1291,
      "num_input_tokens_seen": 161360,
      "step": 290
    },
    {
      "epoch": 5.176991150442478,
      "grad_norm": 10.5625,
      "learning_rate": 4.9993335537324495e-05,
      "loss": 1.1129,
      "num_input_tokens_seen": 164368,
      "step": 295
    },
    {
      "epoch": 5.265486725663717,
      "grad_norm": 9.6875,
      "learning_rate": 4.999310693785516e-05,
      "loss": 1.1924,
      "num_input_tokens_seen": 167360,
      "step": 300
    },
    {
      "epoch": 5.353982300884955,
      "grad_norm": 7.90625,
      "learning_rate": 4.9992874484134653e-05,
      "loss": 0.9392,
      "num_input_tokens_seen": 170672,
      "step": 305
    },
    {
      "epoch": 5.442477876106195,
      "grad_norm": 8.6875,
      "learning_rate": 4.999263817619882e-05,
      "loss": 1.1734,
      "num_input_tokens_seen": 173184,
      "step": 310
    },
    {
      "epoch": 5.530973451327434,
      "grad_norm": 8.3125,
      "learning_rate": 4.9992398014084105e-05,
      "loss": 1.1584,
      "num_input_tokens_seen": 175968,
      "step": 315
    },
    {
      "epoch": 5.619469026548672,
      "grad_norm": 9.625,
      "learning_rate": 4.999215399782754e-05,
      "loss": 1.2579,
      "num_input_tokens_seen": 178608,
      "step": 320
    },
    {
      "epoch": 5.707964601769912,
      "grad_norm": 8.5625,
      "learning_rate": 4.999190612746675e-05,
      "loss": 1.159,
      "num_input_tokens_seen": 181088,
      "step": 325
    },
    {
      "epoch": 5.79646017699115,
      "grad_norm": 8.75,
      "learning_rate": 4.999165440303998e-05,
      "loss": 1.2194,
      "num_input_tokens_seen": 184224,
      "step": 330
    },
    {
      "epoch": 5.88495575221239,
      "grad_norm": 8.8125,
      "learning_rate": 4.999139882458603e-05,
      "loss": 1.0607,
      "num_input_tokens_seen": 187184,
      "step": 335
    },
    {
      "epoch": 5.9734513274336285,
      "grad_norm": 6.84375,
      "learning_rate": 4.9991139392144314e-05,
      "loss": 0.9811,
      "num_input_tokens_seen": 189680,
      "step": 340
    },
    {
      "epoch": 6.053097345132743,
      "grad_norm": 7.9375,
      "learning_rate": 4.999087610575485e-05,
      "loss": 1.0384,
      "num_input_tokens_seen": 192272,
      "step": 345
    },
    {
      "epoch": 6.1415929203539825,
      "grad_norm": 9.0,
      "learning_rate": 4.999060896545824e-05,
      "loss": 1.1132,
      "num_input_tokens_seen": 195200,
      "step": 350
    },
    {
      "epoch": 6.230088495575221,
      "grad_norm": 9.0625,
      "learning_rate": 4.999033797129568e-05,
      "loss": 1.0813,
      "num_input_tokens_seen": 198352,
      "step": 355
    },
    {
      "epoch": 6.31858407079646,
      "grad_norm": 8.625,
      "learning_rate": 4.999006312330894e-05,
      "loss": 1.2567,
      "num_input_tokens_seen": 201024,
      "step": 360
    },
    {
      "epoch": 6.407079646017699,
      "grad_norm": 7.65625,
      "learning_rate": 4.998978442154043e-05,
      "loss": 1.0612,
      "num_input_tokens_seen": 204128,
      "step": 365
    },
    {
      "epoch": 6.495575221238938,
      "grad_norm": 8.375,
      "learning_rate": 4.9989501866033125e-05,
      "loss": 0.9097,
      "num_input_tokens_seen": 206784,
      "step": 370
    },
    {
      "epoch": 6.584070796460177,
      "grad_norm": 9.1875,
      "learning_rate": 4.998921545683059e-05,
      "loss": 1.1683,
      "num_input_tokens_seen": 209680,
      "step": 375
    },
    {
      "epoch": 6.672566371681416,
      "grad_norm": 9.6875,
      "learning_rate": 4.9988925193976996e-05,
      "loss": 1.2941,
      "num_input_tokens_seen": 211984,
      "step": 380
    },
    {
      "epoch": 6.761061946902655,
      "grad_norm": 12.125,
      "learning_rate": 4.998863107751711e-05,
      "loss": 1.2087,
      "num_input_tokens_seen": 214576,
      "step": 385
    },
    {
      "epoch": 6.849557522123893,
      "grad_norm": 8.4375,
      "learning_rate": 4.998833310749629e-05,
      "loss": 1.0459,
      "num_input_tokens_seen": 217024,
      "step": 390
    },
    {
      "epoch": 6.938053097345133,
      "grad_norm": 9.0625,
      "learning_rate": 4.998803128396047e-05,
      "loss": 1.0981,
      "num_input_tokens_seen": 220432,
      "step": 395
    },
    {
      "epoch": 7.017699115044247,
      "grad_norm": 9.1875,
      "learning_rate": 4.9987725606956215e-05,
      "loss": 1.0181,
      "num_input_tokens_seen": 223024,
      "step": 400
    },
    {
      "epoch": 7.017699115044247,
      "eval_loss": 1.0770518779754639,
      "eval_runtime": 0.9301,
      "eval_samples_per_second": 26.878,
      "eval_steps_per_second": 13.976,
      "num_input_tokens_seen": 223024,
      "step": 400
    },
    {
      "epoch": 7.106194690265487,
      "grad_norm": 6.3125,
      "learning_rate": 4.998741607653066e-05,
      "loss": 0.9256,
      "num_input_tokens_seen": 225840,
      "step": 405
    },
    {
      "epoch": 7.1946902654867255,
      "grad_norm": 8.0625,
      "learning_rate": 4.9987102692731523e-05,
      "loss": 1.0519,
      "num_input_tokens_seen": 228224,
      "step": 410
    },
    {
      "epoch": 7.283185840707965,
      "grad_norm": 11.1875,
      "learning_rate": 4.9986785455607157e-05,
      "loss": 1.3414,
      "num_input_tokens_seen": 231088,
      "step": 415
    },
    {
      "epoch": 7.371681415929204,
      "grad_norm": 7.4375,
      "learning_rate": 4.9986464365206456e-05,
      "loss": 1.2015,
      "num_input_tokens_seen": 233968,
      "step": 420
    },
    {
      "epoch": 7.460176991150442,
      "grad_norm": 8.625,
      "learning_rate": 4.9986139421578956e-05,
      "loss": 1.1636,
      "num_input_tokens_seen": 236560,
      "step": 425
    },
    {
      "epoch": 7.548672566371682,
      "grad_norm": 9.5,
      "learning_rate": 4.998581062477477e-05,
      "loss": 1.1511,
      "num_input_tokens_seen": 239360,
      "step": 430
    },
    {
      "epoch": 7.6371681415929205,
      "grad_norm": 8.125,
      "learning_rate": 4.998547797484458e-05,
      "loss": 1.1082,
      "num_input_tokens_seen": 242352,
      "step": 435
    },
    {
      "epoch": 7.725663716814159,
      "grad_norm": 6.1875,
      "learning_rate": 4.9985141471839706e-05,
      "loss": 1.0142,
      "num_input_tokens_seen": 245136,
      "step": 440
    },
    {
      "epoch": 7.814159292035399,
      "grad_norm": 8.4375,
      "learning_rate": 4.998480111581203e-05,
      "loss": 1.0493,
      "num_input_tokens_seen": 247584,
      "step": 445
    },
    {
      "epoch": 7.902654867256637,
      "grad_norm": 8.0,
      "learning_rate": 4.998445690681405e-05,
      "loss": 0.9847,
      "num_input_tokens_seen": 250576,
      "step": 450
    },
    {
      "epoch": 7.991150442477876,
      "grad_norm": 8.125,
      "learning_rate": 4.9984108844898834e-05,
      "loss": 1.1305,
      "num_input_tokens_seen": 253376,
      "step": 455
    },
    {
      "epoch": 8.070796460176991,
      "grad_norm": 6.9375,
      "learning_rate": 4.9983756930120076e-05,
      "loss": 1.0502,
      "num_input_tokens_seen": 255984,
      "step": 460
    },
    {
      "epoch": 8.15929203539823,
      "grad_norm": 7.59375,
      "learning_rate": 4.9983401162532025e-05,
      "loss": 1.2274,
      "num_input_tokens_seen": 258736,
      "step": 465
    },
    {
      "epoch": 8.247787610619469,
      "grad_norm": 9.875,
      "learning_rate": 4.998304154218955e-05,
      "loss": 0.9734,
      "num_input_tokens_seen": 261648,
      "step": 470
    },
    {
      "epoch": 8.336283185840708,
      "grad_norm": 8.75,
      "learning_rate": 4.998267806914812e-05,
      "loss": 1.0464,
      "num_input_tokens_seen": 264160,
      "step": 475
    },
    {
      "epoch": 8.424778761061948,
      "grad_norm": 8.6875,
      "learning_rate": 4.998231074346378e-05,
      "loss": 1.1154,
      "num_input_tokens_seen": 266784,
      "step": 480
    },
    {
      "epoch": 8.513274336283185,
      "grad_norm": 9.6875,
      "learning_rate": 4.998193956519317e-05,
      "loss": 1.0487,
      "num_input_tokens_seen": 269936,
      "step": 485
    },
    {
      "epoch": 8.601769911504425,
      "grad_norm": 8.6875,
      "learning_rate": 4.9981564534393545e-05,
      "loss": 1.1528,
      "num_input_tokens_seen": 272848,
      "step": 490
    },
    {
      "epoch": 8.690265486725664,
      "grad_norm": 6.65625,
      "learning_rate": 4.998118565112272e-05,
      "loss": 0.998,
      "num_input_tokens_seen": 275552,
      "step": 495
    },
    {
      "epoch": 8.778761061946902,
      "grad_norm": 7.34375,
      "learning_rate": 4.998080291543914e-05,
      "loss": 1.003,
      "num_input_tokens_seen": 278016,
      "step": 500
    },
    {
      "epoch": 8.867256637168142,
      "grad_norm": 7.59375,
      "learning_rate": 4.9980416327401826e-05,
      "loss": 1.1562,
      "num_input_tokens_seen": 281040,
      "step": 505
    },
    {
      "epoch": 8.955752212389381,
      "grad_norm": 8.6875,
      "learning_rate": 4.998002588707038e-05,
      "loss": 1.1408,
      "num_input_tokens_seen": 283776,
      "step": 510
    },
    {
      "epoch": 9.035398230088495,
      "grad_norm": 12.75,
      "learning_rate": 4.997963159450503e-05,
      "loss": 1.0989,
      "num_input_tokens_seen": 286152,
      "step": 515
    },
    {
      "epoch": 9.123893805309734,
      "grad_norm": 8.25,
      "learning_rate": 4.9979233449766575e-05,
      "loss": 0.9627,
      "num_input_tokens_seen": 289384,
      "step": 520
    },
    {
      "epoch": 9.212389380530974,
      "grad_norm": 9.4375,
      "learning_rate": 4.997883145291641e-05,
      "loss": 1.1661,
      "num_input_tokens_seen": 291992,
      "step": 525
    },
    {
      "epoch": 9.300884955752213,
      "grad_norm": 6.875,
      "learning_rate": 4.9978425604016536e-05,
      "loss": 0.9506,
      "num_input_tokens_seen": 294776,
      "step": 530
    },
    {
      "epoch": 9.389380530973451,
      "grad_norm": 7.78125,
      "learning_rate": 4.9978015903129536e-05,
      "loss": 0.9976,
      "num_input_tokens_seen": 297528,
      "step": 535
    },
    {
      "epoch": 9.47787610619469,
      "grad_norm": 8.1875,
      "learning_rate": 4.997760235031859e-05,
      "loss": 0.9339,
      "num_input_tokens_seen": 300136,
      "step": 540
    },
    {
      "epoch": 9.56637168141593,
      "grad_norm": 8.3125,
      "learning_rate": 4.9977184945647473e-05,
      "loss": 1.0814,
      "num_input_tokens_seen": 303304,
      "step": 545
    },
    {
      "epoch": 9.654867256637168,
      "grad_norm": 8.625,
      "learning_rate": 4.997676368918055e-05,
      "loss": 1.1517,
      "num_input_tokens_seen": 306216,
      "step": 550
    },
    {
      "epoch": 9.743362831858407,
      "grad_norm": 8.5625,
      "learning_rate": 4.9976338580982794e-05,
      "loss": 1.2526,
      "num_input_tokens_seen": 309048,
      "step": 555
    },
    {
      "epoch": 9.831858407079647,
      "grad_norm": 7.1875,
      "learning_rate": 4.9975909621119755e-05,
      "loss": 1.179,
      "num_input_tokens_seen": 311944,
      "step": 560
    },
    {
      "epoch": 9.920353982300885,
      "grad_norm": 8.3125,
      "learning_rate": 4.997547680965758e-05,
      "loss": 1.1038,
      "num_input_tokens_seen": 314776,
      "step": 565
    },
    {
      "epoch": 10.0,
      "grad_norm": 7.75,
      "learning_rate": 4.997504014666302e-05,
      "loss": 1.2473,
      "num_input_tokens_seen": 316840,
      "step": 570
    },
    {
      "epoch": 10.08849557522124,
      "grad_norm": 7.71875,
      "learning_rate": 4.997459963220342e-05,
      "loss": 1.233,
      "num_input_tokens_seen": 319640,
      "step": 575
    },
    {
      "epoch": 10.176991150442477,
      "grad_norm": 8.1875,
      "learning_rate": 4.997415526634671e-05,
      "loss": 1.2215,
      "num_input_tokens_seen": 322296,
      "step": 580
    },
    {
      "epoch": 10.265486725663717,
      "grad_norm": 7.3125,
      "learning_rate": 4.99737070491614e-05,
      "loss": 1.184,
      "num_input_tokens_seen": 324936,
      "step": 585
    },
    {
      "epoch": 10.353982300884956,
      "grad_norm": 7.34375,
      "learning_rate": 4.997325498071663e-05,
      "loss": 0.9586,
      "num_input_tokens_seen": 327400,
      "step": 590
    },
    {
      "epoch": 10.442477876106194,
      "grad_norm": 7.40625,
      "learning_rate": 4.997279906108211e-05,
      "loss": 1.0311,
      "num_input_tokens_seen": 330376,
      "step": 595
    },
    {
      "epoch": 10.530973451327434,
      "grad_norm": 9.125,
      "learning_rate": 4.9972339290328155e-05,
      "loss": 1.237,
      "num_input_tokens_seen": 332984,
      "step": 600
    },
    {
      "epoch": 10.530973451327434,
      "eval_loss": 1.0585631132125854,
      "eval_runtime": 0.9312,
      "eval_samples_per_second": 26.847,
      "eval_steps_per_second": 13.961,
      "num_input_tokens_seen": 332984,
      "step": 600
    },
    {
      "epoch": 10.619469026548673,
      "grad_norm": 8.125,
      "learning_rate": 4.9971875668525646e-05,
      "loss": 1.0125,
      "num_input_tokens_seen": 335400,
      "step": 605
    },
    {
      "epoch": 10.70796460176991,
      "grad_norm": 9.625,
      "learning_rate": 4.997140819574609e-05,
      "loss": 1.1174,
      "num_input_tokens_seen": 338312,
      "step": 610
    },
    {
      "epoch": 10.79646017699115,
      "grad_norm": 6.25,
      "learning_rate": 4.997093687206159e-05,
      "loss": 0.8649,
      "num_input_tokens_seen": 341304,
      "step": 615
    },
    {
      "epoch": 10.88495575221239,
      "grad_norm": 9.375,
      "learning_rate": 4.997046169754482e-05,
      "loss": 1.08,
      "num_input_tokens_seen": 344376,
      "step": 620
    },
    {
      "epoch": 10.973451327433628,
      "grad_norm": 7.96875,
      "learning_rate": 4.996998267226905e-05,
      "loss": 0.9674,
      "num_input_tokens_seen": 347416,
      "step": 625
    },
    {
      "epoch": 11.053097345132743,
      "grad_norm": 9.1875,
      "learning_rate": 4.996949979630817e-05,
      "loss": 0.9602,
      "num_input_tokens_seen": 349800,
      "step": 630
    },
    {
      "epoch": 11.141592920353983,
      "grad_norm": 9.0625,
      "learning_rate": 4.996901306973663e-05,
      "loss": 1.2431,
      "num_input_tokens_seen": 352344,
      "step": 635
    },
    {
      "epoch": 11.230088495575222,
      "grad_norm": 9.125,
      "learning_rate": 4.996852249262949e-05,
      "loss": 1.0618,
      "num_input_tokens_seen": 355016,
      "step": 640
    },
    {
      "epoch": 11.31858407079646,
      "grad_norm": 12.5,
      "learning_rate": 4.996802806506241e-05,
      "loss": 1.0656,
      "num_input_tokens_seen": 358424,
      "step": 645
    },
    {
      "epoch": 11.4070796460177,
      "grad_norm": 8.5625,
      "learning_rate": 4.996752978711164e-05,
      "loss": 0.9867,
      "num_input_tokens_seen": 361544,
      "step": 650
    },
    {
      "epoch": 11.495575221238939,
      "grad_norm": 8.6875,
      "learning_rate": 4.996702765885401e-05,
      "loss": 1.174,
      "num_input_tokens_seen": 364488,
      "step": 655
    },
    {
      "epoch": 11.584070796460177,
      "grad_norm": 12.375,
      "learning_rate": 4.9966521680366964e-05,
      "loss": 1.1597,
      "num_input_tokens_seen": 367336,
      "step": 660
    },
    {
      "epoch": 11.672566371681416,
      "grad_norm": 6.8125,
      "learning_rate": 4.9966011851728524e-05,
      "loss": 0.9914,
      "num_input_tokens_seen": 370008,
      "step": 665
    },
    {
      "epoch": 11.761061946902656,
      "grad_norm": 8.625,
      "learning_rate": 4.996549817301731e-05,
      "loss": 1.1205,
      "num_input_tokens_seen": 372392,
      "step": 670
    },
    {
      "epoch": 11.849557522123893,
      "grad_norm": 8.875,
      "learning_rate": 4.9964980644312544e-05,
      "loss": 1.0504,
      "num_input_tokens_seen": 375112,
      "step": 675
    },
    {
      "epoch": 11.938053097345133,
      "grad_norm": 7.5,
      "learning_rate": 4.996445926569403e-05,
      "loss": 1.0394,
      "num_input_tokens_seen": 377896,
      "step": 680
    },
    {
      "epoch": 12.017699115044248,
      "grad_norm": 10.625,
      "learning_rate": 4.996393403724218e-05,
      "loss": 1.0184,
      "num_input_tokens_seen": 380672,
      "step": 685
    },
    {
      "epoch": 12.106194690265486,
      "grad_norm": 8.1875,
      "learning_rate": 4.9963404959037985e-05,
      "loss": 0.9884,
      "num_input_tokens_seen": 383376,
      "step": 690
    },
    {
      "epoch": 12.194690265486726,
      "grad_norm": 7.8125,
      "learning_rate": 4.996287203116303e-05,
      "loss": 1.0807,
      "num_input_tokens_seen": 386368,
      "step": 695
    },
    {
      "epoch": 12.283185840707965,
      "grad_norm": 10.9375,
      "learning_rate": 4.996233525369951e-05,
      "loss": 1.1899,
      "num_input_tokens_seen": 389184,
      "step": 700
    },
    {
      "epoch": 12.371681415929203,
      "grad_norm": 8.875,
      "learning_rate": 4.99617946267302e-05,
      "loss": 1.0627,
      "num_input_tokens_seen": 391824,
      "step": 705
    },
    {
      "epoch": 12.460176991150442,
      "grad_norm": 9.1875,
      "learning_rate": 4.996125015033846e-05,
      "loss": 1.0791,
      "num_input_tokens_seen": 395024,
      "step": 710
    },
    {
      "epoch": 12.548672566371682,
      "grad_norm": 6.96875,
      "learning_rate": 4.996070182460827e-05,
      "loss": 0.951,
      "num_input_tokens_seen": 397840,
      "step": 715
    },
    {
      "epoch": 12.63716814159292,
      "grad_norm": 11.6875,
      "learning_rate": 4.996014964962418e-05,
      "loss": 0.9092,
      "num_input_tokens_seen": 400480,
      "step": 720
    },
    {
      "epoch": 12.725663716814159,
      "grad_norm": 7.71875,
      "learning_rate": 4.9959593625471344e-05,
      "loss": 1.0353,
      "num_input_tokens_seen": 403488,
      "step": 725
    },
    {
      "epoch": 12.814159292035399,
      "grad_norm": 9.0,
      "learning_rate": 4.995903375223552e-05,
      "loss": 1.164,
      "num_input_tokens_seen": 406000,
      "step": 730
    },
    {
      "epoch": 12.902654867256636,
      "grad_norm": 7.9375,
      "learning_rate": 4.995847003000302e-05,
      "loss": 1.1499,
      "num_input_tokens_seen": 408800,
      "step": 735
    },
    {
      "epoch": 12.991150442477876,
      "grad_norm": 7.28125,
      "learning_rate": 4.9957902458860804e-05,
      "loss": 1.1274,
      "num_input_tokens_seen": 411440,
      "step": 740
    },
    {
      "epoch": 13.070796460176991,
      "grad_norm": 9.4375,
      "learning_rate": 4.995733103889639e-05,
      "loss": 1.0626,
      "num_input_tokens_seen": 414400,
      "step": 745
    },
    {
      "epoch": 13.15929203539823,
      "grad_norm": 8.8125,
      "learning_rate": 4.99567557701979e-05,
      "loss": 1.036,
      "num_input_tokens_seen": 416832,
      "step": 750
    },
    {
      "epoch": 13.247787610619469,
      "grad_norm": 7.625,
      "learning_rate": 4.995617665285403e-05,
      "loss": 1.0755,
      "num_input_tokens_seen": 419440,
      "step": 755
    },
    {
      "epoch": 13.336283185840708,
      "grad_norm": 8.8125,
      "learning_rate": 4.99555936869541e-05,
      "loss": 0.8651,
      "num_input_tokens_seen": 422352,
      "step": 760
    },
    {
      "epoch": 13.424778761061948,
      "grad_norm": 9.25,
      "learning_rate": 4.995500687258803e-05,
      "loss": 1.2047,
      "num_input_tokens_seen": 424992,
      "step": 765
    },
    {
      "epoch": 13.513274336283185,
      "grad_norm": 9.4375,
      "learning_rate": 4.995441620984628e-05,
      "loss": 0.9927,
      "num_input_tokens_seen": 428032,
      "step": 770
    },
    {
      "epoch": 13.601769911504425,
      "grad_norm": 8.6875,
      "learning_rate": 4.995382169881996e-05,
      "loss": 1.058,
      "num_input_tokens_seen": 430624,
      "step": 775
    },
    {
      "epoch": 13.690265486725664,
      "grad_norm": 9.0625,
      "learning_rate": 4.9953223339600755e-05,
      "loss": 1.0281,
      "num_input_tokens_seen": 433504,
      "step": 780
    },
    {
      "epoch": 13.778761061946902,
      "grad_norm": 7.53125,
      "learning_rate": 4.995262113228091e-05,
      "loss": 1.0287,
      "num_input_tokens_seen": 436512,
      "step": 785
    },
    {
      "epoch": 13.867256637168142,
      "grad_norm": 12.1875,
      "learning_rate": 4.995201507695332e-05,
      "loss": 1.2929,
      "num_input_tokens_seen": 439568,
      "step": 790
    },
    {
      "epoch": 13.955752212389381,
      "grad_norm": 10.375,
      "learning_rate": 4.995140517371144e-05,
      "loss": 1.1704,
      "num_input_tokens_seen": 442464,
      "step": 795
    },
    {
      "epoch": 14.035398230088495,
      "grad_norm": 8.125,
      "learning_rate": 4.995079142264932e-05,
      "loss": 1.1874,
      "num_input_tokens_seen": 444576,
      "step": 800
    },
    {
      "epoch": 14.035398230088495,
      "eval_loss": 1.0446652173995972,
      "eval_runtime": 1.0107,
      "eval_samples_per_second": 24.736,
      "eval_steps_per_second": 12.863,
      "num_input_tokens_seen": 444576,
      "step": 800
    },
    {
      "epoch": 14.123893805309734,
      "grad_norm": 7.15625,
      "learning_rate": 4.995017382386162e-05,
      "loss": 0.9837,
      "num_input_tokens_seen": 447424,
      "step": 805
    },
    {
      "epoch": 14.212389380530974,
      "grad_norm": 7.03125,
      "learning_rate": 4.994955237744356e-05,
      "loss": 1.0797,
      "num_input_tokens_seen": 450080,
      "step": 810
    },
    {
      "epoch": 14.300884955752213,
      "grad_norm": 7.875,
      "learning_rate": 4.994892708349101e-05,
      "loss": 1.0992,
      "num_input_tokens_seen": 452704,
      "step": 815
    },
    {
      "epoch": 14.389380530973451,
      "grad_norm": 7.875,
      "learning_rate": 4.994829794210035e-05,
      "loss": 0.9993,
      "num_input_tokens_seen": 455808,
      "step": 820
    },
    {
      "epoch": 14.47787610619469,
      "grad_norm": 9.125,
      "learning_rate": 4.994766495336864e-05,
      "loss": 1.2195,
      "num_input_tokens_seen": 458576,
      "step": 825
    },
    {
      "epoch": 14.56637168141593,
      "grad_norm": 7.59375,
      "learning_rate": 4.994702811739348e-05,
      "loss": 1.0706,
      "num_input_tokens_seen": 461344,
      "step": 830
    },
    {
      "epoch": 14.654867256637168,
      "grad_norm": 7.34375,
      "learning_rate": 4.994638743427308e-05,
      "loss": 0.8013,
      "num_input_tokens_seen": 464256,
      "step": 835
    },
    {
      "epoch": 14.743362831858407,
      "grad_norm": 9.0,
      "learning_rate": 4.994574290410624e-05,
      "loss": 1.0766,
      "num_input_tokens_seen": 467728,
      "step": 840
    },
    {
      "epoch": 14.831858407079647,
      "grad_norm": 8.6875,
      "learning_rate": 4.9945094526992364e-05,
      "loss": 1.1925,
      "num_input_tokens_seen": 470208,
      "step": 845
    },
    {
      "epoch": 14.920353982300885,
      "grad_norm": 7.8125,
      "learning_rate": 4.994444230303142e-05,
      "loss": 1.0408,
      "num_input_tokens_seen": 473024,
      "step": 850
    },
    {
      "epoch": 15.0,
      "grad_norm": 10.125,
      "learning_rate": 4.994378623232402e-05,
      "loss": 1.1175,
      "num_input_tokens_seen": 475000,
      "step": 855
    },
    {
      "epoch": 15.08849557522124,
      "grad_norm": 10.1875,
      "learning_rate": 4.99431263149713e-05,
      "loss": 1.0947,
      "num_input_tokens_seen": 477960,
      "step": 860
    },
    {
      "epoch": 15.176991150442477,
      "grad_norm": 8.625,
      "learning_rate": 4.9942462551075056e-05,
      "loss": 1.0971,
      "num_input_tokens_seen": 481080,
      "step": 865
    },
    {
      "epoch": 15.265486725663717,
      "grad_norm": 7.875,
      "learning_rate": 4.994179494073764e-05,
      "loss": 0.9375,
      "num_input_tokens_seen": 483768,
      "step": 870
    },
    {
      "epoch": 15.353982300884956,
      "grad_norm": 10.25,
      "learning_rate": 4.9941123484062e-05,
      "loss": 1.2249,
      "num_input_tokens_seen": 486680,
      "step": 875
    },
    {
      "epoch": 15.442477876106194,
      "grad_norm": 6.875,
      "learning_rate": 4.99404481811517e-05,
      "loss": 1.0275,
      "num_input_tokens_seen": 489560,
      "step": 880
    },
    {
      "epoch": 15.530973451327434,
      "grad_norm": 9.4375,
      "learning_rate": 4.9939769032110864e-05,
      "loss": 1.0417,
      "num_input_tokens_seen": 492072,
      "step": 885
    },
    {
      "epoch": 15.619469026548673,
      "grad_norm": 8.875,
      "learning_rate": 4.993908603704423e-05,
      "loss": 1.1103,
      "num_input_tokens_seen": 494744,
      "step": 890
    },
    {
      "epoch": 15.70796460176991,
      "grad_norm": 13.3125,
      "learning_rate": 4.9938399196057126e-05,
      "loss": 1.0616,
      "num_input_tokens_seen": 497368,
      "step": 895
    },
    {
      "epoch": 15.79646017699115,
      "grad_norm": 10.6875,
      "learning_rate": 4.993770850925547e-05,
      "loss": 1.1362,
      "num_input_tokens_seen": 500328,
      "step": 900
    },
    {
      "epoch": 15.88495575221239,
      "grad_norm": 9.625,
      "learning_rate": 4.993701397674577e-05,
      "loss": 1.0152,
      "num_input_tokens_seen": 503304,
      "step": 905
    },
    {
      "epoch": 15.973451327433628,
      "grad_norm": 9.3125,
      "learning_rate": 4.993631559863515e-05,
      "loss": 0.9806,
      "num_input_tokens_seen": 506104,
      "step": 910
    },
    {
      "epoch": 16.053097345132745,
      "grad_norm": 6.34375,
      "learning_rate": 4.9935613375031283e-05,
      "loss": 1.5298,
      "num_input_tokens_seen": 508160,
      "step": 915
    },
    {
      "epoch": 16.141592920353983,
      "grad_norm": 8.3125,
      "learning_rate": 4.993490730604248e-05,
      "loss": 0.9348,
      "num_input_tokens_seen": 511552,
      "step": 920
    },
    {
      "epoch": 16.23008849557522,
      "grad_norm": 12.3125,
      "learning_rate": 4.993419739177761e-05,
      "loss": 0.9761,
      "num_input_tokens_seen": 514752,
      "step": 925
    },
    {
      "epoch": 16.31858407079646,
      "grad_norm": 12.1875,
      "learning_rate": 4.9933483632346164e-05,
      "loss": 1.1229,
      "num_input_tokens_seen": 517696,
      "step": 930
    },
    {
      "epoch": 16.4070796460177,
      "grad_norm": 8.0,
      "learning_rate": 4.993276602785821e-05,
      "loss": 1.3169,
      "num_input_tokens_seen": 520304,
      "step": 935
    },
    {
      "epoch": 16.495575221238937,
      "grad_norm": 7.875,
      "learning_rate": 4.993204457842441e-05,
      "loss": 1.0606,
      "num_input_tokens_seen": 523440,
      "step": 940
    },
    {
      "epoch": 16.58407079646018,
      "grad_norm": 9.5,
      "learning_rate": 4.993131928415602e-05,
      "loss": 1.1164,
      "num_input_tokens_seen": 525936,
      "step": 945
    },
    {
      "epoch": 16.672566371681416,
      "grad_norm": 8.4375,
      "learning_rate": 4.993059014516489e-05,
      "loss": 1.0272,
      "num_input_tokens_seen": 528944,
      "step": 950
    },
    {
      "epoch": 16.761061946902654,
      "grad_norm": 8.1875,
      "learning_rate": 4.9929857161563464e-05,
      "loss": 1.2031,
      "num_input_tokens_seen": 531440,
      "step": 955
    },
    {
      "epoch": 16.849557522123895,
      "grad_norm": 5.90625,
      "learning_rate": 4.992912033346477e-05,
      "loss": 0.9117,
      "num_input_tokens_seen": 534192,
      "step": 960
    },
    {
      "epoch": 16.938053097345133,
      "grad_norm": 8.5,
      "learning_rate": 4.992837966098245e-05,
      "loss": 0.8809,
      "num_input_tokens_seen": 537216,
      "step": 965
    },
    {
      "epoch": 17.01769911504425,
      "grad_norm": 7.75,
      "learning_rate": 4.992763514423071e-05,
      "loss": 1.2746,
      "num_input_tokens_seen": 539288,
      "step": 970
    },
    {
      "epoch": 17.106194690265486,
      "grad_norm": 8.4375,
      "learning_rate": 4.992688678332437e-05,
      "loss": 1.066,
      "num_input_tokens_seen": 541832,
      "step": 975
    },
    {
      "epoch": 17.194690265486727,
      "grad_norm": 8.1875,
      "learning_rate": 4.992613457837884e-05,
      "loss": 1.0156,
      "num_input_tokens_seen": 544712,
      "step": 980
    },
    {
      "epoch": 17.283185840707965,
      "grad_norm": 7.625,
      "learning_rate": 4.992537852951011e-05,
      "loss": 1.2124,
      "num_input_tokens_seen": 547400,
      "step": 985
    },
    {
      "epoch": 17.371681415929203,
      "grad_norm": 9.625,
      "learning_rate": 4.9924618636834785e-05,
      "loss": 1.3081,
      "num_input_tokens_seen": 550024,
      "step": 990
    },
    {
      "epoch": 17.460176991150444,
      "grad_norm": 8.0,
      "learning_rate": 4.9923854900470046e-05,
      "loss": 0.9356,
      "num_input_tokens_seen": 552904,
      "step": 995
    },
    {
      "epoch": 17.548672566371682,
      "grad_norm": 11.4375,
      "learning_rate": 4.992308732053367e-05,
      "loss": 1.0634,
      "num_input_tokens_seen": 555960,
      "step": 1000
    },
    {
      "epoch": 17.548672566371682,
      "eval_loss": 1.0482069253921509,
      "eval_runtime": 0.9414,
      "eval_samples_per_second": 26.555,
      "eval_steps_per_second": 13.809,
      "num_input_tokens_seen": 555960,
      "step": 1000
    },
    {
      "epoch": 17.63716814159292,
      "grad_norm": 10.125,
      "learning_rate": 4.992231589714402e-05,
      "loss": 0.9961,
      "num_input_tokens_seen": 558952,
      "step": 1005
    },
    {
      "epoch": 17.72566371681416,
      "grad_norm": 8.875,
      "learning_rate": 4.992154063042007e-05,
      "loss": 1.0466,
      "num_input_tokens_seen": 561800,
      "step": 1010
    },
    {
      "epoch": 17.8141592920354,
      "grad_norm": 9.0,
      "learning_rate": 4.992076152048136e-05,
      "loss": 1.1201,
      "num_input_tokens_seen": 564296,
      "step": 1015
    },
    {
      "epoch": 17.902654867256636,
      "grad_norm": 9.0625,
      "learning_rate": 4.991997856744807e-05,
      "loss": 1.066,
      "num_input_tokens_seen": 567464,
      "step": 1020
    },
    {
      "epoch": 17.991150442477878,
      "grad_norm": 10.875,
      "learning_rate": 4.9919191771440905e-05,
      "loss": 0.9586,
      "num_input_tokens_seen": 570328,
      "step": 1025
    },
    {
      "epoch": 18.07079646017699,
      "grad_norm": 8.6875,
      "learning_rate": 4.991840113258122e-05,
      "loss": 1.0453,
      "num_input_tokens_seen": 572328,
      "step": 1030
    },
    {
      "epoch": 18.15929203539823,
      "grad_norm": 8.5625,
      "learning_rate": 4.9917606650990933e-05,
      "loss": 1.2064,
      "num_input_tokens_seen": 574920,
      "step": 1035
    },
    {
      "epoch": 18.24778761061947,
      "grad_norm": 8.3125,
      "learning_rate": 4.9916808326792566e-05,
      "loss": 1.135,
      "num_input_tokens_seen": 577736,
      "step": 1040
    },
    {
      "epoch": 18.336283185840706,
      "grad_norm": 7.96875,
      "learning_rate": 4.9916006160109235e-05,
      "loss": 1.1954,
      "num_input_tokens_seen": 580888,
      "step": 1045
    },
    {
      "epoch": 18.424778761061948,
      "grad_norm": 5.625,
      "learning_rate": 4.991520015106464e-05,
      "loss": 1.0382,
      "num_input_tokens_seen": 583832,
      "step": 1050
    },
    {
      "epoch": 18.513274336283185,
      "grad_norm": 6.6875,
      "learning_rate": 4.991439029978308e-05,
      "loss": 1.1793,
      "num_input_tokens_seen": 586536,
      "step": 1055
    },
    {
      "epoch": 18.601769911504427,
      "grad_norm": 8.5,
      "learning_rate": 4.9913576606389434e-05,
      "loss": 1.0636,
      "num_input_tokens_seen": 589304,
      "step": 1060
    },
    {
      "epoch": 18.690265486725664,
      "grad_norm": 8.875,
      "learning_rate": 4.991275907100919e-05,
      "loss": 0.9863,
      "num_input_tokens_seen": 591864,
      "step": 1065
    },
    {
      "epoch": 18.778761061946902,
      "grad_norm": 7.71875,
      "learning_rate": 4.9911937693768434e-05,
      "loss": 0.9451,
      "num_input_tokens_seen": 594792,
      "step": 1070
    },
    {
      "epoch": 18.86725663716814,
      "grad_norm": 6.28125,
      "learning_rate": 4.991111247479382e-05,
      "loss": 0.9605,
      "num_input_tokens_seen": 597816,
      "step": 1075
    },
    {
      "epoch": 18.95575221238938,
      "grad_norm": 7.9375,
      "learning_rate": 4.9910283414212605e-05,
      "loss": 0.9872,
      "num_input_tokens_seen": 600760,
      "step": 1080
    },
    {
      "epoch": 19.035398230088497,
      "grad_norm": 12.3125,
      "learning_rate": 4.990945051215265e-05,
      "loss": 1.0866,
      "num_input_tokens_seen": 603280,
      "step": 1085
    },
    {
      "epoch": 19.123893805309734,
      "grad_norm": 9.4375,
      "learning_rate": 4.99086137687424e-05,
      "loss": 1.0408,
      "num_input_tokens_seen": 605712,
      "step": 1090
    },
    {
      "epoch": 19.212389380530972,
      "grad_norm": 9.5625,
      "learning_rate": 4.9907773184110874e-05,
      "loss": 0.919,
      "num_input_tokens_seen": 608480,
      "step": 1095
    },
    {
      "epoch": 19.300884955752213,
      "grad_norm": 10.375,
      "learning_rate": 4.9906928758387715e-05,
      "loss": 1.2077,
      "num_input_tokens_seen": 610768,
      "step": 1100
    },
    {
      "epoch": 19.38938053097345,
      "grad_norm": 7.6875,
      "learning_rate": 4.9906080491703146e-05,
      "loss": 1.0166,
      "num_input_tokens_seen": 613696,
      "step": 1105
    },
    {
      "epoch": 19.47787610619469,
      "grad_norm": 11.125,
      "learning_rate": 4.990522838418797e-05,
      "loss": 0.9882,
      "num_input_tokens_seen": 616528,
      "step": 1110
    },
    {
      "epoch": 19.56637168141593,
      "grad_norm": 10.6875,
      "learning_rate": 4.9904372435973604e-05,
      "loss": 1.1176,
      "num_input_tokens_seen": 619472,
      "step": 1115
    },
    {
      "epoch": 19.654867256637168,
      "grad_norm": 8.9375,
      "learning_rate": 4.990351264719203e-05,
      "loss": 1.1831,
      "num_input_tokens_seen": 622384,
      "step": 1120
    },
    {
      "epoch": 19.743362831858406,
      "grad_norm": 9.25,
      "learning_rate": 4.990264901797586e-05,
      "loss": 1.1459,
      "num_input_tokens_seen": 625152,
      "step": 1125
    },
    {
      "epoch": 19.831858407079647,
      "grad_norm": 6.75,
      "learning_rate": 4.990178154845826e-05,
      "loss": 1.0141,
      "num_input_tokens_seen": 627792,
      "step": 1130
    },
    {
      "epoch": 19.920353982300885,
      "grad_norm": 11.0,
      "learning_rate": 4.9900910238773014e-05,
      "loss": 1.1473,
      "num_input_tokens_seen": 630800,
      "step": 1135
    },
    {
      "epoch": 20.0,
      "grad_norm": 11.5625,
      "learning_rate": 4.990003508905448e-05,
      "loss": 0.8872,
      "num_input_tokens_seen": 633448,
      "step": 1140
    },
    {
      "epoch": 20.088495575221238,
      "grad_norm": 7.84375,
      "learning_rate": 4.989915609943763e-05,
      "loss": 1.0042,
      "num_input_tokens_seen": 635912,
      "step": 1145
    },
    {
      "epoch": 20.17699115044248,
      "grad_norm": 7.4375,
      "learning_rate": 4.9898273270058e-05,
      "loss": 0.9343,
      "num_input_tokens_seen": 638936,
      "step": 1150
    },
    {
      "epoch": 20.265486725663717,
      "grad_norm": 7.625,
      "learning_rate": 4.989738660105174e-05,
      "loss": 1.0082,
      "num_input_tokens_seen": 641608,
      "step": 1155
    },
    {
      "epoch": 20.353982300884955,
      "grad_norm": 6.96875,
      "learning_rate": 4.989649609255559e-05,
      "loss": 1.1558,
      "num_input_tokens_seen": 644376,
      "step": 1160
    },
    {
      "epoch": 20.442477876106196,
      "grad_norm": 9.25,
      "learning_rate": 4.989560174470687e-05,
      "loss": 1.2662,
      "num_input_tokens_seen": 647512,
      "step": 1165
    },
    {
      "epoch": 20.530973451327434,
      "grad_norm": 7.3125,
      "learning_rate": 4.989470355764351e-05,
      "loss": 0.9805,
      "num_input_tokens_seen": 650488,
      "step": 1170
    },
    {
      "epoch": 20.61946902654867,
      "grad_norm": 8.875,
      "learning_rate": 4.9893801531504e-05,
      "loss": 1.1524,
      "num_input_tokens_seen": 653096,
      "step": 1175
    },
    {
      "epoch": 20.707964601769913,
      "grad_norm": 12.125,
      "learning_rate": 4.9892895666427475e-05,
      "loss": 1.1158,
      "num_input_tokens_seen": 655528,
      "step": 1180
    },
    {
      "epoch": 20.79646017699115,
      "grad_norm": 10.0625,
      "learning_rate": 4.9891985962553606e-05,
      "loss": 1.1163,
      "num_input_tokens_seen": 658440,
      "step": 1185
    },
    {
      "epoch": 20.884955752212388,
      "grad_norm": 11.0,
      "learning_rate": 4.989107242002269e-05,
      "loss": 0.9704,
      "num_input_tokens_seen": 661256,
      "step": 1190
    },
    {
      "epoch": 20.97345132743363,
      "grad_norm": 7.5,
      "learning_rate": 4.989015503897561e-05,
      "loss": 1.048,
      "num_input_tokens_seen": 663832,
      "step": 1195
    },
    {
      "epoch": 21.053097345132745,
      "grad_norm": 10.9375,
      "learning_rate": 4.988923381955383e-05,
      "loss": 1.0207,
      "num_input_tokens_seen": 665952,
      "step": 1200
    },
    {
      "epoch": 21.053097345132745,
      "eval_loss": 1.0312902927398682,
      "eval_runtime": 0.9402,
      "eval_samples_per_second": 26.589,
      "eval_steps_per_second": 13.826,
      "num_input_tokens_seen": 665952,
      "step": 1200
    },
    {
      "epoch": 21.141592920353983,
      "grad_norm": 7.9375,
      "learning_rate": 4.988830876189942e-05,
      "loss": 0.925,
      "num_input_tokens_seen": 669168,
      "step": 1205
    },
    {
      "epoch": 21.23008849557522,
      "grad_norm": 8.375,
      "learning_rate": 4.988737986615503e-05,
      "loss": 1.2331,
      "num_input_tokens_seen": 672000,
      "step": 1210
    },
    {
      "epoch": 21.31858407079646,
      "grad_norm": 8.375,
      "learning_rate": 4.988644713246391e-05,
      "loss": 1.068,
      "num_input_tokens_seen": 675120,
      "step": 1215
    },
    {
      "epoch": 21.4070796460177,
      "grad_norm": 8.5625,
      "learning_rate": 4.988551056096991e-05,
      "loss": 1.0869,
      "num_input_tokens_seen": 677728,
      "step": 1220
    },
    {
      "epoch": 21.495575221238937,
      "grad_norm": 8.375,
      "learning_rate": 4.988457015181743e-05,
      "loss": 0.9693,
      "num_input_tokens_seen": 680192,
      "step": 1225
    },
    {
      "epoch": 21.58407079646018,
      "grad_norm": 10.25,
      "learning_rate": 4.988362590515153e-05,
      "loss": 1.0582,
      "num_input_tokens_seen": 682688,
      "step": 1230
    },
    {
      "epoch": 21.672566371681416,
      "grad_norm": 7.28125,
      "learning_rate": 4.9882677821117805e-05,
      "loss": 0.9963,
      "num_input_tokens_seen": 686448,
      "step": 1235
    },
    {
      "epoch": 21.761061946902654,
      "grad_norm": 11.0,
      "learning_rate": 4.988172589986246e-05,
      "loss": 1.271,
      "num_input_tokens_seen": 689008,
      "step": 1240
    },
    {
      "epoch": 21.849557522123895,
      "grad_norm": 10.625,
      "learning_rate": 4.9880770141532304e-05,
      "loss": 0.9951,
      "num_input_tokens_seen": 691712,
      "step": 1245
    },
    {
      "epoch": 21.938053097345133,
      "grad_norm": 11.375,
      "learning_rate": 4.987981054627472e-05,
      "loss": 1.02,
      "num_input_tokens_seen": 694496,
      "step": 1250
    },
    {
      "epoch": 22.01769911504425,
      "grad_norm": 6.5,
      "learning_rate": 4.987884711423769e-05,
      "loss": 0.9748,
      "num_input_tokens_seen": 696624,
      "step": 1255
    },
    {
      "epoch": 22.106194690265486,
      "grad_norm": 7.96875,
      "learning_rate": 4.9877879845569784e-05,
      "loss": 0.9616,
      "num_input_tokens_seen": 699600,
      "step": 1260
    },
    {
      "epoch": 22.194690265486727,
      "grad_norm": 8.0625,
      "learning_rate": 4.9876908740420175e-05,
      "loss": 1.1923,
      "num_input_tokens_seen": 702208,
      "step": 1265
    },
    {
      "epoch": 22.283185840707965,
      "grad_norm": 7.28125,
      "learning_rate": 4.987593379893861e-05,
      "loss": 0.8483,
      "num_input_tokens_seen": 705488,
      "step": 1270
    },
    {
      "epoch": 22.371681415929203,
      "grad_norm": 5.96875,
      "learning_rate": 4.987495502127545e-05,
      "loss": 0.9639,
      "num_input_tokens_seen": 708880,
      "step": 1275
    },
    {
      "epoch": 22.460176991150444,
      "grad_norm": 8.5625,
      "learning_rate": 4.987397240758162e-05,
      "loss": 0.9429,
      "num_input_tokens_seen": 711520,
      "step": 1280
    },
    {
      "epoch": 22.548672566371682,
      "grad_norm": 9.3125,
      "learning_rate": 4.9872985958008664e-05,
      "loss": 1.1297,
      "num_input_tokens_seen": 714576,
      "step": 1285
    },
    {
      "epoch": 22.63716814159292,
      "grad_norm": 7.0,
      "learning_rate": 4.987199567270871e-05,
      "loss": 1.005,
      "num_input_tokens_seen": 717200,
      "step": 1290
    },
    {
      "epoch": 22.72566371681416,
      "grad_norm": 7.0625,
      "learning_rate": 4.9871001551834444e-05,
      "loss": 1.3031,
      "num_input_tokens_seen": 719888,
      "step": 1295
    },
    {
      "epoch": 22.8141592920354,
      "grad_norm": 7.5625,
      "learning_rate": 4.98700035955392e-05,
      "loss": 1.0828,
      "num_input_tokens_seen": 722624,
      "step": 1300
    },
    {
      "epoch": 22.902654867256636,
      "grad_norm": 9.1875,
      "learning_rate": 4.986900180397686e-05,
      "loss": 1.1717,
      "num_input_tokens_seen": 725120,
      "step": 1305
    },
    {
      "epoch": 22.991150442477878,
      "grad_norm": 8.5625,
      "learning_rate": 4.9867996177301926e-05,
      "loss": 1.1508,
      "num_input_tokens_seen": 727680,
      "step": 1310
    },
    {
      "epoch": 23.07079646017699,
      "grad_norm": 8.6875,
      "learning_rate": 4.9866986715669464e-05,
      "loss": 1.219,
      "num_input_tokens_seen": 730192,
      "step": 1315
    },
    {
      "epoch": 23.15929203539823,
      "grad_norm": 8.625,
      "learning_rate": 4.9865973419235155e-05,
      "loss": 1.1313,
      "num_input_tokens_seen": 733040,
      "step": 1320
    },
    {
      "epoch": 23.24778761061947,
      "grad_norm": 12.5,
      "learning_rate": 4.986495628815526e-05,
      "loss": 0.9784,
      "num_input_tokens_seen": 736064,
      "step": 1325
    },
    {
      "epoch": 23.336283185840706,
      "grad_norm": 7.3125,
      "learning_rate": 4.986393532258663e-05,
      "loss": 1.0711,
      "num_input_tokens_seen": 738896,
      "step": 1330
    },
    {
      "epoch": 23.424778761061948,
      "grad_norm": 8.4375,
      "learning_rate": 4.986291052268671e-05,
      "loss": 1.0025,
      "num_input_tokens_seen": 741424,
      "step": 1335
    },
    {
      "epoch": 23.513274336283185,
      "grad_norm": 8.3125,
      "learning_rate": 4.986188188861355e-05,
      "loss": 1.0236,
      "num_input_tokens_seen": 744256,
      "step": 1340
    },
    {
      "epoch": 23.601769911504427,
      "grad_norm": 7.09375,
      "learning_rate": 4.9860849420525766e-05,
      "loss": 1.0254,
      "num_input_tokens_seen": 747296,
      "step": 1345
    },
    {
      "epoch": 23.690265486725664,
      "grad_norm": 9.125,
      "learning_rate": 4.9859813118582575e-05,
      "loss": 1.2374,
      "num_input_tokens_seen": 750144,
      "step": 1350
    },
    {
      "epoch": 23.778761061946902,
      "grad_norm": 7.0,
      "learning_rate": 4.98587729829438e-05,
      "loss": 0.9905,
      "num_input_tokens_seen": 752800,
      "step": 1355
    },
    {
      "epoch": 23.86725663716814,
      "grad_norm": 7.40625,
      "learning_rate": 4.985772901376983e-05,
      "loss": 1.1329,
      "num_input_tokens_seen": 755424,
      "step": 1360
    },
    {
      "epoch": 23.95575221238938,
      "grad_norm": 8.1875,
      "learning_rate": 4.9856681211221666e-05,
      "loss": 0.9668,
      "num_input_tokens_seen": 758560,
      "step": 1365
    },
    {
      "epoch": 24.035398230088497,
      "grad_norm": 10.6875,
      "learning_rate": 4.985562957546089e-05,
      "loss": 1.1538,
      "num_input_tokens_seen": 760952,
      "step": 1370
    },
    {
      "epoch": 24.123893805309734,
      "grad_norm": 7.15625,
      "learning_rate": 4.9854574106649686e-05,
      "loss": 1.1139,
      "num_input_tokens_seen": 763640,
      "step": 1375
    },
    {
      "epoch": 24.212389380530972,
      "grad_norm": 10.25,
      "learning_rate": 4.985351480495081e-05,
      "loss": 1.0397,
      "num_input_tokens_seen": 766568,
      "step": 1380
    },
    {
      "epoch": 24.300884955752213,
      "grad_norm": 8.0,
      "learning_rate": 4.985245167052762e-05,
      "loss": 1.1334,
      "num_input_tokens_seen": 769272,
      "step": 1385
    },
    {
      "epoch": 24.38938053097345,
      "grad_norm": 9.5,
      "learning_rate": 4.9851384703544066e-05,
      "loss": 0.994,
      "num_input_tokens_seen": 771864,
      "step": 1390
    },
    {
      "epoch": 24.47787610619469,
      "grad_norm": 9.0,
      "learning_rate": 4.985031390416469e-05,
      "loss": 0.9371,
      "num_input_tokens_seen": 774392,
      "step": 1395
    },
    {
      "epoch": 24.56637168141593,
      "grad_norm": 7.90625,
      "learning_rate": 4.984923927255461e-05,
      "loss": 0.8781,
      "num_input_tokens_seen": 777608,
      "step": 1400
    },
    {
      "epoch": 24.56637168141593,
      "eval_loss": 1.0357418060302734,
      "eval_runtime": 0.9369,
      "eval_samples_per_second": 26.683,
      "eval_steps_per_second": 13.875,
      "num_input_tokens_seen": 777608,
      "step": 1400
    },
    {
      "epoch": 24.654867256637168,
      "grad_norm": 7.6875,
      "learning_rate": 4.984816080887958e-05,
      "loss": 1.137,
      "num_input_tokens_seen": 780376,
      "step": 1405
    },
    {
      "epoch": 24.743362831858406,
      "grad_norm": 9.8125,
      "learning_rate": 4.9847078513305875e-05,
      "loss": 1.1902,
      "num_input_tokens_seen": 783272,
      "step": 1410
    },
    {
      "epoch": 24.831858407079647,
      "grad_norm": 9.75,
      "learning_rate": 4.984599238600043e-05,
      "loss": 1.0681,
      "num_input_tokens_seen": 785816,
      "step": 1415
    },
    {
      "epoch": 24.920353982300885,
      "grad_norm": 8.4375,
      "learning_rate": 4.9844902427130716e-05,
      "loss": 1.1753,
      "num_input_tokens_seen": 788440,
      "step": 1420
    },
    {
      "epoch": 25.0,
      "grad_norm": 7.5625,
      "learning_rate": 4.984380863686482e-05,
      "loss": 1.0449,
      "num_input_tokens_seen": 791152,
      "step": 1425
    },
    {
      "epoch": 25.088495575221238,
      "grad_norm": 6.59375,
      "learning_rate": 4.984271101537143e-05,
      "loss": 0.9099,
      "num_input_tokens_seen": 793840,
      "step": 1430
    },
    {
      "epoch": 25.17699115044248,
      "grad_norm": 7.5,
      "learning_rate": 4.9841609562819816e-05,
      "loss": 0.9884,
      "num_input_tokens_seen": 797040,
      "step": 1435
    },
    {
      "epoch": 25.265486725663717,
      "grad_norm": 10.5,
      "learning_rate": 4.984050427937983e-05,
      "loss": 1.0289,
      "num_input_tokens_seen": 799936,
      "step": 1440
    },
    {
      "epoch": 25.353982300884955,
      "grad_norm": 6.53125,
      "learning_rate": 4.983939516522191e-05,
      "loss": 1.0098,
      "num_input_tokens_seen": 802688,
      "step": 1445
    },
    {
      "epoch": 25.442477876106196,
      "grad_norm": 10.875,
      "learning_rate": 4.983828222051711e-05,
      "loss": 1.398,
      "num_input_tokens_seen": 805232,
      "step": 1450
    },
    {
      "epoch": 25.530973451327434,
      "grad_norm": 8.125,
      "learning_rate": 4.983716544543705e-05,
      "loss": 0.9409,
      "num_input_tokens_seen": 807904,
      "step": 1455
    },
    {
      "epoch": 25.61946902654867,
      "grad_norm": 7.71875,
      "learning_rate": 4.983604484015395e-05,
      "loss": 1.018,
      "num_input_tokens_seen": 811248,
      "step": 1460
    },
    {
      "epoch": 25.707964601769913,
      "grad_norm": 9.25,
      "learning_rate": 4.983492040484064e-05,
      "loss": 1.0736,
      "num_input_tokens_seen": 813600,
      "step": 1465
    },
    {
      "epoch": 25.79646017699115,
      "grad_norm": 7.1875,
      "learning_rate": 4.98337921396705e-05,
      "loss": 1.0907,
      "num_input_tokens_seen": 816736,
      "step": 1470
    },
    {
      "epoch": 25.884955752212388,
      "grad_norm": 9.5,
      "learning_rate": 4.983266004481753e-05,
      "loss": 1.0712,
      "num_input_tokens_seen": 819376,
      "step": 1475
    },
    {
      "epoch": 25.97345132743363,
      "grad_norm": 8.0,
      "learning_rate": 4.9831524120456316e-05,
      "loss": 1.1007,
      "num_input_tokens_seen": 822320,
      "step": 1480
    },
    {
      "epoch": 26.053097345132745,
      "grad_norm": 8.125,
      "learning_rate": 4.9830384366762026e-05,
      "loss": 1.4928,
      "num_input_tokens_seen": 824520,
      "step": 1485
    },
    {
      "epoch": 26.141592920353983,
      "grad_norm": 8.6875,
      "learning_rate": 4.9829240783910436e-05,
      "loss": 1.1321,
      "num_input_tokens_seen": 827512,
      "step": 1490
    },
    {
      "epoch": 26.23008849557522,
      "grad_norm": 11.0,
      "learning_rate": 4.982809337207789e-05,
      "loss": 1.1141,
      "num_input_tokens_seen": 829752,
      "step": 1495
    },
    {
      "epoch": 26.31858407079646,
      "grad_norm": 10.75,
      "learning_rate": 4.9826942131441337e-05,
      "loss": 1.2,
      "num_input_tokens_seen": 832552,
      "step": 1500
    },
    {
      "epoch": 26.4070796460177,
      "grad_norm": 7.9375,
      "learning_rate": 4.9825787062178315e-05,
      "loss": 0.9735,
      "num_input_tokens_seen": 835480,
      "step": 1505
    },
    {
      "epoch": 26.495575221238937,
      "grad_norm": 7.09375,
      "learning_rate": 4.9824628164466945e-05,
      "loss": 1.0058,
      "num_input_tokens_seen": 838520,
      "step": 1510
    },
    {
      "epoch": 26.58407079646018,
      "grad_norm": 8.125,
      "learning_rate": 4.982346543848595e-05,
      "loss": 1.1446,
      "num_input_tokens_seen": 841432,
      "step": 1515
    },
    {
      "epoch": 26.672566371681416,
      "grad_norm": 7.53125,
      "learning_rate": 4.9822298884414626e-05,
      "loss": 0.9709,
      "num_input_tokens_seen": 843928,
      "step": 1520
    },
    {
      "epoch": 26.761061946902654,
      "grad_norm": 12.0,
      "learning_rate": 4.982112850243288e-05,
      "loss": 1.3262,
      "num_input_tokens_seen": 846904,
      "step": 1525
    },
    {
      "epoch": 26.849557522123895,
      "grad_norm": 6.90625,
      "learning_rate": 4.98199542927212e-05,
      "loss": 0.9535,
      "num_input_tokens_seen": 850056,
      "step": 1530
    },
    {
      "epoch": 26.938053097345133,
      "grad_norm": 4.375,
      "learning_rate": 4.981877625546066e-05,
      "loss": 0.9786,
      "num_input_tokens_seen": 852488,
      "step": 1535
    },
    {
      "epoch": 27.01769911504425,
      "grad_norm": 8.6875,
      "learning_rate": 4.981759439083293e-05,
      "loss": 1.1158,
      "num_input_tokens_seen": 854680,
      "step": 1540
    },
    {
      "epoch": 27.106194690265486,
      "grad_norm": 8.25,
      "learning_rate": 4.981640869902027e-05,
      "loss": 1.0113,
      "num_input_tokens_seen": 857608,
      "step": 1545
    },
    {
      "epoch": 27.194690265486727,
      "grad_norm": 8.375,
      "learning_rate": 4.9815219180205517e-05,
      "loss": 1.0476,
      "num_input_tokens_seen": 860456,
      "step": 1550
    },
    {
      "epoch": 27.283185840707965,
      "grad_norm": 8.25,
      "learning_rate": 4.9814025834572126e-05,
      "loss": 1.1957,
      "num_input_tokens_seen": 863112,
      "step": 1555
    },
    {
      "epoch": 27.371681415929203,
      "grad_norm": 9.75,
      "learning_rate": 4.981282866230411e-05,
      "loss": 1.0573,
      "num_input_tokens_seen": 865544,
      "step": 1560
    },
    {
      "epoch": 27.460176991150444,
      "grad_norm": 6.09375,
      "learning_rate": 4.981162766358611e-05,
      "loss": 1.0589,
      "num_input_tokens_seen": 868472,
      "step": 1565
    },
    {
      "epoch": 27.548672566371682,
      "grad_norm": 8.375,
      "learning_rate": 4.9810422838603316e-05,
      "loss": 0.8599,
      "num_input_tokens_seen": 871336,
      "step": 1570
    },
    {
      "epoch": 27.63716814159292,
      "grad_norm": 7.1875,
      "learning_rate": 4.9809214187541533e-05,
      "loss": 1.0889,
      "num_input_tokens_seen": 873992,
      "step": 1575
    },
    {
      "epoch": 27.72566371681416,
      "grad_norm": 6.90625,
      "learning_rate": 4.980800171058715e-05,
      "loss": 1.2083,
      "num_input_tokens_seen": 876792,
      "step": 1580
    },
    {
      "epoch": 27.8141592920354,
      "grad_norm": 7.28125,
      "learning_rate": 4.980678540792715e-05,
      "loss": 1.1742,
      "num_input_tokens_seen": 879608,
      "step": 1585
    },
    {
      "epoch": 27.902654867256636,
      "grad_norm": 5.625,
      "learning_rate": 4.980556527974909e-05,
      "loss": 0.844,
      "num_input_tokens_seen": 882760,
      "step": 1590
    },
    {
      "epoch": 27.991150442477878,
      "grad_norm": 9.75,
      "learning_rate": 4.980434132624114e-05,
      "loss": 1.1307,
      "num_input_tokens_seen": 885640,
      "step": 1595
    },
    {
      "epoch": 28.07079646017699,
      "grad_norm": 8.4375,
      "learning_rate": 4.980311354759205e-05,
      "loss": 1.1403,
      "num_input_tokens_seen": 887904,
      "step": 1600
    },
    {
      "epoch": 28.07079646017699,
      "eval_loss": 1.0395183563232422,
      "eval_runtime": 0.941,
      "eval_samples_per_second": 26.566,
      "eval_steps_per_second": 13.815,
      "num_input_tokens_seen": 887904,
      "step": 1600
    },
    {
      "epoch": 28.15929203539823,
      "grad_norm": 8.25,
      "learning_rate": 4.980188194399116e-05,
      "loss": 1.0493,
      "num_input_tokens_seen": 891024,
      "step": 1605
    },
    {
      "epoch": 28.24778761061947,
      "grad_norm": 7.90625,
      "learning_rate": 4.9800646515628384e-05,
      "loss": 0.944,
      "num_input_tokens_seen": 893936,
      "step": 1610
    },
    {
      "epoch": 28.336283185840706,
      "grad_norm": 11.6875,
      "learning_rate": 4.979940726269426e-05,
      "loss": 1.2937,
      "num_input_tokens_seen": 896784,
      "step": 1615
    },
    {
      "epoch": 28.424778761061948,
      "grad_norm": 6.8125,
      "learning_rate": 4.979816418537988e-05,
      "loss": 1.038,
      "num_input_tokens_seen": 899152,
      "step": 1620
    },
    {
      "epoch": 28.513274336283185,
      "grad_norm": 7.71875,
      "learning_rate": 4.979691728387696e-05,
      "loss": 1.0823,
      "num_input_tokens_seen": 902176,
      "step": 1625
    },
    {
      "epoch": 28.601769911504427,
      "grad_norm": 8.625,
      "learning_rate": 4.979566655837776e-05,
      "loss": 1.0829,
      "num_input_tokens_seen": 904704,
      "step": 1630
    },
    {
      "epoch": 28.690265486725664,
      "grad_norm": 15.375,
      "learning_rate": 4.9794412009075184e-05,
      "loss": 1.2817,
      "num_input_tokens_seen": 907632,
      "step": 1635
    },
    {
      "epoch": 28.778761061946902,
      "grad_norm": 8.9375,
      "learning_rate": 4.979315363616269e-05,
      "loss": 1.0367,
      "num_input_tokens_seen": 910448,
      "step": 1640
    },
    {
      "epoch": 28.86725663716814,
      "grad_norm": 8.1875,
      "learning_rate": 4.979189143983434e-05,
      "loss": 0.9599,
      "num_input_tokens_seen": 913440,
      "step": 1645
    },
    {
      "epoch": 28.95575221238938,
      "grad_norm": 12.1875,
      "learning_rate": 4.979062542028478e-05,
      "loss": 0.8981,
      "num_input_tokens_seen": 916464,
      "step": 1650
    },
    {
      "epoch": 29.035398230088497,
      "grad_norm": 7.0,
      "learning_rate": 4.978935557770923e-05,
      "loss": 0.9245,
      "num_input_tokens_seen": 918528,
      "step": 1655
    },
    {
      "epoch": 29.123893805309734,
      "grad_norm": 11.3125,
      "learning_rate": 4.978808191230353e-05,
      "loss": 1.1781,
      "num_input_tokens_seen": 921312,
      "step": 1660
    },
    {
      "epoch": 29.212389380530972,
      "grad_norm": 8.1875,
      "learning_rate": 4.9786804424264085e-05,
      "loss": 1.1253,
      "num_input_tokens_seen": 924352,
      "step": 1665
    },
    {
      "epoch": 29.300884955752213,
      "grad_norm": 8.5625,
      "learning_rate": 4.978552311378792e-05,
      "loss": 1.0448,
      "num_input_tokens_seen": 927392,
      "step": 1670
    },
    {
      "epoch": 29.38938053097345,
      "grad_norm": 7.34375,
      "learning_rate": 4.978423798107261e-05,
      "loss": 0.9992,
      "num_input_tokens_seen": 929888,
      "step": 1675
    },
    {
      "epoch": 29.47787610619469,
      "grad_norm": 6.53125,
      "learning_rate": 4.978294902631635e-05,
      "loss": 1.0099,
      "num_input_tokens_seen": 932752,
      "step": 1680
    },
    {
      "epoch": 29.56637168141593,
      "grad_norm": 9.125,
      "learning_rate": 4.9781656249717914e-05,
      "loss": 1.0536,
      "num_input_tokens_seen": 935728,
      "step": 1685
    },
    {
      "epoch": 29.654867256637168,
      "grad_norm": 8.375,
      "learning_rate": 4.9780359651476645e-05,
      "loss": 1.1189,
      "num_input_tokens_seen": 938240,
      "step": 1690
    },
    {
      "epoch": 29.743362831858406,
      "grad_norm": 8.5,
      "learning_rate": 4.977905923179251e-05,
      "loss": 1.1198,
      "num_input_tokens_seen": 940992,
      "step": 1695
    },
    {
      "epoch": 29.831858407079647,
      "grad_norm": 8.3125,
      "learning_rate": 4.977775499086606e-05,
      "loss": 0.9366,
      "num_input_tokens_seen": 944256,
      "step": 1700
    },
    {
      "epoch": 29.920353982300885,
      "grad_norm": 11.5625,
      "learning_rate": 4.97764469288984e-05,
      "loss": 1.2763,
      "num_input_tokens_seen": 946832,
      "step": 1705
    },
    {
      "epoch": 30.0,
      "grad_norm": 14.0,
      "learning_rate": 4.977513504609127e-05,
      "loss": 0.9251,
      "num_input_tokens_seen": 949256,
      "step": 1710
    },
    {
      "epoch": 30.088495575221238,
      "grad_norm": 9.375,
      "learning_rate": 4.9773819342646965e-05,
      "loss": 0.9223,
      "num_input_tokens_seen": 951912,
      "step": 1715
    },
    {
      "epoch": 30.17699115044248,
      "grad_norm": 7.09375,
      "learning_rate": 4.97724998187684e-05,
      "loss": 1.1281,
      "num_input_tokens_seen": 954344,
      "step": 1720
    },
    {
      "epoch": 30.265486725663717,
      "grad_norm": 10.9375,
      "learning_rate": 4.9771176474659045e-05,
      "loss": 1.0481,
      "num_input_tokens_seen": 957576,
      "step": 1725
    },
    {
      "epoch": 30.353982300884955,
      "grad_norm": 9.4375,
      "learning_rate": 4.976984931052299e-05,
      "loss": 1.1988,
      "num_input_tokens_seen": 960104,
      "step": 1730
    },
    {
      "epoch": 30.442477876106196,
      "grad_norm": 7.25,
      "learning_rate": 4.976851832656489e-05,
      "loss": 0.9444,
      "num_input_tokens_seen": 963224,
      "step": 1735
    },
    {
      "epoch": 30.530973451327434,
      "grad_norm": 10.25,
      "learning_rate": 4.9767183522990004e-05,
      "loss": 1.0891,
      "num_input_tokens_seen": 966184,
      "step": 1740
    },
    {
      "epoch": 30.61946902654867,
      "grad_norm": 11.4375,
      "learning_rate": 4.9765844900004176e-05,
      "loss": 1.192,
      "num_input_tokens_seen": 969128,
      "step": 1745
    },
    {
      "epoch": 30.707964601769913,
      "grad_norm": 7.5,
      "learning_rate": 4.9764502457813834e-05,
      "loss": 1.0545,
      "num_input_tokens_seen": 971960,
      "step": 1750
    },
    {
      "epoch": 30.79646017699115,
      "grad_norm": 7.53125,
      "learning_rate": 4.9763156196626005e-05,
      "loss": 1.0994,
      "num_input_tokens_seen": 974952,
      "step": 1755
    },
    {
      "epoch": 30.884955752212388,
      "grad_norm": 6.46875,
      "learning_rate": 4.97618061166483e-05,
      "loss": 1.0485,
      "num_input_tokens_seen": 977752,
      "step": 1760
    },
    {
      "epoch": 30.97345132743363,
      "grad_norm": 7.25,
      "learning_rate": 4.9760452218088915e-05,
      "loss": 1.0559,
      "num_input_tokens_seen": 980232,
      "step": 1765
    },
    {
      "epoch": 31.053097345132745,
      "grad_norm": 6.34375,
      "learning_rate": 4.975909450115663e-05,
      "loss": 1.0037,
      "num_input_tokens_seen": 982648,
      "step": 1770
    },
    {
      "epoch": 31.141592920353983,
      "grad_norm": 9.1875,
      "learning_rate": 4.975773296606084e-05,
      "loss": 1.0364,
      "num_input_tokens_seen": 985768,
      "step": 1775
    },
    {
      "epoch": 31.23008849557522,
      "grad_norm": 8.125,
      "learning_rate": 4.97563676130115e-05,
      "loss": 1.1497,
      "num_input_tokens_seen": 988840,
      "step": 1780
    },
    {
      "epoch": 31.31858407079646,
      "grad_norm": 7.375,
      "learning_rate": 4.9754998442219166e-05,
      "loss": 1.0466,
      "num_input_tokens_seen": 991656,
      "step": 1785
    },
    {
      "epoch": 31.4070796460177,
      "grad_norm": 8.5625,
      "learning_rate": 4.9753625453894984e-05,
      "loss": 1.0299,
      "num_input_tokens_seen": 994584,
      "step": 1790
    },
    {
      "epoch": 31.495575221238937,
      "grad_norm": 6.875,
      "learning_rate": 4.975224864825068e-05,
      "loss": 1.0418,
      "num_input_tokens_seen": 997000,
      "step": 1795
    },
    {
      "epoch": 31.58407079646018,
      "grad_norm": 8.5,
      "learning_rate": 4.9750868025498576e-05,
      "loss": 1.2085,
      "num_input_tokens_seen": 999464,
      "step": 1800
    },
    {
      "epoch": 31.58407079646018,
      "eval_loss": 1.0357142686843872,
      "eval_runtime": 0.9307,
      "eval_samples_per_second": 26.861,
      "eval_steps_per_second": 13.967,
      "num_input_tokens_seen": 999464,
      "step": 1800
    },
    {
      "epoch": 31.672566371681416,
      "grad_norm": 8.375,
      "learning_rate": 4.974948358585158e-05,
      "loss": 0.9874,
      "num_input_tokens_seen": 1002152,
      "step": 1805
    },
    {
      "epoch": 31.761061946902654,
      "grad_norm": 12.6875,
      "learning_rate": 4.9748095329523205e-05,
      "loss": 1.1453,
      "num_input_tokens_seen": 1004728,
      "step": 1810
    },
    {
      "epoch": 31.849557522123895,
      "grad_norm": 8.4375,
      "learning_rate": 4.974670325672752e-05,
      "loss": 1.1323,
      "num_input_tokens_seen": 1007656,
      "step": 1815
    },
    {
      "epoch": 31.938053097345133,
      "grad_norm": 7.71875,
      "learning_rate": 4.974530736767921e-05,
      "loss": 1.0198,
      "num_input_tokens_seen": 1010488,
      "step": 1820
    },
    {
      "epoch": 32.017699115044245,
      "grad_norm": 8.875,
      "learning_rate": 4.9743907662593524e-05,
      "loss": 0.8373,
      "num_input_tokens_seen": 1012952,
      "step": 1825
    },
    {
      "epoch": 32.10619469026549,
      "grad_norm": 7.875,
      "learning_rate": 4.974250414168633e-05,
      "loss": 0.9198,
      "num_input_tokens_seen": 1015512,
      "step": 1830
    },
    {
      "epoch": 32.19469026548673,
      "grad_norm": 8.625,
      "learning_rate": 4.974109680517407e-05,
      "loss": 1.0892,
      "num_input_tokens_seen": 1018664,
      "step": 1835
    },
    {
      "epoch": 32.283185840707965,
      "grad_norm": 6.8125,
      "learning_rate": 4.973968565327376e-05,
      "loss": 0.9898,
      "num_input_tokens_seen": 1021176,
      "step": 1840
    },
    {
      "epoch": 32.3716814159292,
      "grad_norm": 10.0625,
      "learning_rate": 4.973827068620303e-05,
      "loss": 1.0656,
      "num_input_tokens_seen": 1024008,
      "step": 1845
    },
    {
      "epoch": 32.46017699115044,
      "grad_norm": 7.96875,
      "learning_rate": 4.973685190418008e-05,
      "loss": 1.1066,
      "num_input_tokens_seen": 1026520,
      "step": 1850
    },
    {
      "epoch": 32.54867256637168,
      "grad_norm": 8.125,
      "learning_rate": 4.97354293074237e-05,
      "loss": 0.9665,
      "num_input_tokens_seen": 1029656,
      "step": 1855
    },
    {
      "epoch": 32.63716814159292,
      "grad_norm": 6.375,
      "learning_rate": 4.9734002896153276e-05,
      "loss": 1.0488,
      "num_input_tokens_seen": 1032808,
      "step": 1860
    },
    {
      "epoch": 32.72566371681416,
      "grad_norm": 6.78125,
      "learning_rate": 4.973257267058877e-05,
      "loss": 0.9107,
      "num_input_tokens_seen": 1035944,
      "step": 1865
    },
    {
      "epoch": 32.8141592920354,
      "grad_norm": 8.375,
      "learning_rate": 4.973113863095076e-05,
      "loss": 1.2498,
      "num_input_tokens_seen": 1038792,
      "step": 1870
    },
    {
      "epoch": 32.902654867256636,
      "grad_norm": 9.625,
      "learning_rate": 4.9729700777460384e-05,
      "loss": 1.047,
      "num_input_tokens_seen": 1041480,
      "step": 1875
    },
    {
      "epoch": 32.991150442477874,
      "grad_norm": 8.4375,
      "learning_rate": 4.972825911033937e-05,
      "loss": 1.2335,
      "num_input_tokens_seen": 1043896,
      "step": 1880
    },
    {
      "epoch": 33.07079646017699,
      "grad_norm": 7.96875,
      "learning_rate": 4.9726813629810056e-05,
      "loss": 1.2356,
      "num_input_tokens_seen": 1046456,
      "step": 1885
    },
    {
      "epoch": 33.15929203539823,
      "grad_norm": 9.625,
      "learning_rate": 4.9725364336095326e-05,
      "loss": 1.1956,
      "num_input_tokens_seen": 1049272,
      "step": 1890
    },
    {
      "epoch": 33.24778761061947,
      "grad_norm": 9.375,
      "learning_rate": 4.972391122941871e-05,
      "loss": 1.2616,
      "num_input_tokens_seen": 1051784,
      "step": 1895
    },
    {
      "epoch": 33.336283185840706,
      "grad_norm": 9.0,
      "learning_rate": 4.972245431000428e-05,
      "loss": 0.9355,
      "num_input_tokens_seen": 1054504,
      "step": 1900
    },
    {
      "epoch": 33.424778761061944,
      "grad_norm": 8.375,
      "learning_rate": 4.972099357807671e-05,
      "loss": 1.0065,
      "num_input_tokens_seen": 1057304,
      "step": 1905
    },
    {
      "epoch": 33.51327433628319,
      "grad_norm": 6.375,
      "learning_rate": 4.971952903386127e-05,
      "loss": 1.1536,
      "num_input_tokens_seen": 1060072,
      "step": 1910
    },
    {
      "epoch": 33.60176991150443,
      "grad_norm": 7.375,
      "learning_rate": 4.971806067758381e-05,
      "loss": 0.9852,
      "num_input_tokens_seen": 1062872,
      "step": 1915
    },
    {
      "epoch": 33.690265486725664,
      "grad_norm": 6.375,
      "learning_rate": 4.971658850947076e-05,
      "loss": 0.8722,
      "num_input_tokens_seen": 1065944,
      "step": 1920
    },
    {
      "epoch": 33.7787610619469,
      "grad_norm": 7.9375,
      "learning_rate": 4.9715112529749165e-05,
      "loss": 0.8978,
      "num_input_tokens_seen": 1068328,
      "step": 1925
    },
    {
      "epoch": 33.86725663716814,
      "grad_norm": 10.625,
      "learning_rate": 4.9713632738646624e-05,
      "loss": 1.1624,
      "num_input_tokens_seen": 1071512,
      "step": 1930
    },
    {
      "epoch": 33.95575221238938,
      "grad_norm": 12.625,
      "learning_rate": 4.971214913639134e-05,
      "loss": 1.127,
      "num_input_tokens_seen": 1074136,
      "step": 1935
    },
    {
      "epoch": 34.0353982300885,
      "grad_norm": 8.3125,
      "learning_rate": 4.9710661723212104e-05,
      "loss": 1.2572,
      "num_input_tokens_seen": 1076632,
      "step": 1940
    },
    {
      "epoch": 34.123893805309734,
      "grad_norm": 9.4375,
      "learning_rate": 4.9709170499338295e-05,
      "loss": 1.1555,
      "num_input_tokens_seen": 1079688,
      "step": 1945
    },
    {
      "epoch": 34.21238938053097,
      "grad_norm": 8.6875,
      "learning_rate": 4.9707675464999895e-05,
      "loss": 1.1171,
      "num_input_tokens_seen": 1082648,
      "step": 1950
    },
    {
      "epoch": 34.30088495575221,
      "grad_norm": 11.375,
      "learning_rate": 4.970617662042743e-05,
      "loss": 1.1405,
      "num_input_tokens_seen": 1085464,
      "step": 1955
    },
    {
      "epoch": 34.389380530973455,
      "grad_norm": 7.25,
      "learning_rate": 4.970467396585206e-05,
      "loss": 1.0895,
      "num_input_tokens_seen": 1088328,
      "step": 1960
    },
    {
      "epoch": 34.47787610619469,
      "grad_norm": 8.75,
      "learning_rate": 4.97031675015055e-05,
      "loss": 0.864,
      "num_input_tokens_seen": 1091880,
      "step": 1965
    },
    {
      "epoch": 34.56637168141593,
      "grad_norm": 6.75,
      "learning_rate": 4.9701657227620075e-05,
      "loss": 1.0265,
      "num_input_tokens_seen": 1094328,
      "step": 1970
    },
    {
      "epoch": 34.65486725663717,
      "grad_norm": 8.375,
      "learning_rate": 4.9700143144428685e-05,
      "loss": 1.0528,
      "num_input_tokens_seen": 1096728,
      "step": 1975
    },
    {
      "epoch": 34.743362831858406,
      "grad_norm": 9.9375,
      "learning_rate": 4.969862525216482e-05,
      "loss": 0.892,
      "num_input_tokens_seen": 1099640,
      "step": 1980
    },
    {
      "epoch": 34.83185840707964,
      "grad_norm": 9.125,
      "learning_rate": 4.9697103551062556e-05,
      "loss": 1.2152,
      "num_input_tokens_seen": 1102600,
      "step": 1985
    },
    {
      "epoch": 34.92035398230089,
      "grad_norm": 8.625,
      "learning_rate": 4.9695578041356565e-05,
      "loss": 1.1194,
      "num_input_tokens_seen": 1105080,
      "step": 1990
    },
    {
      "epoch": 35.0,
      "grad_norm": 9.25,
      "learning_rate": 4.969404872328209e-05,
      "loss": 0.8329,
      "num_input_tokens_seen": 1107456,
      "step": 1995
    },
    {
      "epoch": 35.08849557522124,
      "grad_norm": 8.25,
      "learning_rate": 4.969251559707498e-05,
      "loss": 1.01,
      "num_input_tokens_seen": 1110640,
      "step": 2000
    },
    {
      "epoch": 35.08849557522124,
      "eval_loss": 1.036898136138916,
      "eval_runtime": 0.9316,
      "eval_samples_per_second": 26.834,
      "eval_steps_per_second": 13.954,
      "num_input_tokens_seen": 1110640,
      "step": 2000
    },
    {
      "epoch": 35.176991150442475,
      "grad_norm": 7.03125,
      "learning_rate": 4.9690978662971674e-05,
      "loss": 0.9603,
      "num_input_tokens_seen": 1113104,
      "step": 2005
    },
    {
      "epoch": 35.26548672566372,
      "grad_norm": 7.625,
      "learning_rate": 4.968943792120916e-05,
      "loss": 1.2564,
      "num_input_tokens_seen": 1115968,
      "step": 2010
    },
    {
      "epoch": 35.35398230088496,
      "grad_norm": 6.9375,
      "learning_rate": 4.9687893372025046e-05,
      "loss": 1.0183,
      "num_input_tokens_seen": 1118912,
      "step": 2015
    },
    {
      "epoch": 35.442477876106196,
      "grad_norm": 8.5625,
      "learning_rate": 4.9686345015657535e-05,
      "loss": 1.1004,
      "num_input_tokens_seen": 1121792,
      "step": 2020
    },
    {
      "epoch": 35.530973451327434,
      "grad_norm": 7.59375,
      "learning_rate": 4.968479285234538e-05,
      "loss": 1.0572,
      "num_input_tokens_seen": 1124720,
      "step": 2025
    },
    {
      "epoch": 35.61946902654867,
      "grad_norm": 7.84375,
      "learning_rate": 4.9683236882327974e-05,
      "loss": 1.1318,
      "num_input_tokens_seen": 1127392,
      "step": 2030
    },
    {
      "epoch": 35.70796460176991,
      "grad_norm": 7.90625,
      "learning_rate": 4.968167710584526e-05,
      "loss": 0.9216,
      "num_input_tokens_seen": 1129952,
      "step": 2035
    },
    {
      "epoch": 35.796460176991154,
      "grad_norm": 8.875,
      "learning_rate": 4.968011352313775e-05,
      "loss": 1.1395,
      "num_input_tokens_seen": 1132624,
      "step": 2040
    },
    {
      "epoch": 35.88495575221239,
      "grad_norm": 10.9375,
      "learning_rate": 4.967854613444659e-05,
      "loss": 1.2753,
      "num_input_tokens_seen": 1135200,
      "step": 2045
    },
    {
      "epoch": 35.97345132743363,
      "grad_norm": 6.84375,
      "learning_rate": 4.967697494001349e-05,
      "loss": 0.9162,
      "num_input_tokens_seen": 1138352,
      "step": 2050
    },
    {
      "epoch": 36.05309734513274,
      "grad_norm": 6.875,
      "learning_rate": 4.9675399940080736e-05,
      "loss": 1.0396,
      "num_input_tokens_seen": 1140704,
      "step": 2055
    },
    {
      "epoch": 36.14159292035398,
      "grad_norm": 7.03125,
      "learning_rate": 4.9673821134891226e-05,
      "loss": 1.1045,
      "num_input_tokens_seen": 1144048,
      "step": 2060
    },
    {
      "epoch": 36.230088495575224,
      "grad_norm": 7.40625,
      "learning_rate": 4.967223852468842e-05,
      "loss": 1.0029,
      "num_input_tokens_seen": 1146752,
      "step": 2065
    },
    {
      "epoch": 36.31858407079646,
      "grad_norm": 8.9375,
      "learning_rate": 4.967065210971639e-05,
      "loss": 1.0706,
      "num_input_tokens_seen": 1149552,
      "step": 2070
    },
    {
      "epoch": 36.4070796460177,
      "grad_norm": 8.75,
      "learning_rate": 4.966906189021977e-05,
      "loss": 1.2111,
      "num_input_tokens_seen": 1152176,
      "step": 2075
    },
    {
      "epoch": 36.49557522123894,
      "grad_norm": 6.6875,
      "learning_rate": 4.966746786644379e-05,
      "loss": 1.0159,
      "num_input_tokens_seen": 1154928,
      "step": 2080
    },
    {
      "epoch": 36.584070796460175,
      "grad_norm": 8.375,
      "learning_rate": 4.966587003863429e-05,
      "loss": 0.9057,
      "num_input_tokens_seen": 1157520,
      "step": 2085
    },
    {
      "epoch": 36.67256637168141,
      "grad_norm": 8.375,
      "learning_rate": 4.966426840703765e-05,
      "loss": 1.1692,
      "num_input_tokens_seen": 1160656,
      "step": 2090
    },
    {
      "epoch": 36.76106194690266,
      "grad_norm": 8.8125,
      "learning_rate": 4.9662662971900875e-05,
      "loss": 1.0794,
      "num_input_tokens_seen": 1163376,
      "step": 2095
    },
    {
      "epoch": 36.849557522123895,
      "grad_norm": 6.21875,
      "learning_rate": 4.9661053733471534e-05,
      "loss": 1.1199,
      "num_input_tokens_seen": 1165936,
      "step": 2100
    },
    {
      "epoch": 36.93805309734513,
      "grad_norm": 11.625,
      "learning_rate": 4.965944069199781e-05,
      "loss": 0.9257,
      "num_input_tokens_seen": 1169040,
      "step": 2105
    },
    {
      "epoch": 37.017699115044245,
      "grad_norm": 6.8125,
      "learning_rate": 4.965782384772842e-05,
      "loss": 1.0265,
      "num_input_tokens_seen": 1171560,
      "step": 2110
    },
    {
      "epoch": 37.10619469026549,
      "grad_norm": 7.5625,
      "learning_rate": 4.9656203200912734e-05,
      "loss": 0.9614,
      "num_input_tokens_seen": 1174840,
      "step": 2115
    },
    {
      "epoch": 37.19469026548673,
      "grad_norm": 8.9375,
      "learning_rate": 4.965457875180067e-05,
      "loss": 1.0877,
      "num_input_tokens_seen": 1177608,
      "step": 2120
    },
    {
      "epoch": 37.283185840707965,
      "grad_norm": 7.90625,
      "learning_rate": 4.9652950500642724e-05,
      "loss": 1.0401,
      "num_input_tokens_seen": 1180408,
      "step": 2125
    },
    {
      "epoch": 37.3716814159292,
      "grad_norm": 9.25,
      "learning_rate": 4.965131844769001e-05,
      "loss": 1.0571,
      "num_input_tokens_seen": 1183288,
      "step": 2130
    },
    {
      "epoch": 37.46017699115044,
      "grad_norm": 7.625,
      "learning_rate": 4.96496825931942e-05,
      "loss": 1.0746,
      "num_input_tokens_seen": 1185784,
      "step": 2135
    },
    {
      "epoch": 37.54867256637168,
      "grad_norm": 11.6875,
      "learning_rate": 4.9648042937407566e-05,
      "loss": 1.0645,
      "num_input_tokens_seen": 1188424,
      "step": 2140
    },
    {
      "epoch": 37.63716814159292,
      "grad_norm": 7.9375,
      "learning_rate": 4.964639948058297e-05,
      "loss": 1.1439,
      "num_input_tokens_seen": 1192104,
      "step": 2145
    },
    {
      "epoch": 37.72566371681416,
      "grad_norm": 8.0625,
      "learning_rate": 4.9644752222973846e-05,
      "loss": 1.1344,
      "num_input_tokens_seen": 1194664,
      "step": 2150
    },
    {
      "epoch": 37.8141592920354,
      "grad_norm": 7.90625,
      "learning_rate": 4.964310116483422e-05,
      "loss": 1.1902,
      "num_input_tokens_seen": 1197608,
      "step": 2155
    },
    {
      "epoch": 37.902654867256636,
      "grad_norm": 6.125,
      "learning_rate": 4.964144630641872e-05,
      "loss": 0.9806,
      "num_input_tokens_seen": 1200008,
      "step": 2160
    },
    {
      "epoch": 37.991150442477874,
      "grad_norm": 9.1875,
      "learning_rate": 4.9639787647982525e-05,
      "loss": 1.0412,
      "num_input_tokens_seen": 1202712,
      "step": 2165
    },
    {
      "epoch": 38.07079646017699,
      "grad_norm": 11.625,
      "learning_rate": 4.963812518978143e-05,
      "loss": 1.0241,
      "num_input_tokens_seen": 1205056,
      "step": 2170
    },
    {
      "epoch": 38.15929203539823,
      "grad_norm": 9.875,
      "learning_rate": 4.963645893207182e-05,
      "loss": 1.0711,
      "num_input_tokens_seen": 1207968,
      "step": 2175
    },
    {
      "epoch": 38.24778761061947,
      "grad_norm": 8.3125,
      "learning_rate": 4.963478887511063e-05,
      "loss": 0.9696,
      "num_input_tokens_seen": 1211024,
      "step": 2180
    },
    {
      "epoch": 38.336283185840706,
      "grad_norm": 7.78125,
      "learning_rate": 4.963311501915542e-05,
      "loss": 1.1743,
      "num_input_tokens_seen": 1213648,
      "step": 2185
    },
    {
      "epoch": 38.424778761061944,
      "grad_norm": 7.0625,
      "learning_rate": 4.963143736446432e-05,
      "loss": 1.1472,
      "num_input_tokens_seen": 1216400,
      "step": 2190
    },
    {
      "epoch": 38.51327433628319,
      "grad_norm": 8.625,
      "learning_rate": 4.962975591129603e-05,
      "loss": 0.9087,
      "num_input_tokens_seen": 1218784,
      "step": 2195
    },
    {
      "epoch": 38.60176991150443,
      "grad_norm": 11.125,
      "learning_rate": 4.962807065990986e-05,
      "loss": 1.0684,
      "num_input_tokens_seen": 1222144,
      "step": 2200
    },
    {
      "epoch": 38.60176991150443,
      "eval_loss": 1.034595251083374,
      "eval_runtime": 0.9317,
      "eval_samples_per_second": 26.833,
      "eval_steps_per_second": 13.953,
      "num_input_tokens_seen": 1222144,
      "step": 2200
    },
    {
      "epoch": 38.690265486725664,
      "grad_norm": 9.4375,
      "learning_rate": 4.9626381610565714e-05,
      "loss": 1.1904,
      "num_input_tokens_seen": 1224752,
      "step": 2205
    },
    {
      "epoch": 38.7787610619469,
      "grad_norm": 8.875,
      "learning_rate": 4.9624688763524043e-05,
      "loss": 1.2077,
      "num_input_tokens_seen": 1227456,
      "step": 2210
    },
    {
      "epoch": 38.86725663716814,
      "grad_norm": 10.0,
      "learning_rate": 4.962299211904591e-05,
      "loss": 0.9789,
      "num_input_tokens_seen": 1230512,
      "step": 2215
    },
    {
      "epoch": 38.95575221238938,
      "grad_norm": 7.34375,
      "learning_rate": 4.962129167739296e-05,
      "loss": 0.9492,
      "num_input_tokens_seen": 1233136,
      "step": 2220
    },
    {
      "epoch": 39.0353982300885,
      "grad_norm": 8.125,
      "learning_rate": 4.961958743882742e-05,
      "loss": 0.9823,
      "num_input_tokens_seen": 1235624,
      "step": 2225
    },
    {
      "epoch": 39.123893805309734,
      "grad_norm": 9.125,
      "learning_rate": 4.961787940361211e-05,
      "loss": 1.2504,
      "num_input_tokens_seen": 1238184,
      "step": 2230
    },
    {
      "epoch": 39.21238938053097,
      "grad_norm": 6.78125,
      "learning_rate": 4.961616757201043e-05,
      "loss": 0.9359,
      "num_input_tokens_seen": 1241400,
      "step": 2235
    },
    {
      "epoch": 39.30088495575221,
      "grad_norm": 6.71875,
      "learning_rate": 4.961445194428637e-05,
      "loss": 1.0819,
      "num_input_tokens_seen": 1243720,
      "step": 2240
    },
    {
      "epoch": 39.389380530973455,
      "grad_norm": 10.1875,
      "learning_rate": 4.9612732520704486e-05,
      "loss": 1.0039,
      "num_input_tokens_seen": 1246504,
      "step": 2245
    },
    {
      "epoch": 39.47787610619469,
      "grad_norm": 6.9375,
      "learning_rate": 4.961100930152994e-05,
      "loss": 1.0412,
      "num_input_tokens_seen": 1249576,
      "step": 2250
    },
    {
      "epoch": 39.56637168141593,
      "grad_norm": 13.4375,
      "learning_rate": 4.960928228702849e-05,
      "loss": 1.1657,
      "num_input_tokens_seen": 1252808,
      "step": 2255
    },
    {
      "epoch": 39.65486725663717,
      "grad_norm": 7.71875,
      "learning_rate": 4.960755147746645e-05,
      "loss": 0.9832,
      "num_input_tokens_seen": 1255400,
      "step": 2260
    },
    {
      "epoch": 39.743362831858406,
      "grad_norm": 9.625,
      "learning_rate": 4.9605816873110736e-05,
      "loss": 1.2113,
      "num_input_tokens_seen": 1258408,
      "step": 2265
    },
    {
      "epoch": 39.83185840707964,
      "grad_norm": 8.5,
      "learning_rate": 4.960407847422883e-05,
      "loss": 1.2047,
      "num_input_tokens_seen": 1261016,
      "step": 2270
    },
    {
      "epoch": 39.92035398230089,
      "grad_norm": 7.84375,
      "learning_rate": 4.960233628108885e-05,
      "loss": 0.8355,
      "num_input_tokens_seen": 1263720,
      "step": 2275
    },
    {
      "epoch": 40.0,
      "grad_norm": 7.78125,
      "learning_rate": 4.960059029395942e-05,
      "loss": 0.9988,
      "num_input_tokens_seen": 1266024,
      "step": 2280
    },
    {
      "epoch": 40.08849557522124,
      "grad_norm": 8.5,
      "learning_rate": 4.959884051310983e-05,
      "loss": 1.0047,
      "num_input_tokens_seen": 1268904,
      "step": 2285
    },
    {
      "epoch": 40.176991150442475,
      "grad_norm": 10.3125,
      "learning_rate": 4.959708693880991e-05,
      "loss": 1.26,
      "num_input_tokens_seen": 1271736,
      "step": 2290
    },
    {
      "epoch": 40.26548672566372,
      "grad_norm": 7.40625,
      "learning_rate": 4.9595329571330074e-05,
      "loss": 1.004,
      "num_input_tokens_seen": 1274952,
      "step": 2295
    },
    {
      "epoch": 40.35398230088496,
      "grad_norm": 7.65625,
      "learning_rate": 4.9593568410941326e-05,
      "loss": 0.9221,
      "num_input_tokens_seen": 1278184,
      "step": 2300
    },
    {
      "epoch": 40.442477876106196,
      "grad_norm": 9.75,
      "learning_rate": 4.959180345791528e-05,
      "loss": 1.1199,
      "num_input_tokens_seen": 1280616,
      "step": 2305
    },
    {
      "epoch": 40.530973451327434,
      "grad_norm": 5.3125,
      "learning_rate": 4.9590034712524086e-05,
      "loss": 1.0061,
      "num_input_tokens_seen": 1283128,
      "step": 2310
    },
    {
      "epoch": 40.61946902654867,
      "grad_norm": 9.0625,
      "learning_rate": 4.958826217504053e-05,
      "loss": 0.932,
      "num_input_tokens_seen": 1285704,
      "step": 2315
    },
    {
      "epoch": 40.70796460176991,
      "grad_norm": 6.78125,
      "learning_rate": 4.958648584573795e-05,
      "loss": 1.0591,
      "num_input_tokens_seen": 1288248,
      "step": 2320
    },
    {
      "epoch": 40.796460176991154,
      "grad_norm": 7.8125,
      "learning_rate": 4.958470572489028e-05,
      "loss": 1.1401,
      "num_input_tokens_seen": 1291208,
      "step": 2325
    },
    {
      "epoch": 40.88495575221239,
      "grad_norm": 7.84375,
      "learning_rate": 4.958292181277203e-05,
      "loss": 1.05,
      "num_input_tokens_seen": 1293976,
      "step": 2330
    },
    {
      "epoch": 40.97345132743363,
      "grad_norm": 9.25,
      "learning_rate": 4.958113410965832e-05,
      "loss": 1.1935,
      "num_input_tokens_seen": 1296664,
      "step": 2335
    },
    {
      "epoch": 41.05309734513274,
      "grad_norm": 5.53125,
      "learning_rate": 4.957934261582481e-05,
      "loss": 0.9401,
      "num_input_tokens_seen": 1299216,
      "step": 2340
    },
    {
      "epoch": 41.14159292035398,
      "grad_norm": 8.1875,
      "learning_rate": 4.95775473315478e-05,
      "loss": 1.0954,
      "num_input_tokens_seen": 1302544,
      "step": 2345
    },
    {
      "epoch": 41.230088495575224,
      "grad_norm": 8.0625,
      "learning_rate": 4.9575748257104124e-05,
      "loss": 0.9991,
      "num_input_tokens_seen": 1305408,
      "step": 2350
    },
    {
      "epoch": 41.31858407079646,
      "grad_norm": 8.0625,
      "learning_rate": 4.9573945392771224e-05,
      "loss": 1.228,
      "num_input_tokens_seen": 1308336,
      "step": 2355
    },
    {
      "epoch": 41.4070796460177,
      "grad_norm": 8.4375,
      "learning_rate": 4.9572138738827134e-05,
      "loss": 0.9996,
      "num_input_tokens_seen": 1311152,
      "step": 2360
    },
    {
      "epoch": 41.49557522123894,
      "grad_norm": 7.84375,
      "learning_rate": 4.957032829555046e-05,
      "loss": 1.0235,
      "num_input_tokens_seen": 1313760,
      "step": 2365
    },
    {
      "epoch": 41.584070796460175,
      "grad_norm": 8.625,
      "learning_rate": 4.956851406322039e-05,
      "loss": 1.2356,
      "num_input_tokens_seen": 1316624,
      "step": 2370
    },
    {
      "epoch": 41.67256637168141,
      "grad_norm": 7.78125,
      "learning_rate": 4.9566696042116704e-05,
      "loss": 1.0768,
      "num_input_tokens_seen": 1319520,
      "step": 2375
    },
    {
      "epoch": 41.76106194690266,
      "grad_norm": 10.875,
      "learning_rate": 4.9564874232519766e-05,
      "loss": 1.0013,
      "num_input_tokens_seen": 1322208,
      "step": 2380
    },
    {
      "epoch": 41.849557522123895,
      "grad_norm": 7.53125,
      "learning_rate": 4.9563048634710516e-05,
      "loss": 0.8422,
      "num_input_tokens_seen": 1324912,
      "step": 2385
    },
    {
      "epoch": 41.93805309734513,
      "grad_norm": 9.125,
      "learning_rate": 4.956121924897049e-05,
      "loss": 1.2606,
      "num_input_tokens_seen": 1327184,
      "step": 2390
    },
    {
      "epoch": 42.017699115044245,
      "grad_norm": 9.4375,
      "learning_rate": 4.955938607558181e-05,
      "loss": 1.0276,
      "num_input_tokens_seen": 1329568,
      "step": 2395
    },
    {
      "epoch": 42.10619469026549,
      "grad_norm": 9.3125,
      "learning_rate": 4.955754911482715e-05,
      "loss": 1.1009,
      "num_input_tokens_seen": 1332096,
      "step": 2400
    },
    {
      "epoch": 42.10619469026549,
      "eval_loss": 1.0410767793655396,
      "eval_runtime": 0.9368,
      "eval_samples_per_second": 26.685,
      "eval_steps_per_second": 13.876,
      "num_input_tokens_seen": 1332096,
      "step": 2400
    },
    {
      "epoch": 42.19469026548673,
      "grad_norm": 8.625,
      "learning_rate": 4.9555708366989804e-05,
      "loss": 1.1461,
      "num_input_tokens_seen": 1334816,
      "step": 2405
    },
    {
      "epoch": 42.283185840707965,
      "grad_norm": 9.0625,
      "learning_rate": 4.9553863832353655e-05,
      "loss": 0.9442,
      "num_input_tokens_seen": 1337504,
      "step": 2410
    },
    {
      "epoch": 42.3716814159292,
      "grad_norm": 7.90625,
      "learning_rate": 4.955201551120313e-05,
      "loss": 1.2566,
      "num_input_tokens_seen": 1340800,
      "step": 2415
    },
    {
      "epoch": 42.46017699115044,
      "grad_norm": 9.4375,
      "learning_rate": 4.955016340382328e-05,
      "loss": 0.8846,
      "num_input_tokens_seen": 1343600,
      "step": 2420
    },
    {
      "epoch": 42.54867256637168,
      "grad_norm": 9.6875,
      "learning_rate": 4.954830751049972e-05,
      "loss": 0.9403,
      "num_input_tokens_seen": 1346416,
      "step": 2425
    },
    {
      "epoch": 42.63716814159292,
      "grad_norm": 12.8125,
      "learning_rate": 4.954644783151864e-05,
      "loss": 1.0466,
      "num_input_tokens_seen": 1349328,
      "step": 2430
    },
    {
      "epoch": 42.72566371681416,
      "grad_norm": 8.875,
      "learning_rate": 4.954458436716684e-05,
      "loss": 1.1436,
      "num_input_tokens_seen": 1351760,
      "step": 2435
    },
    {
      "epoch": 42.8141592920354,
      "grad_norm": 7.125,
      "learning_rate": 4.954271711773168e-05,
      "loss": 1.0057,
      "num_input_tokens_seen": 1354240,
      "step": 2440
    },
    {
      "epoch": 42.902654867256636,
      "grad_norm": 8.6875,
      "learning_rate": 4.9540846083501115e-05,
      "loss": 1.1506,
      "num_input_tokens_seen": 1356944,
      "step": 2445
    },
    {
      "epoch": 42.991150442477874,
      "grad_norm": 7.59375,
      "learning_rate": 4.953897126476369e-05,
      "loss": 1.0824,
      "num_input_tokens_seen": 1360208,
      "step": 2450
    },
    {
      "epoch": 43.07079646017699,
      "grad_norm": 8.5,
      "learning_rate": 4.9537092661808514e-05,
      "loss": 0.8338,
      "num_input_tokens_seen": 1362560,
      "step": 2455
    },
    {
      "epoch": 43.15929203539823,
      "grad_norm": 12.0,
      "learning_rate": 4.9535210274925306e-05,
      "loss": 1.0029,
      "num_input_tokens_seen": 1365840,
      "step": 2460
    },
    {
      "epoch": 43.24778761061947,
      "grad_norm": 8.125,
      "learning_rate": 4.953332410440435e-05,
      "loss": 1.1018,
      "num_input_tokens_seen": 1368928,
      "step": 2465
    },
    {
      "epoch": 43.336283185840706,
      "grad_norm": 8.4375,
      "learning_rate": 4.9531434150536496e-05,
      "loss": 1.0809,
      "num_input_tokens_seen": 1371456,
      "step": 2470
    },
    {
      "epoch": 43.424778761061944,
      "grad_norm": 9.1875,
      "learning_rate": 4.952954041361322e-05,
      "loss": 1.0611,
      "num_input_tokens_seen": 1373936,
      "step": 2475
    },
    {
      "epoch": 43.51327433628319,
      "grad_norm": 9.8125,
      "learning_rate": 4.952764289392655e-05,
      "loss": 1.1373,
      "num_input_tokens_seen": 1377184,
      "step": 2480
    },
    {
      "epoch": 43.60176991150443,
      "grad_norm": 8.1875,
      "learning_rate": 4.952574159176912e-05,
      "loss": 1.128,
      "num_input_tokens_seen": 1379840,
      "step": 2485
    },
    {
      "epoch": 43.690265486725664,
      "grad_norm": 8.5625,
      "learning_rate": 4.952383650743413e-05,
      "loss": 1.1107,
      "num_input_tokens_seen": 1382320,
      "step": 2490
    },
    {
      "epoch": 43.7787610619469,
      "grad_norm": 8.625,
      "learning_rate": 4.952192764121536e-05,
      "loss": 1.0807,
      "num_input_tokens_seen": 1384768,
      "step": 2495
    },
    {
      "epoch": 43.86725663716814,
      "grad_norm": 9.375,
      "learning_rate": 4.9520014993407185e-05,
      "loss": 0.9164,
      "num_input_tokens_seen": 1387632,
      "step": 2500
    },
    {
      "epoch": 43.95575221238938,
      "grad_norm": 8.5625,
      "learning_rate": 4.951809856430456e-05,
      "loss": 1.0785,
      "num_input_tokens_seen": 1390544,
      "step": 2505
    },
    {
      "epoch": 44.0353982300885,
      "grad_norm": 8.5625,
      "learning_rate": 4.951617835420303e-05,
      "loss": 1.0361,
      "num_input_tokens_seen": 1392776,
      "step": 2510
    },
    {
      "epoch": 44.123893805309734,
      "grad_norm": 8.375,
      "learning_rate": 4.951425436339869e-05,
      "loss": 1.0071,
      "num_input_tokens_seen": 1395496,
      "step": 2515
    },
    {
      "epoch": 44.21238938053097,
      "grad_norm": 7.5,
      "learning_rate": 4.9512326592188274e-05,
      "loss": 1.1831,
      "num_input_tokens_seen": 1398504,
      "step": 2520
    },
    {
      "epoch": 44.30088495575221,
      "grad_norm": 11.5,
      "learning_rate": 4.9510395040869054e-05,
      "loss": 1.0001,
      "num_input_tokens_seen": 1401176,
      "step": 2525
    },
    {
      "epoch": 44.389380530973455,
      "grad_norm": 11.125,
      "learning_rate": 4.9508459709738905e-05,
      "loss": 1.1393,
      "num_input_tokens_seen": 1404312,
      "step": 2530
    },
    {
      "epoch": 44.47787610619469,
      "grad_norm": 9.3125,
      "learning_rate": 4.950652059909627e-05,
      "loss": 1.0886,
      "num_input_tokens_seen": 1406776,
      "step": 2535
    },
    {
      "epoch": 44.56637168141593,
      "grad_norm": 8.125,
      "learning_rate": 4.95045777092402e-05,
      "loss": 1.0657,
      "num_input_tokens_seen": 1409912,
      "step": 2540
    },
    {
      "epoch": 44.65486725663717,
      "grad_norm": 7.125,
      "learning_rate": 4.950263104047031e-05,
      "loss": 1.053,
      "num_input_tokens_seen": 1412632,
      "step": 2545
    },
    {
      "epoch": 44.743362831858406,
      "grad_norm": 8.3125,
      "learning_rate": 4.9500680593086775e-05,
      "loss": 0.8453,
      "num_input_tokens_seen": 1415928,
      "step": 2550
    },
    {
      "epoch": 44.83185840707964,
      "grad_norm": 8.25,
      "learning_rate": 4.94987263673904e-05,
      "loss": 1.0167,
      "num_input_tokens_seen": 1418904,
      "step": 2555
    },
    {
      "epoch": 44.92035398230089,
      "grad_norm": 9.1875,
      "learning_rate": 4.949676836368256e-05,
      "loss": 1.0468,
      "num_input_tokens_seen": 1421656,
      "step": 2560
    },
    {
      "epoch": 45.0,
      "grad_norm": 9.3125,
      "learning_rate": 4.949480658226518e-05,
      "loss": 1.2257,
      "num_input_tokens_seen": 1423840,
      "step": 2565
    },
    {
      "epoch": 45.08849557522124,
      "grad_norm": 9.375,
      "learning_rate": 4.949284102344082e-05,
      "loss": 1.066,
      "num_input_tokens_seen": 1426960,
      "step": 2570
    },
    {
      "epoch": 45.176991150442475,
      "grad_norm": 7.75,
      "learning_rate": 4.9490871687512565e-05,
      "loss": 1.1606,
      "num_input_tokens_seen": 1429648,
      "step": 2575
    },
    {
      "epoch": 45.26548672566372,
      "grad_norm": 13.4375,
      "learning_rate": 4.948889857478413e-05,
      "loss": 0.9771,
      "num_input_tokens_seen": 1432480,
      "step": 2580
    },
    {
      "epoch": 45.35398230088496,
      "grad_norm": 9.625,
      "learning_rate": 4.948692168555978e-05,
      "loss": 1.1605,
      "num_input_tokens_seen": 1435360,
      "step": 2585
    },
    {
      "epoch": 45.442477876106196,
      "grad_norm": 5.53125,
      "learning_rate": 4.94849410201444e-05,
      "loss": 1.0444,
      "num_input_tokens_seen": 1438240,
      "step": 2590
    },
    {
      "epoch": 45.530973451327434,
      "grad_norm": 11.875,
      "learning_rate": 4.948295657884341e-05,
      "loss": 1.1787,
      "num_input_tokens_seen": 1440976,
      "step": 2595
    },
    {
      "epoch": 45.61946902654867,
      "grad_norm": 8.1875,
      "learning_rate": 4.9480968361962835e-05,
      "loss": 1.0705,
      "num_input_tokens_seen": 1443792,
      "step": 2600
    },
    {
      "epoch": 45.61946902654867,
      "eval_loss": 1.0365309715270996,
      "eval_runtime": 0.9361,
      "eval_samples_per_second": 26.708,
      "eval_steps_per_second": 13.888,
      "num_input_tokens_seen": 1443792,
      "step": 2600
    },
    {
      "epoch": 45.70796460176991,
      "grad_norm": 10.3125,
      "learning_rate": 4.9478976369809305e-05,
      "loss": 1.0247,
      "num_input_tokens_seen": 1446128,
      "step": 2605
    },
    {
      "epoch": 45.796460176991154,
      "grad_norm": 6.78125,
      "learning_rate": 4.947698060268999e-05,
      "loss": 0.8802,
      "num_input_tokens_seen": 1449216,
      "step": 2610
    },
    {
      "epoch": 45.88495575221239,
      "grad_norm": 10.9375,
      "learning_rate": 4.9474981060912665e-05,
      "loss": 1.1404,
      "num_input_tokens_seen": 1452368,
      "step": 2615
    },
    {
      "epoch": 45.97345132743363,
      "grad_norm": 10.375,
      "learning_rate": 4.94729777447857e-05,
      "loss": 1.1044,
      "num_input_tokens_seen": 1455008,
      "step": 2620
    },
    {
      "epoch": 46.05309734513274,
      "grad_norm": 6.6875,
      "learning_rate": 4.947097065461801e-05,
      "loss": 1.1003,
      "num_input_tokens_seen": 1457200,
      "step": 2625
    },
    {
      "epoch": 46.14159292035398,
      "grad_norm": 7.8125,
      "learning_rate": 4.9468959790719125e-05,
      "loss": 0.6578,
      "num_input_tokens_seen": 1460192,
      "step": 2630
    },
    {
      "epoch": 46.230088495575224,
      "grad_norm": 16.75,
      "learning_rate": 4.9466945153399146e-05,
      "loss": 1.3081,
      "num_input_tokens_seen": 1462960,
      "step": 2635
    },
    {
      "epoch": 46.31858407079646,
      "grad_norm": 8.25,
      "learning_rate": 4.9464926742968755e-05,
      "loss": 1.0785,
      "num_input_tokens_seen": 1466032,
      "step": 2640
    },
    {
      "epoch": 46.4070796460177,
      "grad_norm": 10.0625,
      "learning_rate": 4.946290455973921e-05,
      "loss": 1.2016,
      "num_input_tokens_seen": 1468704,
      "step": 2645
    },
    {
      "epoch": 46.49557522123894,
      "grad_norm": 9.25,
      "learning_rate": 4.9460878604022365e-05,
      "loss": 1.0973,
      "num_input_tokens_seen": 1471616,
      "step": 2650
    },
    {
      "epoch": 46.584070796460175,
      "grad_norm": 7.28125,
      "learning_rate": 4.945884887613065e-05,
      "loss": 1.0145,
      "num_input_tokens_seen": 1474224,
      "step": 2655
    },
    {
      "epoch": 46.67256637168141,
      "grad_norm": 12.625,
      "learning_rate": 4.9456815376377055e-05,
      "loss": 1.1635,
      "num_input_tokens_seen": 1477264,
      "step": 2660
    },
    {
      "epoch": 46.76106194690266,
      "grad_norm": 8.125,
      "learning_rate": 4.9454778105075195e-05,
      "loss": 0.9729,
      "num_input_tokens_seen": 1479600,
      "step": 2665
    },
    {
      "epoch": 46.849557522123895,
      "grad_norm": 8.375,
      "learning_rate": 4.945273706253924e-05,
      "loss": 0.8822,
      "num_input_tokens_seen": 1482512,
      "step": 2670
    },
    {
      "epoch": 46.93805309734513,
      "grad_norm": 9.125,
      "learning_rate": 4.9450692249083925e-05,
      "loss": 1.0949,
      "num_input_tokens_seen": 1485184,
      "step": 2675
    },
    {
      "epoch": 47.017699115044245,
      "grad_norm": 10.0,
      "learning_rate": 4.9448643665024605e-05,
      "loss": 1.0174,
      "num_input_tokens_seen": 1487456,
      "step": 2680
    },
    {
      "epoch": 47.10619469026549,
      "grad_norm": 7.6875,
      "learning_rate": 4.944659131067719e-05,
      "loss": 1.1195,
      "num_input_tokens_seen": 1490592,
      "step": 2685
    },
    {
      "epoch": 47.19469026548673,
      "grad_norm": 6.40625,
      "learning_rate": 4.944453518635818e-05,
      "loss": 0.9324,
      "num_input_tokens_seen": 1493520,
      "step": 2690
    },
    {
      "epoch": 47.283185840707965,
      "grad_norm": 7.5625,
      "learning_rate": 4.944247529238465e-05,
      "loss": 1.2133,
      "num_input_tokens_seen": 1496336,
      "step": 2695
    },
    {
      "epoch": 47.3716814159292,
      "grad_norm": 7.84375,
      "learning_rate": 4.944041162907427e-05,
      "loss": 1.1062,
      "num_input_tokens_seen": 1499040,
      "step": 2700
    },
    {
      "epoch": 47.46017699115044,
      "grad_norm": 7.34375,
      "learning_rate": 4.943834419674529e-05,
      "loss": 1.1706,
      "num_input_tokens_seen": 1501888,
      "step": 2705
    },
    {
      "epoch": 47.54867256637168,
      "grad_norm": 10.875,
      "learning_rate": 4.9436272995716506e-05,
      "loss": 1.0725,
      "num_input_tokens_seen": 1504512,
      "step": 2710
    },
    {
      "epoch": 47.63716814159292,
      "grad_norm": 7.4375,
      "learning_rate": 4.943419802630735e-05,
      "loss": 0.9417,
      "num_input_tokens_seen": 1507568,
      "step": 2715
    },
    {
      "epoch": 47.72566371681416,
      "grad_norm": 6.5,
      "learning_rate": 4.94321192888378e-05,
      "loss": 1.0919,
      "num_input_tokens_seen": 1510192,
      "step": 2720
    },
    {
      "epoch": 47.8141592920354,
      "grad_norm": 7.75,
      "learning_rate": 4.943003678362842e-05,
      "loss": 1.0099,
      "num_input_tokens_seen": 1512464,
      "step": 2725
    },
    {
      "epoch": 47.902654867256636,
      "grad_norm": 6.90625,
      "learning_rate": 4.942795051100036e-05,
      "loss": 1.0711,
      "num_input_tokens_seen": 1514816,
      "step": 2730
    },
    {
      "epoch": 47.991150442477874,
      "grad_norm": 8.5625,
      "learning_rate": 4.942586047127536e-05,
      "loss": 0.9587,
      "num_input_tokens_seen": 1517840,
      "step": 2735
    },
    {
      "epoch": 48.07079646017699,
      "grad_norm": 6.625,
      "learning_rate": 4.942376666477571e-05,
      "loss": 1.0307,
      "num_input_tokens_seen": 1520312,
      "step": 2740
    },
    {
      "epoch": 48.15929203539823,
      "grad_norm": 8.5625,
      "learning_rate": 4.9421669091824304e-05,
      "loss": 1.084,
      "num_input_tokens_seen": 1523352,
      "step": 2745
    },
    {
      "epoch": 48.24778761061947,
      "grad_norm": 8.5625,
      "learning_rate": 4.9419567752744634e-05,
      "loss": 0.9973,
      "num_input_tokens_seen": 1525992,
      "step": 2750
    },
    {
      "epoch": 48.336283185840706,
      "grad_norm": 12.625,
      "learning_rate": 4.941746264786074e-05,
      "loss": 0.9243,
      "num_input_tokens_seen": 1528632,
      "step": 2755
    },
    {
      "epoch": 48.424778761061944,
      "grad_norm": 6.96875,
      "learning_rate": 4.9415353777497254e-05,
      "loss": 1.1191,
      "num_input_tokens_seen": 1531416,
      "step": 2760
    },
    {
      "epoch": 48.51327433628319,
      "grad_norm": 7.5625,
      "learning_rate": 4.9413241141979394e-05,
      "loss": 1.216,
      "num_input_tokens_seen": 1534296,
      "step": 2765
    },
    {
      "epoch": 48.60176991150443,
      "grad_norm": 8.0625,
      "learning_rate": 4.9411124741632956e-05,
      "loss": 1.1385,
      "num_input_tokens_seen": 1537544,
      "step": 2770
    },
    {
      "epoch": 48.690265486725664,
      "grad_norm": 12.9375,
      "learning_rate": 4.940900457678431e-05,
      "loss": 0.9148,
      "num_input_tokens_seen": 1539944,
      "step": 2775
    },
    {
      "epoch": 48.7787610619469,
      "grad_norm": 9.4375,
      "learning_rate": 4.9406880647760425e-05,
      "loss": 1.0459,
      "num_input_tokens_seen": 1542488,
      "step": 2780
    },
    {
      "epoch": 48.86725663716814,
      "grad_norm": 8.375,
      "learning_rate": 4.9404752954888824e-05,
      "loss": 0.9952,
      "num_input_tokens_seen": 1545464,
      "step": 2785
    },
    {
      "epoch": 48.95575221238938,
      "grad_norm": 8.25,
      "learning_rate": 4.940262149849762e-05,
      "loss": 1.237,
      "num_input_tokens_seen": 1548152,
      "step": 2790
    },
    {
      "epoch": 49.0353982300885,
      "grad_norm": 8.125,
      "learning_rate": 4.9400486278915526e-05,
      "loss": 1.1286,
      "num_input_tokens_seen": 1550720,
      "step": 2795
    },
    {
      "epoch": 49.123893805309734,
      "grad_norm": 9.375,
      "learning_rate": 4.939834729647181e-05,
      "loss": 1.0641,
      "num_input_tokens_seen": 1553600,
      "step": 2800
    },
    {
      "epoch": 49.123893805309734,
      "eval_loss": 1.0337246656417847,
      "eval_runtime": 0.937,
      "eval_samples_per_second": 26.68,
      "eval_steps_per_second": 13.873,
      "num_input_tokens_seen": 1553600,
      "step": 2800
    },
    {
      "epoch": 49.21238938053097,
      "grad_norm": 8.625,
      "learning_rate": 4.9396204551496326e-05,
      "loss": 1.1942,
      "num_input_tokens_seen": 1555936,
      "step": 2805
    },
    {
      "epoch": 49.30088495575221,
      "grad_norm": 6.5,
      "learning_rate": 4.939405804431952e-05,
      "loss": 0.9292,
      "num_input_tokens_seen": 1559264,
      "step": 2810
    },
    {
      "epoch": 49.389380530973455,
      "grad_norm": 7.28125,
      "learning_rate": 4.9391907775272414e-05,
      "loss": 0.9311,
      "num_input_tokens_seen": 1562096,
      "step": 2815
    },
    {
      "epoch": 49.47787610619469,
      "grad_norm": 8.25,
      "learning_rate": 4.9389753744686604e-05,
      "loss": 1.2517,
      "num_input_tokens_seen": 1564672,
      "step": 2820
    },
    {
      "epoch": 49.56637168141593,
      "grad_norm": 7.90625,
      "learning_rate": 4.938759595289426e-05,
      "loss": 1.0261,
      "num_input_tokens_seen": 1567664,
      "step": 2825
    },
    {
      "epoch": 49.65486725663717,
      "grad_norm": 7.9375,
      "learning_rate": 4.938543440022815e-05,
      "loss": 0.9091,
      "num_input_tokens_seen": 1570480,
      "step": 2830
    },
    {
      "epoch": 49.743362831858406,
      "grad_norm": 7.53125,
      "learning_rate": 4.938326908702161e-05,
      "loss": 1.0903,
      "num_input_tokens_seen": 1573120,
      "step": 2835
    },
    {
      "epoch": 49.83185840707964,
      "grad_norm": 7.25,
      "learning_rate": 4.9381100013608554e-05,
      "loss": 1.0519,
      "num_input_tokens_seen": 1576016,
      "step": 2840
    },
    {
      "epoch": 49.92035398230089,
      "grad_norm": 8.0625,
      "learning_rate": 4.9378927180323485e-05,
      "loss": 1.0855,
      "num_input_tokens_seen": 1578272,
      "step": 2845
    },
    {
      "epoch": 50.0,
      "grad_norm": 12.0,
      "learning_rate": 4.937675058750148e-05,
      "loss": 1.0891,
      "num_input_tokens_seen": 1580800,
      "step": 2850
    },
    {
      "epoch": 50.08849557522124,
      "grad_norm": 9.625,
      "learning_rate": 4.937457023547819e-05,
      "loss": 1.2194,
      "num_input_tokens_seen": 1583792,
      "step": 2855
    },
    {
      "epoch": 50.176991150442475,
      "grad_norm": 9.6875,
      "learning_rate": 4.9372386124589876e-05,
      "loss": 0.8434,
      "num_input_tokens_seen": 1586784,
      "step": 2860
    },
    {
      "epoch": 50.26548672566372,
      "grad_norm": 11.125,
      "learning_rate": 4.937019825517333e-05,
      "loss": 1.0576,
      "num_input_tokens_seen": 1589360,
      "step": 2865
    },
    {
      "epoch": 50.35398230088496,
      "grad_norm": 6.96875,
      "learning_rate": 4.9368006627565954e-05,
      "loss": 1.1608,
      "num_input_tokens_seen": 1591680,
      "step": 2870
    },
    {
      "epoch": 50.442477876106196,
      "grad_norm": 11.75,
      "learning_rate": 4.936581124210573e-05,
      "loss": 0.9917,
      "num_input_tokens_seen": 1594064,
      "step": 2875
    },
    {
      "epoch": 50.530973451327434,
      "grad_norm": 8.3125,
      "learning_rate": 4.9363612099131216e-05,
      "loss": 1.2948,
      "num_input_tokens_seen": 1596624,
      "step": 2880
    },
    {
      "epoch": 50.61946902654867,
      "grad_norm": 8.1875,
      "learning_rate": 4.936140919898155e-05,
      "loss": 1.0662,
      "num_input_tokens_seen": 1599664,
      "step": 2885
    },
    {
      "epoch": 50.70796460176991,
      "grad_norm": 8.25,
      "learning_rate": 4.9359202541996426e-05,
      "loss": 0.9519,
      "num_input_tokens_seen": 1602720,
      "step": 2890
    },
    {
      "epoch": 50.796460176991154,
      "grad_norm": 6.65625,
      "learning_rate": 4.935699212851616e-05,
      "loss": 0.9199,
      "num_input_tokens_seen": 1605664,
      "step": 2895
    },
    {
      "epoch": 50.88495575221239,
      "grad_norm": 9.5,
      "learning_rate": 4.935477795888162e-05,
      "loss": 1.1664,
      "num_input_tokens_seen": 1608928,
      "step": 2900
    },
    {
      "epoch": 50.97345132743363,
      "grad_norm": 8.0625,
      "learning_rate": 4.935256003343426e-05,
      "loss": 1.0509,
      "num_input_tokens_seen": 1611744,
      "step": 2905
    },
    {
      "epoch": 51.05309734513274,
      "grad_norm": 8.125,
      "learning_rate": 4.93503383525161e-05,
      "loss": 1.1642,
      "num_input_tokens_seen": 1614048,
      "step": 2910
    },
    {
      "epoch": 51.14159292035398,
      "grad_norm": 11.25,
      "learning_rate": 4.934811291646977e-05,
      "loss": 0.8886,
      "num_input_tokens_seen": 1617120,
      "step": 2915
    },
    {
      "epoch": 51.230088495575224,
      "grad_norm": 9.4375,
      "learning_rate": 4.934588372563845e-05,
      "loss": 1.1167,
      "num_input_tokens_seen": 1619840,
      "step": 2920
    },
    {
      "epoch": 51.31858407079646,
      "grad_norm": 8.0,
      "learning_rate": 4.93436507803659e-05,
      "loss": 1.0726,
      "num_input_tokens_seen": 1622864,
      "step": 2925
    },
    {
      "epoch": 51.4070796460177,
      "grad_norm": 11.0625,
      "learning_rate": 4.934141408099649e-05,
      "loss": 1.0717,
      "num_input_tokens_seen": 1625888,
      "step": 2930
    },
    {
      "epoch": 51.49557522123894,
      "grad_norm": 6.875,
      "learning_rate": 4.9339173627875135e-05,
      "loss": 0.9769,
      "num_input_tokens_seen": 1628736,
      "step": 2935
    },
    {
      "epoch": 51.584070796460175,
      "grad_norm": 8.5,
      "learning_rate": 4.9336929421347335e-05,
      "loss": 1.0558,
      "num_input_tokens_seen": 1631328,
      "step": 2940
    },
    {
      "epoch": 51.67256637168141,
      "grad_norm": 7.8125,
      "learning_rate": 4.933468146175918e-05,
      "loss": 1.0486,
      "num_input_tokens_seen": 1634432,
      "step": 2945
    },
    {
      "epoch": 51.76106194690266,
      "grad_norm": 9.125,
      "learning_rate": 4.933242974945734e-05,
      "loss": 1.1221,
      "num_input_tokens_seen": 1637232,
      "step": 2950
    },
    {
      "epoch": 51.849557522123895,
      "grad_norm": 9.0,
      "learning_rate": 4.933017428478906e-05,
      "loss": 1.0418,
      "num_input_tokens_seen": 1639920,
      "step": 2955
    },
    {
      "epoch": 51.93805309734513,
      "grad_norm": 7.46875,
      "learning_rate": 4.932791506810214e-05,
      "loss": 1.1377,
      "num_input_tokens_seen": 1642368,
      "step": 2960
    },
    {
      "epoch": 52.017699115044245,
      "grad_norm": 9.5,
      "learning_rate": 4.932565209974499e-05,
      "loss": 0.9259,
      "num_input_tokens_seen": 1645144,
      "step": 2965
    },
    {
      "epoch": 52.10619469026549,
      "grad_norm": 6.9375,
      "learning_rate": 4.93233853800666e-05,
      "loss": 1.1173,
      "num_input_tokens_seen": 1647544,
      "step": 2970
    },
    {
      "epoch": 52.19469026548673,
      "grad_norm": 8.875,
      "learning_rate": 4.932111490941651e-05,
      "loss": 0.9252,
      "num_input_tokens_seen": 1651080,
      "step": 2975
    },
    {
      "epoch": 52.283185840707965,
      "grad_norm": 9.125,
      "learning_rate": 4.9318840688144876e-05,
      "loss": 1.1238,
      "num_input_tokens_seen": 1653944,
      "step": 2980
    },
    {
      "epoch": 52.3716814159292,
      "grad_norm": 8.0625,
      "learning_rate": 4.9316562716602387e-05,
      "loss": 1.1846,
      "num_input_tokens_seen": 1656392,
      "step": 2985
    },
    {
      "epoch": 52.46017699115044,
      "grad_norm": 7.1875,
      "learning_rate": 4.9314280995140346e-05,
      "loss": 1.1333,
      "num_input_tokens_seen": 1658744,
      "step": 2990
    },
    {
      "epoch": 52.54867256637168,
      "grad_norm": 9.0,
      "learning_rate": 4.931199552411063e-05,
      "loss": 1.1994,
      "num_input_tokens_seen": 1661224,
      "step": 2995
    },
    {
      "epoch": 52.63716814159292,
      "grad_norm": 7.09375,
      "learning_rate": 4.930970630386568e-05,
      "loss": 1.0718,
      "num_input_tokens_seen": 1664296,
      "step": 3000
    },
    {
      "epoch": 52.63716814159292,
      "eval_loss": 1.0373647212982178,
      "eval_runtime": 0.933,
      "eval_samples_per_second": 26.796,
      "eval_steps_per_second": 13.934,
      "num_input_tokens_seen": 1664296,
      "step": 3000
    },
    {
      "epoch": 52.72566371681416,
      "grad_norm": 8.0,
      "learning_rate": 4.9307413334758524e-05,
      "loss": 0.9431,
      "num_input_tokens_seen": 1667080,
      "step": 3005
    },
    {
      "epoch": 52.8141592920354,
      "grad_norm": 8.25,
      "learning_rate": 4.930511661714276e-05,
      "loss": 0.9939,
      "num_input_tokens_seen": 1669976,
      "step": 3010
    },
    {
      "epoch": 52.902654867256636,
      "grad_norm": 8.75,
      "learning_rate": 4.9302816151372576e-05,
      "loss": 1.0408,
      "num_input_tokens_seen": 1672536,
      "step": 3015
    },
    {
      "epoch": 52.991150442477874,
      "grad_norm": 8.0,
      "learning_rate": 4.930051193780274e-05,
      "loss": 0.9894,
      "num_input_tokens_seen": 1675704,
      "step": 3020
    },
    {
      "epoch": 53.07079646017699,
      "grad_norm": 10.875,
      "learning_rate": 4.929820397678858e-05,
      "loss": 1.1276,
      "num_input_tokens_seen": 1678288,
      "step": 3025
    },
    {
      "epoch": 53.15929203539823,
      "grad_norm": 9.75,
      "learning_rate": 4.9295892268686015e-05,
      "loss": 0.9697,
      "num_input_tokens_seen": 1681232,
      "step": 3030
    },
    {
      "epoch": 53.24778761061947,
      "grad_norm": 9.875,
      "learning_rate": 4.9293576813851536e-05,
      "loss": 1.0657,
      "num_input_tokens_seen": 1684160,
      "step": 3035
    },
    {
      "epoch": 53.336283185840706,
      "grad_norm": 8.8125,
      "learning_rate": 4.929125761264223e-05,
      "loss": 0.9177,
      "num_input_tokens_seen": 1687008,
      "step": 3040
    },
    {
      "epoch": 53.424778761061944,
      "grad_norm": 7.75,
      "learning_rate": 4.928893466541573e-05,
      "loss": 1.0389,
      "num_input_tokens_seen": 1689776,
      "step": 3045
    },
    {
      "epoch": 53.51327433628319,
      "grad_norm": 7.53125,
      "learning_rate": 4.928660797253027e-05,
      "loss": 1.0129,
      "num_input_tokens_seen": 1692896,
      "step": 3050
    },
    {
      "epoch": 53.60176991150443,
      "grad_norm": 8.875,
      "learning_rate": 4.928427753434467e-05,
      "loss": 1.2072,
      "num_input_tokens_seen": 1695696,
      "step": 3055
    },
    {
      "epoch": 53.690265486725664,
      "grad_norm": 8.4375,
      "learning_rate": 4.9281943351218286e-05,
      "loss": 0.9551,
      "num_input_tokens_seen": 1698544,
      "step": 3060
    },
    {
      "epoch": 53.7787610619469,
      "grad_norm": 8.5,
      "learning_rate": 4.9279605423511095e-05,
      "loss": 1.0342,
      "num_input_tokens_seen": 1701344,
      "step": 3065
    },
    {
      "epoch": 53.86725663716814,
      "grad_norm": 6.0625,
      "learning_rate": 4.927726375158363e-05,
      "loss": 1.0681,
      "num_input_tokens_seen": 1703712,
      "step": 3070
    },
    {
      "epoch": 53.95575221238938,
      "grad_norm": 8.125,
      "learning_rate": 4.9274918335797004e-05,
      "loss": 0.9698,
      "num_input_tokens_seen": 1706400,
      "step": 3075
    },
    {
      "epoch": 54.0353982300885,
      "grad_norm": 7.90625,
      "learning_rate": 4.927256917651292e-05,
      "loss": 1.0973,
      "num_input_tokens_seen": 1708720,
      "step": 3080
    },
    {
      "epoch": 54.123893805309734,
      "grad_norm": 8.8125,
      "learning_rate": 4.927021627409364e-05,
      "loss": 1.017,
      "num_input_tokens_seen": 1711680,
      "step": 3085
    },
    {
      "epoch": 54.21238938053097,
      "grad_norm": 7.0,
      "learning_rate": 4.9267859628902005e-05,
      "loss": 1.0901,
      "num_input_tokens_seen": 1714240,
      "step": 3090
    },
    {
      "epoch": 54.30088495575221,
      "grad_norm": 6.6875,
      "learning_rate": 4.9265499241301454e-05,
      "loss": 0.9609,
      "num_input_tokens_seen": 1716848,
      "step": 3095
    },
    {
      "epoch": 54.389380530973455,
      "grad_norm": 10.3125,
      "learning_rate": 4.926313511165598e-05,
      "loss": 1.0088,
      "num_input_tokens_seen": 1719856,
      "step": 3100
    },
    {
      "epoch": 54.47787610619469,
      "grad_norm": 7.9375,
      "learning_rate": 4.926076724033016e-05,
      "loss": 0.901,
      "num_input_tokens_seen": 1723120,
      "step": 3105
    },
    {
      "epoch": 54.56637168141593,
      "grad_norm": 9.6875,
      "learning_rate": 4.9258395627689146e-05,
      "loss": 0.9077,
      "num_input_tokens_seen": 1725552,
      "step": 3110
    },
    {
      "epoch": 54.65486725663717,
      "grad_norm": 14.25,
      "learning_rate": 4.925602027409868e-05,
      "loss": 1.2118,
      "num_input_tokens_seen": 1727712,
      "step": 3115
    },
    {
      "epoch": 54.743362831858406,
      "grad_norm": 8.3125,
      "learning_rate": 4.925364117992507e-05,
      "loss": 1.0978,
      "num_input_tokens_seen": 1731040,
      "step": 3120
    },
    {
      "epoch": 54.83185840707964,
      "grad_norm": 8.8125,
      "learning_rate": 4.92512583455352e-05,
      "loss": 1.1139,
      "num_input_tokens_seen": 1733744,
      "step": 3125
    },
    {
      "epoch": 54.92035398230089,
      "grad_norm": 11.5625,
      "learning_rate": 4.9248871771296536e-05,
      "loss": 1.2298,
      "num_input_tokens_seen": 1736640,
      "step": 3130
    },
    {
      "epoch": 55.0,
      "grad_norm": 8.625,
      "learning_rate": 4.924648145757711e-05,
      "loss": 1.0574,
      "num_input_tokens_seen": 1739376,
      "step": 3135
    },
    {
      "epoch": 55.08849557522124,
      "grad_norm": 8.875,
      "learning_rate": 4.924408740474554e-05,
      "loss": 1.0567,
      "num_input_tokens_seen": 1741904,
      "step": 3140
    },
    {
      "epoch": 55.176991150442475,
      "grad_norm": 8.8125,
      "learning_rate": 4.924168961317103e-05,
      "loss": 1.1512,
      "num_input_tokens_seen": 1744832,
      "step": 3145
    },
    {
      "epoch": 55.26548672566372,
      "grad_norm": 7.9375,
      "learning_rate": 4.9239288083223334e-05,
      "loss": 1.0069,
      "num_input_tokens_seen": 1747984,
      "step": 3150
    },
    {
      "epoch": 55.35398230088496,
      "grad_norm": 7.3125,
      "learning_rate": 4.9236882815272803e-05,
      "loss": 1.1237,
      "num_input_tokens_seen": 1750304,
      "step": 3155
    },
    {
      "epoch": 55.442477876106196,
      "grad_norm": 8.25,
      "learning_rate": 4.9234473809690365e-05,
      "loss": 0.9591,
      "num_input_tokens_seen": 1753088,
      "step": 3160
    },
    {
      "epoch": 55.530973451327434,
      "grad_norm": 8.25,
      "learning_rate": 4.923206106684752e-05,
      "loss": 1.1108,
      "num_input_tokens_seen": 1756048,
      "step": 3165
    },
    {
      "epoch": 55.61946902654867,
      "grad_norm": 8.4375,
      "learning_rate": 4.922964458711634e-05,
      "loss": 0.9815,
      "num_input_tokens_seen": 1758960,
      "step": 3170
    },
    {
      "epoch": 55.70796460176991,
      "grad_norm": 9.5,
      "learning_rate": 4.9227224370869474e-05,
      "loss": 1.079,
      "num_input_tokens_seen": 1761600,
      "step": 3175
    },
    {
      "epoch": 55.796460176991154,
      "grad_norm": 6.0625,
      "learning_rate": 4.9224800418480155e-05,
      "loss": 1.0799,
      "num_input_tokens_seen": 1764432,
      "step": 3180
    },
    {
      "epoch": 55.88495575221239,
      "grad_norm": 8.9375,
      "learning_rate": 4.9222372730322176e-05,
      "loss": 0.9036,
      "num_input_tokens_seen": 1767120,
      "step": 3185
    },
    {
      "epoch": 55.97345132743363,
      "grad_norm": 7.46875,
      "learning_rate": 4.921994130676993e-05,
      "loss": 1.1217,
      "num_input_tokens_seen": 1769920,
      "step": 3190
    },
    {
      "epoch": 56.05309734513274,
      "grad_norm": 8.5625,
      "learning_rate": 4.9217506148198366e-05,
      "loss": 1.2944,
      "num_input_tokens_seen": 1772224,
      "step": 3195
    },
    {
      "epoch": 56.14159292035398,
      "grad_norm": 7.3125,
      "learning_rate": 4.921506725498302e-05,
      "loss": 1.0614,
      "num_input_tokens_seen": 1775264,
      "step": 3200
    },
    {
      "epoch": 56.14159292035398,
      "eval_loss": 1.026594877243042,
      "eval_runtime": 0.9358,
      "eval_samples_per_second": 26.716,
      "eval_steps_per_second": 13.892,
      "num_input_tokens_seen": 1775264,
      "step": 3200
    },
    {
      "epoch": 56.230088495575224,
      "grad_norm": 7.875,
      "learning_rate": 4.9212624627499994e-05,
      "loss": 1.0091,
      "num_input_tokens_seen": 1777616,
      "step": 3205
    },
    {
      "epoch": 56.31858407079646,
      "grad_norm": 11.375,
      "learning_rate": 4.921017826612597e-05,
      "loss": 0.9704,
      "num_input_tokens_seen": 1780592,
      "step": 3210
    },
    {
      "epoch": 56.4070796460177,
      "grad_norm": 12.375,
      "learning_rate": 4.9207728171238223e-05,
      "loss": 1.0009,
      "num_input_tokens_seen": 1783824,
      "step": 3215
    },
    {
      "epoch": 56.49557522123894,
      "grad_norm": 7.0625,
      "learning_rate": 4.920527434321458e-05,
      "loss": 1.1396,
      "num_input_tokens_seen": 1786848,
      "step": 3220
    },
    {
      "epoch": 56.584070796460175,
      "grad_norm": 7.34375,
      "learning_rate": 4.920281678243345e-05,
      "loss": 0.9987,
      "num_input_tokens_seen": 1789184,
      "step": 3225
    },
    {
      "epoch": 56.67256637168141,
      "grad_norm": 6.5,
      "learning_rate": 4.920035548927381e-05,
      "loss": 1.1379,
      "num_input_tokens_seen": 1792048,
      "step": 3230
    },
    {
      "epoch": 56.76106194690266,
      "grad_norm": 8.0625,
      "learning_rate": 4.919789046411525e-05,
      "loss": 1.0619,
      "num_input_tokens_seen": 1794816,
      "step": 3235
    },
    {
      "epoch": 56.849557522123895,
      "grad_norm": 7.90625,
      "learning_rate": 4.919542170733787e-05,
      "loss": 1.1517,
      "num_input_tokens_seen": 1797456,
      "step": 3240
    },
    {
      "epoch": 56.93805309734513,
      "grad_norm": 7.78125,
      "learning_rate": 4.919294921932242e-05,
      "loss": 1.0474,
      "num_input_tokens_seen": 1800624,
      "step": 3245
    },
    {
      "epoch": 57.017699115044245,
      "grad_norm": 8.5625,
      "learning_rate": 4.919047300045016e-05,
      "loss": 0.8836,
      "num_input_tokens_seen": 1803088,
      "step": 3250
    },
    {
      "epoch": 57.10619469026549,
      "grad_norm": 8.6875,
      "learning_rate": 4.918799305110299e-05,
      "loss": 1.1675,
      "num_input_tokens_seen": 1805680,
      "step": 3255
    },
    {
      "epoch": 57.19469026548673,
      "grad_norm": 8.4375,
      "learning_rate": 4.918550937166331e-05,
      "loss": 1.0039,
      "num_input_tokens_seen": 1808352,
      "step": 3260
    },
    {
      "epoch": 57.283185840707965,
      "grad_norm": 8.625,
      "learning_rate": 4.918302196251415e-05,
      "loss": 1.0393,
      "num_input_tokens_seen": 1811584,
      "step": 3265
    },
    {
      "epoch": 57.3716814159292,
      "grad_norm": 9.875,
      "learning_rate": 4.91805308240391e-05,
      "loss": 1.0434,
      "num_input_tokens_seen": 1814080,
      "step": 3270
    },
    {
      "epoch": 57.46017699115044,
      "grad_norm": 9.25,
      "learning_rate": 4.9178035956622326e-05,
      "loss": 0.9162,
      "num_input_tokens_seen": 1816576,
      "step": 3275
    },
    {
      "epoch": 57.54867256637168,
      "grad_norm": 8.375,
      "learning_rate": 4.917553736064857e-05,
      "loss": 1.1836,
      "num_input_tokens_seen": 1819152,
      "step": 3280
    },
    {
      "epoch": 57.63716814159292,
      "grad_norm": 7.6875,
      "learning_rate": 4.917303503650314e-05,
      "loss": 1.0258,
      "num_input_tokens_seen": 1822112,
      "step": 3285
    },
    {
      "epoch": 57.72566371681416,
      "grad_norm": 10.1875,
      "learning_rate": 4.917052898457194e-05,
      "loss": 0.9212,
      "num_input_tokens_seen": 1824864,
      "step": 3290
    },
    {
      "epoch": 57.8141592920354,
      "grad_norm": 8.9375,
      "learning_rate": 4.916801920524141e-05,
      "loss": 1.0736,
      "num_input_tokens_seen": 1827376,
      "step": 3295
    },
    {
      "epoch": 57.902654867256636,
      "grad_norm": 10.875,
      "learning_rate": 4.916550569889862e-05,
      "loss": 1.2072,
      "num_input_tokens_seen": 1830400,
      "step": 3300
    },
    {
      "epoch": 57.991150442477874,
      "grad_norm": 8.5,
      "learning_rate": 4.916298846593116e-05,
      "loss": 1.0081,
      "num_input_tokens_seen": 1833872,
      "step": 3305
    },
    {
      "epoch": 58.07079646017699,
      "grad_norm": 7.15625,
      "learning_rate": 4.916046750672722e-05,
      "loss": 1.0211,
      "num_input_tokens_seen": 1836176,
      "step": 3310
    },
    {
      "epoch": 58.15929203539823,
      "grad_norm": 7.625,
      "learning_rate": 4.915794282167559e-05,
      "loss": 1.1414,
      "num_input_tokens_seen": 1838544,
      "step": 3315
    },
    {
      "epoch": 58.24778761061947,
      "grad_norm": 8.3125,
      "learning_rate": 4.915541441116558e-05,
      "loss": 1.1066,
      "num_input_tokens_seen": 1841648,
      "step": 3320
    },
    {
      "epoch": 58.336283185840706,
      "grad_norm": 11.9375,
      "learning_rate": 4.915288227558711e-05,
      "loss": 0.9964,
      "num_input_tokens_seen": 1845088,
      "step": 3325
    },
    {
      "epoch": 58.424778761061944,
      "grad_norm": 9.5625,
      "learning_rate": 4.915034641533066e-05,
      "loss": 1.1723,
      "num_input_tokens_seen": 1847696,
      "step": 3330
    },
    {
      "epoch": 58.51327433628319,
      "grad_norm": 10.875,
      "learning_rate": 4.914780683078731e-05,
      "loss": 1.0077,
      "num_input_tokens_seen": 1850640,
      "step": 3335
    },
    {
      "epoch": 58.60176991150443,
      "grad_norm": 8.5625,
      "learning_rate": 4.9145263522348695e-05,
      "loss": 0.9295,
      "num_input_tokens_seen": 1853392,
      "step": 3340
    },
    {
      "epoch": 58.690265486725664,
      "grad_norm": 9.0625,
      "learning_rate": 4.9142716490407e-05,
      "loss": 1.2047,
      "num_input_tokens_seen": 1856032,
      "step": 3345
    },
    {
      "epoch": 58.7787610619469,
      "grad_norm": 10.0,
      "learning_rate": 4.914016573535504e-05,
      "loss": 1.2962,
      "num_input_tokens_seen": 1858640,
      "step": 3350
    },
    {
      "epoch": 58.86725663716814,
      "grad_norm": 7.4375,
      "learning_rate": 4.9137611257586154e-05,
      "loss": 0.969,
      "num_input_tokens_seen": 1861248,
      "step": 3355
    },
    {
      "epoch": 58.95575221238938,
      "grad_norm": 15.375,
      "learning_rate": 4.9135053057494274e-05,
      "loss": 0.9193,
      "num_input_tokens_seen": 1864064,
      "step": 3360
    },
    {
      "epoch": 59.0353982300885,
      "grad_norm": 10.1875,
      "learning_rate": 4.913249113547392e-05,
      "loss": 0.7664,
      "num_input_tokens_seen": 1866416,
      "step": 3365
    },
    {
      "epoch": 59.123893805309734,
      "grad_norm": 8.125,
      "learning_rate": 4.912992549192016e-05,
      "loss": 1.0872,
      "num_input_tokens_seen": 1868704,
      "step": 3370
    },
    {
      "epoch": 59.21238938053097,
      "grad_norm": 8.125,
      "learning_rate": 4.9127356127228665e-05,
      "loss": 1.0612,
      "num_input_tokens_seen": 1871312,
      "step": 3375
    },
    {
      "epoch": 59.30088495575221,
      "grad_norm": 7.28125,
      "learning_rate": 4.912478304179564e-05,
      "loss": 1.0459,
      "num_input_tokens_seen": 1874160,
      "step": 3380
    },
    {
      "epoch": 59.389380530973455,
      "grad_norm": 8.375,
      "learning_rate": 4.9122206236017896e-05,
      "loss": 1.0538,
      "num_input_tokens_seen": 1877056,
      "step": 3385
    },
    {
      "epoch": 59.47787610619469,
      "grad_norm": 6.0,
      "learning_rate": 4.911962571029282e-05,
      "loss": 0.9863,
      "num_input_tokens_seen": 1880240,
      "step": 3390
    },
    {
      "epoch": 59.56637168141593,
      "grad_norm": 11.625,
      "learning_rate": 4.9117041465018353e-05,
      "loss": 1.2876,
      "num_input_tokens_seen": 1883472,
      "step": 3395
    },
    {
      "epoch": 59.65486725663717,
      "grad_norm": 8.75,
      "learning_rate": 4.911445350059302e-05,
      "loss": 1.0873,
      "num_input_tokens_seen": 1885968,
      "step": 3400
    },
    {
      "epoch": 59.65486725663717,
      "eval_loss": 1.037407636642456,
      "eval_runtime": 0.9333,
      "eval_samples_per_second": 26.785,
      "eval_steps_per_second": 13.928,
      "num_input_tokens_seen": 1885968,
      "step": 3400
    },
    {
      "epoch": 59.743362831858406,
      "grad_norm": 7.625,
      "learning_rate": 4.9111861817415905e-05,
      "loss": 1.0778,
      "num_input_tokens_seen": 1888896,
      "step": 3405
    },
    {
      "epoch": 59.83185840707964,
      "grad_norm": 7.03125,
      "learning_rate": 4.91092664158867e-05,
      "loss": 0.8853,
      "num_input_tokens_seen": 1891824,
      "step": 3410
    },
    {
      "epoch": 59.92035398230089,
      "grad_norm": 11.0,
      "learning_rate": 4.910666729640563e-05,
      "loss": 0.9945,
      "num_input_tokens_seen": 1894560,
      "step": 3415
    },
    {
      "epoch": 60.0,
      "grad_norm": 24.75,
      "learning_rate": 4.910406445937353e-05,
      "loss": 1.3435,
      "num_input_tokens_seen": 1897016,
      "step": 3420
    },
    {
      "epoch": 60.08849557522124,
      "grad_norm": 6.65625,
      "learning_rate": 4.9101457905191774e-05,
      "loss": 0.8803,
      "num_input_tokens_seen": 1899528,
      "step": 3425
    },
    {
      "epoch": 60.176991150442475,
      "grad_norm": 9.5,
      "learning_rate": 4.909884763426233e-05,
      "loss": 0.8334,
      "num_input_tokens_seen": 1902536,
      "step": 3430
    },
    {
      "epoch": 60.26548672566372,
      "grad_norm": 10.25,
      "learning_rate": 4.9096233646987736e-05,
      "loss": 1.0572,
      "num_input_tokens_seen": 1905464,
      "step": 3435
    },
    {
      "epoch": 60.35398230088496,
      "grad_norm": 8.3125,
      "learning_rate": 4.9093615943771104e-05,
      "loss": 1.1374,
      "num_input_tokens_seen": 1908184,
      "step": 3440
    },
    {
      "epoch": 60.442477876106196,
      "grad_norm": 6.90625,
      "learning_rate": 4.909099452501611e-05,
      "loss": 0.8739,
      "num_input_tokens_seen": 1910968,
      "step": 3445
    },
    {
      "epoch": 60.530973451327434,
      "grad_norm": 12.125,
      "learning_rate": 4.908836939112702e-05,
      "loss": 1.2575,
      "num_input_tokens_seen": 1913672,
      "step": 3450
    },
    {
      "epoch": 60.61946902654867,
      "grad_norm": 13.625,
      "learning_rate": 4.908574054250865e-05,
      "loss": 1.2822,
      "num_input_tokens_seen": 1916392,
      "step": 3455
    },
    {
      "epoch": 60.70796460176991,
      "grad_norm": 7.71875,
      "learning_rate": 4.9083107979566414e-05,
      "loss": 1.002,
      "num_input_tokens_seen": 1919288,
      "step": 3460
    },
    {
      "epoch": 60.796460176991154,
      "grad_norm": 9.0,
      "learning_rate": 4.908047170270628e-05,
      "loss": 1.2225,
      "num_input_tokens_seen": 1922168,
      "step": 3465
    },
    {
      "epoch": 60.88495575221239,
      "grad_norm": 8.125,
      "learning_rate": 4.9077831712334784e-05,
      "loss": 1.1258,
      "num_input_tokens_seen": 1924968,
      "step": 3470
    },
    {
      "epoch": 60.97345132743363,
      "grad_norm": 7.34375,
      "learning_rate": 4.907518800885907e-05,
      "loss": 0.8904,
      "num_input_tokens_seen": 1928136,
      "step": 3475
    },
    {
      "epoch": 61.05309734513274,
      "grad_norm": 8.25,
      "learning_rate": 4.907254059268681e-05,
      "loss": 1.2968,
      "num_input_tokens_seen": 1930760,
      "step": 3480
    },
    {
      "epoch": 61.14159292035398,
      "grad_norm": 5.5625,
      "learning_rate": 4.906988946422628e-05,
      "loss": 0.9641,
      "num_input_tokens_seen": 1933464,
      "step": 3485
    },
    {
      "epoch": 61.230088495575224,
      "grad_norm": 7.9375,
      "learning_rate": 4.9067234623886315e-05,
      "loss": 0.9357,
      "num_input_tokens_seen": 1936376,
      "step": 3490
    },
    {
      "epoch": 61.31858407079646,
      "grad_norm": 7.65625,
      "learning_rate": 4.9064576072076316e-05,
      "loss": 1.2214,
      "num_input_tokens_seen": 1938904,
      "step": 3495
    },
    {
      "epoch": 61.4070796460177,
      "grad_norm": 8.3125,
      "learning_rate": 4.906191380920628e-05,
      "loss": 1.1469,
      "num_input_tokens_seen": 1941688,
      "step": 3500
    },
    {
      "epoch": 61.49557522123894,
      "grad_norm": 8.125,
      "learning_rate": 4.905924783568675e-05,
      "loss": 1.1151,
      "num_input_tokens_seen": 1943960,
      "step": 3505
    },
    {
      "epoch": 61.584070796460175,
      "grad_norm": 8.5625,
      "learning_rate": 4.905657815192886e-05,
      "loss": 0.9633,
      "num_input_tokens_seen": 1946696,
      "step": 3510
    },
    {
      "epoch": 61.67256637168141,
      "grad_norm": 10.0625,
      "learning_rate": 4.90539047583443e-05,
      "loss": 0.9453,
      "num_input_tokens_seen": 1949864,
      "step": 3515
    },
    {
      "epoch": 61.76106194690266,
      "grad_norm": 10.5,
      "learning_rate": 4.905122765534534e-05,
      "loss": 1.1147,
      "num_input_tokens_seen": 1952376,
      "step": 3520
    },
    {
      "epoch": 61.849557522123895,
      "grad_norm": 10.375,
      "learning_rate": 4.9048546843344846e-05,
      "loss": 1.1334,
      "num_input_tokens_seen": 1955112,
      "step": 3525
    },
    {
      "epoch": 61.93805309734513,
      "grad_norm": 8.25,
      "learning_rate": 4.9045862322756206e-05,
      "loss": 1.0344,
      "num_input_tokens_seen": 1957992,
      "step": 3530
    },
    {
      "epoch": 62.017699115044245,
      "grad_norm": 8.5,
      "learning_rate": 4.904317409399342e-05,
      "loss": 1.3761,
      "num_input_tokens_seen": 1960640,
      "step": 3535
    },
    {
      "epoch": 62.10619469026549,
      "grad_norm": 8.0625,
      "learning_rate": 4.904048215747104e-05,
      "loss": 1.1361,
      "num_input_tokens_seen": 1963536,
      "step": 3540
    },
    {
      "epoch": 62.19469026548673,
      "grad_norm": 9.5,
      "learning_rate": 4.90377865136042e-05,
      "loss": 0.9951,
      "num_input_tokens_seen": 1966560,
      "step": 3545
    },
    {
      "epoch": 62.283185840707965,
      "grad_norm": 8.3125,
      "learning_rate": 4.90350871628086e-05,
      "loss": 1.1082,
      "num_input_tokens_seen": 1969696,
      "step": 3550
    },
    {
      "epoch": 62.3716814159292,
      "grad_norm": 8.8125,
      "learning_rate": 4.903238410550052e-05,
      "loss": 1.2363,
      "num_input_tokens_seen": 1972016,
      "step": 3555
    },
    {
      "epoch": 62.46017699115044,
      "grad_norm": 13.9375,
      "learning_rate": 4.90296773420968e-05,
      "loss": 1.0162,
      "num_input_tokens_seen": 1974672,
      "step": 3560
    },
    {
      "epoch": 62.54867256637168,
      "grad_norm": 8.0625,
      "learning_rate": 4.902696687301486e-05,
      "loss": 1.2063,
      "num_input_tokens_seen": 1977584,
      "step": 3565
    },
    {
      "epoch": 62.63716814159292,
      "grad_norm": 10.25,
      "learning_rate": 4.902425269867268e-05,
      "loss": 1.0131,
      "num_input_tokens_seen": 1980144,
      "step": 3570
    },
    {
      "epoch": 62.72566371681416,
      "grad_norm": 13.0625,
      "learning_rate": 4.902153481948883e-05,
      "loss": 1.0049,
      "num_input_tokens_seen": 1983072,
      "step": 3575
    },
    {
      "epoch": 62.8141592920354,
      "grad_norm": 6.09375,
      "learning_rate": 4.901881323588244e-05,
      "loss": 0.9809,
      "num_input_tokens_seen": 1986064,
      "step": 3580
    },
    {
      "epoch": 62.902654867256636,
      "grad_norm": 7.40625,
      "learning_rate": 4.90160879482732e-05,
      "loss": 0.8637,
      "num_input_tokens_seen": 1988960,
      "step": 3585
    },
    {
      "epoch": 62.991150442477874,
      "grad_norm": 8.5625,
      "learning_rate": 4.9013358957081405e-05,
      "loss": 1.1144,
      "num_input_tokens_seen": 1991136,
      "step": 3590
    },
    {
      "epoch": 63.07079646017699,
      "grad_norm": 8.125,
      "learning_rate": 4.901062626272789e-05,
      "loss": 1.0899,
      "num_input_tokens_seen": 1993496,
      "step": 3595
    },
    {
      "epoch": 63.15929203539823,
      "grad_norm": 10.875,
      "learning_rate": 4.900788986563406e-05,
      "loss": 0.9122,
      "num_input_tokens_seen": 1996440,
      "step": 3600
    },
    {
      "epoch": 63.15929203539823,
      "eval_loss": 1.0292187929153442,
      "eval_runtime": 0.9315,
      "eval_samples_per_second": 26.839,
      "eval_steps_per_second": 13.956,
      "num_input_tokens_seen": 1996440,
      "step": 3600
    },
    {
      "epoch": 63.24778761061947,
      "grad_norm": 7.3125,
      "learning_rate": 4.9005149766221915e-05,
      "loss": 0.9871,
      "num_input_tokens_seen": 1999128,
      "step": 3605
    },
    {
      "epoch": 63.336283185840706,
      "grad_norm": 8.0625,
      "learning_rate": 4.9002405964914e-05,
      "loss": 1.2309,
      "num_input_tokens_seen": 2001960,
      "step": 3610
    },
    {
      "epoch": 63.424778761061944,
      "grad_norm": 6.78125,
      "learning_rate": 4.899965846213346e-05,
      "loss": 1.1073,
      "num_input_tokens_seen": 2004840,
      "step": 3615
    },
    {
      "epoch": 63.51327433628319,
      "grad_norm": 8.8125,
      "learning_rate": 4.899690725830399e-05,
      "loss": 1.0038,
      "num_input_tokens_seen": 2007688,
      "step": 3620
    },
    {
      "epoch": 63.60176991150443,
      "grad_norm": 8.0,
      "learning_rate": 4.899415235384985e-05,
      "loss": 1.1712,
      "num_input_tokens_seen": 2010328,
      "step": 3625
    },
    {
      "epoch": 63.690265486725664,
      "grad_norm": 8.8125,
      "learning_rate": 4.899139374919589e-05,
      "loss": 1.0468,
      "num_input_tokens_seen": 2013064,
      "step": 3630
    },
    {
      "epoch": 63.7787610619469,
      "grad_norm": 8.75,
      "learning_rate": 4.898863144476752e-05,
      "loss": 0.9869,
      "num_input_tokens_seen": 2016136,
      "step": 3635
    },
    {
      "epoch": 63.86725663716814,
      "grad_norm": 7.125,
      "learning_rate": 4.898586544099072e-05,
      "loss": 0.8907,
      "num_input_tokens_seen": 2018872,
      "step": 3640
    },
    {
      "epoch": 63.95575221238938,
      "grad_norm": 7.84375,
      "learning_rate": 4.898309573829204e-05,
      "loss": 1.1264,
      "num_input_tokens_seen": 2021512,
      "step": 3645
    },
    {
      "epoch": 64.03539823008849,
      "grad_norm": 9.5,
      "learning_rate": 4.898032233709862e-05,
      "loss": 1.7888,
      "num_input_tokens_seen": 2023712,
      "step": 3650
    },
    {
      "epoch": 64.12389380530973,
      "grad_norm": 11.9375,
      "learning_rate": 4.8977545237838123e-05,
      "loss": 1.0711,
      "num_input_tokens_seen": 2026592,
      "step": 3655
    },
    {
      "epoch": 64.21238938053098,
      "grad_norm": 7.9375,
      "learning_rate": 4.8974764440938836e-05,
      "loss": 1.1135,
      "num_input_tokens_seen": 2029104,
      "step": 3660
    },
    {
      "epoch": 64.30088495575221,
      "grad_norm": 7.40625,
      "learning_rate": 4.897197994682959e-05,
      "loss": 1.119,
      "num_input_tokens_seen": 2032032,
      "step": 3665
    },
    {
      "epoch": 64.38938053097345,
      "grad_norm": 9.375,
      "learning_rate": 4.8969191755939786e-05,
      "loss": 1.086,
      "num_input_tokens_seen": 2034800,
      "step": 3670
    },
    {
      "epoch": 64.47787610619469,
      "grad_norm": 8.875,
      "learning_rate": 4.8966399868699396e-05,
      "loss": 1.1577,
      "num_input_tokens_seen": 2037792,
      "step": 3675
    },
    {
      "epoch": 64.56637168141593,
      "grad_norm": 10.3125,
      "learning_rate": 4.8963604285538965e-05,
      "loss": 1.0748,
      "num_input_tokens_seen": 2040272,
      "step": 3680
    },
    {
      "epoch": 64.65486725663717,
      "grad_norm": 8.75,
      "learning_rate": 4.8960805006889604e-05,
      "loss": 0.9191,
      "num_input_tokens_seen": 2043136,
      "step": 3685
    },
    {
      "epoch": 64.7433628318584,
      "grad_norm": 13.25,
      "learning_rate": 4.8958002033183004e-05,
      "loss": 0.939,
      "num_input_tokens_seen": 2045632,
      "step": 3690
    },
    {
      "epoch": 64.83185840707965,
      "grad_norm": 7.59375,
      "learning_rate": 4.8955195364851414e-05,
      "loss": 1.0619,
      "num_input_tokens_seen": 2049280,
      "step": 3695
    },
    {
      "epoch": 64.92035398230088,
      "grad_norm": 9.625,
      "learning_rate": 4.895238500232766e-05,
      "loss": 1.0257,
      "num_input_tokens_seen": 2052304,
      "step": 3700
    },
    {
      "epoch": 65.0,
      "grad_norm": 8.1875,
      "learning_rate": 4.8949570946045143e-05,
      "loss": 0.9667,
      "num_input_tokens_seen": 2054512,
      "step": 3705
    },
    {
      "epoch": 65.08849557522124,
      "grad_norm": 6.875,
      "learning_rate": 4.89467531964378e-05,
      "loss": 1.0141,
      "num_input_tokens_seen": 2057216,
      "step": 3710
    },
    {
      "epoch": 65.17699115044248,
      "grad_norm": 9.0625,
      "learning_rate": 4.894393175394019e-05,
      "loss": 1.0098,
      "num_input_tokens_seen": 2059840,
      "step": 3715
    },
    {
      "epoch": 65.26548672566372,
      "grad_norm": 10.3125,
      "learning_rate": 4.8941106618987406e-05,
      "loss": 1.0205,
      "num_input_tokens_seen": 2062592,
      "step": 3720
    },
    {
      "epoch": 65.35398230088495,
      "grad_norm": 8.875,
      "learning_rate": 4.893827779201512e-05,
      "loss": 1.0698,
      "num_input_tokens_seen": 2065392,
      "step": 3725
    },
    {
      "epoch": 65.4424778761062,
      "grad_norm": 8.0,
      "learning_rate": 4.893544527345957e-05,
      "loss": 0.9832,
      "num_input_tokens_seen": 2068576,
      "step": 3730
    },
    {
      "epoch": 65.53097345132744,
      "grad_norm": 11.9375,
      "learning_rate": 4.8932609063757563e-05,
      "loss": 1.0627,
      "num_input_tokens_seen": 2071408,
      "step": 3735
    },
    {
      "epoch": 65.61946902654867,
      "grad_norm": 7.3125,
      "learning_rate": 4.8929769163346484e-05,
      "loss": 1.0982,
      "num_input_tokens_seen": 2073952,
      "step": 3740
    },
    {
      "epoch": 65.70796460176992,
      "grad_norm": 9.625,
      "learning_rate": 4.892692557266429e-05,
      "loss": 1.1629,
      "num_input_tokens_seen": 2077168,
      "step": 3745
    },
    {
      "epoch": 65.79646017699115,
      "grad_norm": 7.9375,
      "learning_rate": 4.8924078292149464e-05,
      "loss": 1.09,
      "num_input_tokens_seen": 2079744,
      "step": 3750
    },
    {
      "epoch": 65.88495575221239,
      "grad_norm": 8.8125,
      "learning_rate": 4.892122732224114e-05,
      "loss": 1.0887,
      "num_input_tokens_seen": 2082576,
      "step": 3755
    },
    {
      "epoch": 65.97345132743362,
      "grad_norm": 6.9375,
      "learning_rate": 4.8918372663378944e-05,
      "loss": 1.035,
      "num_input_tokens_seen": 2085376,
      "step": 3760
    },
    {
      "epoch": 66.05309734513274,
      "grad_norm": 6.90625,
      "learning_rate": 4.89155143160031e-05,
      "loss": 1.078,
      "num_input_tokens_seen": 2087672,
      "step": 3765
    },
    {
      "epoch": 66.14159292035399,
      "grad_norm": 5.71875,
      "learning_rate": 4.891265228055441e-05,
      "loss": 0.8322,
      "num_input_tokens_seen": 2090904,
      "step": 3770
    },
    {
      "epoch": 66.23008849557522,
      "grad_norm": 8.6875,
      "learning_rate": 4.890978655747424e-05,
      "loss": 0.8313,
      "num_input_tokens_seen": 2093992,
      "step": 3775
    },
    {
      "epoch": 66.31858407079646,
      "grad_norm": 9.6875,
      "learning_rate": 4.89069171472045e-05,
      "loss": 1.1195,
      "num_input_tokens_seen": 2096456,
      "step": 3780
    },
    {
      "epoch": 66.40707964601769,
      "grad_norm": 7.75,
      "learning_rate": 4.890404405018772e-05,
      "loss": 1.2301,
      "num_input_tokens_seen": 2099544,
      "step": 3785
    },
    {
      "epoch": 66.49557522123894,
      "grad_norm": 7.78125,
      "learning_rate": 4.8901167266866934e-05,
      "loss": 1.1037,
      "num_input_tokens_seen": 2101880,
      "step": 3790
    },
    {
      "epoch": 66.58407079646018,
      "grad_norm": 10.5625,
      "learning_rate": 4.88982867976858e-05,
      "loss": 1.1912,
      "num_input_tokens_seen": 2104664,
      "step": 3795
    },
    {
      "epoch": 66.67256637168141,
      "grad_norm": 8.1875,
      "learning_rate": 4.889540264308852e-05,
      "loss": 1.1853,
      "num_input_tokens_seen": 2107400,
      "step": 3800
    },
    {
      "epoch": 66.67256637168141,
      "eval_loss": 1.0412622690200806,
      "eval_runtime": 0.9319,
      "eval_samples_per_second": 26.827,
      "eval_steps_per_second": 13.95,
      "num_input_tokens_seen": 2107400,
      "step": 3800
    },
    {
      "epoch": 66.76106194690266,
      "grad_norm": 8.0625,
      "learning_rate": 4.889251480351986e-05,
      "loss": 1.0038,
      "num_input_tokens_seen": 2110552,
      "step": 3805
    },
    {
      "epoch": 66.84955752212389,
      "grad_norm": 8.1875,
      "learning_rate": 4.888962327942517e-05,
      "loss": 1.0386,
      "num_input_tokens_seen": 2113064,
      "step": 3810
    },
    {
      "epoch": 66.93805309734513,
      "grad_norm": 8.9375,
      "learning_rate": 4.8886728071250356e-05,
      "loss": 0.956,
      "num_input_tokens_seen": 2116088,
      "step": 3815
    },
    {
      "epoch": 67.01769911504425,
      "grad_norm": 8.25,
      "learning_rate": 4.8883829179441884e-05,
      "loss": 0.9496,
      "num_input_tokens_seen": 2118696,
      "step": 3820
    },
    {
      "epoch": 67.10619469026548,
      "grad_norm": 8.5,
      "learning_rate": 4.888092660444682e-05,
      "loss": 1.252,
      "num_input_tokens_seen": 2121240,
      "step": 3825
    },
    {
      "epoch": 67.19469026548673,
      "grad_norm": 9.75,
      "learning_rate": 4.887802034671276e-05,
      "loss": 1.0999,
      "num_input_tokens_seen": 2123832,
      "step": 3830
    },
    {
      "epoch": 67.28318584070796,
      "grad_norm": 8.25,
      "learning_rate": 4.88751104066879e-05,
      "loss": 0.9364,
      "num_input_tokens_seen": 2126392,
      "step": 3835
    },
    {
      "epoch": 67.3716814159292,
      "grad_norm": 8.375,
      "learning_rate": 4.887219678482098e-05,
      "loss": 1.1722,
      "num_input_tokens_seen": 2128776,
      "step": 3840
    },
    {
      "epoch": 67.46017699115045,
      "grad_norm": 8.1875,
      "learning_rate": 4.8869279481561316e-05,
      "loss": 0.9925,
      "num_input_tokens_seen": 2132312,
      "step": 3845
    },
    {
      "epoch": 67.54867256637168,
      "grad_norm": 7.53125,
      "learning_rate": 4.88663584973588e-05,
      "loss": 1.1251,
      "num_input_tokens_seen": 2134824,
      "step": 3850
    },
    {
      "epoch": 67.63716814159292,
      "grad_norm": 7.96875,
      "learning_rate": 4.8863433832663874e-05,
      "loss": 1.0803,
      "num_input_tokens_seen": 2137944,
      "step": 3855
    },
    {
      "epoch": 67.72566371681415,
      "grad_norm": 7.0625,
      "learning_rate": 4.886050548792757e-05,
      "loss": 0.9451,
      "num_input_tokens_seen": 2141000,
      "step": 3860
    },
    {
      "epoch": 67.8141592920354,
      "grad_norm": 6.625,
      "learning_rate": 4.8857573463601465e-05,
      "loss": 1.0805,
      "num_input_tokens_seen": 2143864,
      "step": 3865
    },
    {
      "epoch": 67.90265486725664,
      "grad_norm": 10.0,
      "learning_rate": 4.885463776013772e-05,
      "loss": 0.9503,
      "num_input_tokens_seen": 2146904,
      "step": 3870
    },
    {
      "epoch": 67.99115044247787,
      "grad_norm": 6.65625,
      "learning_rate": 4.8851698377989056e-05,
      "loss": 1.041,
      "num_input_tokens_seen": 2149656,
      "step": 3875
    },
    {
      "epoch": 68.070796460177,
      "grad_norm": 8.625,
      "learning_rate": 4.884875531760876e-05,
      "loss": 1.1188,
      "num_input_tokens_seen": 2152792,
      "step": 3880
    },
    {
      "epoch": 68.15929203539822,
      "grad_norm": 8.1875,
      "learning_rate": 4.88458085794507e-05,
      "loss": 0.9176,
      "num_input_tokens_seen": 2155688,
      "step": 3885
    },
    {
      "epoch": 68.24778761061947,
      "grad_norm": 11.4375,
      "learning_rate": 4.884285816396929e-05,
      "loss": 1.0054,
      "num_input_tokens_seen": 2157944,
      "step": 3890
    },
    {
      "epoch": 68.33628318584071,
      "grad_norm": 8.75,
      "learning_rate": 4.8839904071619526e-05,
      "loss": 0.9863,
      "num_input_tokens_seen": 2160904,
      "step": 3895
    },
    {
      "epoch": 68.42477876106194,
      "grad_norm": 10.9375,
      "learning_rate": 4.8836946302856955e-05,
      "loss": 1.066,
      "num_input_tokens_seen": 2163912,
      "step": 3900
    },
    {
      "epoch": 68.51327433628319,
      "grad_norm": 10.875,
      "learning_rate": 4.8833984858137715e-05,
      "loss": 1.19,
      "num_input_tokens_seen": 2166520,
      "step": 3905
    },
    {
      "epoch": 68.60176991150442,
      "grad_norm": 7.78125,
      "learning_rate": 4.8831019737918494e-05,
      "loss": 1.2362,
      "num_input_tokens_seen": 2169176,
      "step": 3910
    },
    {
      "epoch": 68.69026548672566,
      "grad_norm": 5.84375,
      "learning_rate": 4.882805094265655e-05,
      "loss": 0.9795,
      "num_input_tokens_seen": 2171640,
      "step": 3915
    },
    {
      "epoch": 68.77876106194691,
      "grad_norm": 5.875,
      "learning_rate": 4.8825078472809706e-05,
      "loss": 0.7832,
      "num_input_tokens_seen": 2174840,
      "step": 3920
    },
    {
      "epoch": 68.86725663716814,
      "grad_norm": 7.75,
      "learning_rate": 4.882210232883635e-05,
      "loss": 1.0928,
      "num_input_tokens_seen": 2177592,
      "step": 3925
    },
    {
      "epoch": 68.95575221238938,
      "grad_norm": 8.375,
      "learning_rate": 4.881912251119546e-05,
      "loss": 1.2172,
      "num_input_tokens_seen": 2180488,
      "step": 3930
    },
    {
      "epoch": 69.03539823008849,
      "grad_norm": 10.0,
      "learning_rate": 4.881613902034654e-05,
      "loss": 1.1043,
      "num_input_tokens_seen": 2182816,
      "step": 3935
    },
    {
      "epoch": 69.12389380530973,
      "grad_norm": 8.0,
      "learning_rate": 4.88131518567497e-05,
      "loss": 1.1269,
      "num_input_tokens_seen": 2185376,
      "step": 3940
    },
    {
      "epoch": 69.21238938053098,
      "grad_norm": 7.9375,
      "learning_rate": 4.881016102086558e-05,
      "loss": 1.0507,
      "num_input_tokens_seen": 2188176,
      "step": 3945
    },
    {
      "epoch": 69.30088495575221,
      "grad_norm": 8.3125,
      "learning_rate": 4.8807166513155425e-05,
      "loss": 1.0213,
      "num_input_tokens_seen": 2191232,
      "step": 3950
    },
    {
      "epoch": 69.38938053097345,
      "grad_norm": 8.0,
      "learning_rate": 4.8804168334081004e-05,
      "loss": 1.1796,
      "num_input_tokens_seen": 2193840,
      "step": 3955
    },
    {
      "epoch": 69.47787610619469,
      "grad_norm": 10.25,
      "learning_rate": 4.880116648410468e-05,
      "loss": 1.1956,
      "num_input_tokens_seen": 2196464,
      "step": 3960
    },
    {
      "epoch": 69.56637168141593,
      "grad_norm": 8.6875,
      "learning_rate": 4.879816096368939e-05,
      "loss": 1.0898,
      "num_input_tokens_seen": 2199312,
      "step": 3965
    },
    {
      "epoch": 69.65486725663717,
      "grad_norm": 6.5,
      "learning_rate": 4.879515177329861e-05,
      "loss": 0.7104,
      "num_input_tokens_seen": 2202080,
      "step": 3970
    },
    {
      "epoch": 69.7433628318584,
      "grad_norm": 12.375,
      "learning_rate": 4.8792138913396394e-05,
      "loss": 1.1217,
      "num_input_tokens_seen": 2205296,
      "step": 3975
    },
    {
      "epoch": 69.83185840707965,
      "grad_norm": 9.125,
      "learning_rate": 4.8789122384447374e-05,
      "loss": 0.8916,
      "num_input_tokens_seen": 2208160,
      "step": 3980
    },
    {
      "epoch": 69.92035398230088,
      "grad_norm": 7.25,
      "learning_rate": 4.878610218691673e-05,
      "loss": 0.9879,
      "num_input_tokens_seen": 2210896,
      "step": 3985
    },
    {
      "epoch": 70.0,
      "grad_norm": 8.3125,
      "learning_rate": 4.87830783212702e-05,
      "loss": 1.1441,
      "num_input_tokens_seen": 2213232,
      "step": 3990
    },
    {
      "epoch": 70.08849557522124,
      "grad_norm": 10.0625,
      "learning_rate": 4.878005078797413e-05,
      "loss": 1.292,
      "num_input_tokens_seen": 2215680,
      "step": 3995
    },
    {
      "epoch": 70.17699115044248,
      "grad_norm": 6.90625,
      "learning_rate": 4.877701958749539e-05,
      "loss": 0.9562,
      "num_input_tokens_seen": 2218352,
      "step": 4000
    },
    {
      "epoch": 70.17699115044248,
      "eval_loss": 1.033484935760498,
      "eval_runtime": 0.931,
      "eval_samples_per_second": 26.854,
      "eval_steps_per_second": 13.964,
      "num_input_tokens_seen": 2218352,
      "step": 4000
    },
    {
      "epoch": 70.26548672566372,
      "grad_norm": 7.65625,
      "learning_rate": 4.877398472030142e-05,
      "loss": 0.9688,
      "num_input_tokens_seen": 2220944,
      "step": 4005
    },
    {
      "epoch": 70.35398230088495,
      "grad_norm": 8.25,
      "learning_rate": 4.877094618686024e-05,
      "loss": 1.225,
      "num_input_tokens_seen": 2223712,
      "step": 4010
    },
    {
      "epoch": 70.4424778761062,
      "grad_norm": 8.0,
      "learning_rate": 4.876790398764045e-05,
      "loss": 0.939,
      "num_input_tokens_seen": 2226496,
      "step": 4015
    },
    {
      "epoch": 70.53097345132744,
      "grad_norm": 9.0625,
      "learning_rate": 4.8764858123111167e-05,
      "loss": 1.0228,
      "num_input_tokens_seen": 2229712,
      "step": 4020
    },
    {
      "epoch": 70.61946902654867,
      "grad_norm": 5.4375,
      "learning_rate": 4.876180859374212e-05,
      "loss": 1.1074,
      "num_input_tokens_seen": 2232528,
      "step": 4025
    },
    {
      "epoch": 70.70796460176992,
      "grad_norm": 7.125,
      "learning_rate": 4.875875540000357e-05,
      "loss": 1.0118,
      "num_input_tokens_seen": 2235248,
      "step": 4030
    },
    {
      "epoch": 70.79646017699115,
      "grad_norm": 8.0625,
      "learning_rate": 4.8755698542366376e-05,
      "loss": 1.2002,
      "num_input_tokens_seen": 2237744,
      "step": 4035
    },
    {
      "epoch": 70.88495575221239,
      "grad_norm": 7.625,
      "learning_rate": 4.875263802130193e-05,
      "loss": 0.9116,
      "num_input_tokens_seen": 2241152,
      "step": 4040
    },
    {
      "epoch": 70.97345132743362,
      "grad_norm": 9.8125,
      "learning_rate": 4.8749573837282207e-05,
      "loss": 1.025,
      "num_input_tokens_seen": 2243888,
      "step": 4045
    },
    {
      "epoch": 71.05309734513274,
      "grad_norm": 5.78125,
      "learning_rate": 4.874650599077974e-05,
      "loss": 1.1629,
      "num_input_tokens_seen": 2246104,
      "step": 4050
    },
    {
      "epoch": 71.14159292035399,
      "grad_norm": 8.9375,
      "learning_rate": 4.874343448226764e-05,
      "loss": 1.3138,
      "num_input_tokens_seen": 2249144,
      "step": 4055
    },
    {
      "epoch": 71.23008849557522,
      "grad_norm": 9.5625,
      "learning_rate": 4.874035931221955e-05,
      "loss": 1.0354,
      "num_input_tokens_seen": 2251624,
      "step": 4060
    },
    {
      "epoch": 71.31858407079646,
      "grad_norm": 8.625,
      "learning_rate": 4.8737280481109724e-05,
      "loss": 1.1266,
      "num_input_tokens_seen": 2254488,
      "step": 4065
    },
    {
      "epoch": 71.40707964601769,
      "grad_norm": 10.75,
      "learning_rate": 4.873419798941294e-05,
      "loss": 0.9863,
      "num_input_tokens_seen": 2257288,
      "step": 4070
    },
    {
      "epoch": 71.49557522123894,
      "grad_norm": 8.4375,
      "learning_rate": 4.873111183760458e-05,
      "loss": 1.0226,
      "num_input_tokens_seen": 2260040,
      "step": 4075
    },
    {
      "epoch": 71.58407079646018,
      "grad_norm": 12.1875,
      "learning_rate": 4.8728022026160537e-05,
      "loss": 1.0441,
      "num_input_tokens_seen": 2262664,
      "step": 4080
    },
    {
      "epoch": 71.67256637168141,
      "grad_norm": 8.625,
      "learning_rate": 4.872492855555732e-05,
      "loss": 0.984,
      "num_input_tokens_seen": 2265336,
      "step": 4085
    },
    {
      "epoch": 71.76106194690266,
      "grad_norm": 6.59375,
      "learning_rate": 4.8721831426271956e-05,
      "loss": 0.9154,
      "num_input_tokens_seen": 2268792,
      "step": 4090
    },
    {
      "epoch": 71.84955752212389,
      "grad_norm": 14.875,
      "learning_rate": 4.87187306387821e-05,
      "loss": 1.1737,
      "num_input_tokens_seen": 2271288,
      "step": 4095
    },
    {
      "epoch": 71.93805309734513,
      "grad_norm": 8.0,
      "learning_rate": 4.87156261935659e-05,
      "loss": 1.0144,
      "num_input_tokens_seen": 2274200,
      "step": 4100
    },
    {
      "epoch": 72.01769911504425,
      "grad_norm": 7.03125,
      "learning_rate": 4.871251809110211e-05,
      "loss": 0.9508,
      "num_input_tokens_seen": 2276720,
      "step": 4105
    },
    {
      "epoch": 72.10619469026548,
      "grad_norm": 8.5625,
      "learning_rate": 4.8709406331870044e-05,
      "loss": 1.0365,
      "num_input_tokens_seen": 2279584,
      "step": 4110
    },
    {
      "epoch": 72.19469026548673,
      "grad_norm": 8.3125,
      "learning_rate": 4.8706290916349574e-05,
      "loss": 1.1531,
      "num_input_tokens_seen": 2281984,
      "step": 4115
    },
    {
      "epoch": 72.28318584070796,
      "grad_norm": 7.15625,
      "learning_rate": 4.8703171845021134e-05,
      "loss": 0.9221,
      "num_input_tokens_seen": 2284704,
      "step": 4120
    },
    {
      "epoch": 72.3716814159292,
      "grad_norm": 8.9375,
      "learning_rate": 4.870004911836572e-05,
      "loss": 1.1203,
      "num_input_tokens_seen": 2287872,
      "step": 4125
    },
    {
      "epoch": 72.46017699115045,
      "grad_norm": 7.0625,
      "learning_rate": 4.869692273686489e-05,
      "loss": 1.1609,
      "num_input_tokens_seen": 2290272,
      "step": 4130
    },
    {
      "epoch": 72.54867256637168,
      "grad_norm": 8.8125,
      "learning_rate": 4.869379270100079e-05,
      "loss": 1.0417,
      "num_input_tokens_seen": 2293088,
      "step": 4135
    },
    {
      "epoch": 72.63716814159292,
      "grad_norm": 8.75,
      "learning_rate": 4.86906590112561e-05,
      "loss": 0.9548,
      "num_input_tokens_seen": 2295744,
      "step": 4140
    },
    {
      "epoch": 72.72566371681415,
      "grad_norm": 10.0,
      "learning_rate": 4.8687521668114064e-05,
      "loss": 0.9396,
      "num_input_tokens_seen": 2298624,
      "step": 4145
    },
    {
      "epoch": 72.8141592920354,
      "grad_norm": 9.125,
      "learning_rate": 4.868438067205853e-05,
      "loss": 1.3139,
      "num_input_tokens_seen": 2301728,
      "step": 4150
    },
    {
      "epoch": 72.90265486725664,
      "grad_norm": 8.8125,
      "learning_rate": 4.8681236023573844e-05,
      "loss": 0.9878,
      "num_input_tokens_seen": 2304400,
      "step": 4155
    },
    {
      "epoch": 72.99115044247787,
      "grad_norm": 7.28125,
      "learning_rate": 4.867808772314497e-05,
      "loss": 1.0482,
      "num_input_tokens_seen": 2307440,
      "step": 4160
    },
    {
      "epoch": 73.070796460177,
      "grad_norm": 8.1875,
      "learning_rate": 4.867493577125741e-05,
      "loss": 0.9905,
      "num_input_tokens_seen": 2309592,
      "step": 4165
    },
    {
      "epoch": 73.15929203539822,
      "grad_norm": 8.875,
      "learning_rate": 4.867178016839725e-05,
      "loss": 1.0967,
      "num_input_tokens_seen": 2312808,
      "step": 4170
    },
    {
      "epoch": 73.24778761061947,
      "grad_norm": 10.75,
      "learning_rate": 4.8668620915051094e-05,
      "loss": 1.0735,
      "num_input_tokens_seen": 2315992,
      "step": 4175
    },
    {
      "epoch": 73.33628318584071,
      "grad_norm": 9.9375,
      "learning_rate": 4.866545801170616e-05,
      "loss": 1.2157,
      "num_input_tokens_seen": 2318424,
      "step": 4180
    },
    {
      "epoch": 73.42477876106194,
      "grad_norm": 7.75,
      "learning_rate": 4.86622914588502e-05,
      "loss": 0.9438,
      "num_input_tokens_seen": 2321304,
      "step": 4185
    },
    {
      "epoch": 73.51327433628319,
      "grad_norm": 7.71875,
      "learning_rate": 4.865912125697154e-05,
      "loss": 1.0769,
      "num_input_tokens_seen": 2324520,
      "step": 4190
    },
    {
      "epoch": 73.60176991150442,
      "grad_norm": 12.6875,
      "learning_rate": 4.865594740655907e-05,
      "loss": 1.1058,
      "num_input_tokens_seen": 2326920,
      "step": 4195
    },
    {
      "epoch": 73.69026548672566,
      "grad_norm": 7.84375,
      "learning_rate": 4.865276990810222e-05,
      "loss": 0.9934,
      "num_input_tokens_seen": 2330072,
      "step": 4200
    },
    {
      "epoch": 73.69026548672566,
      "eval_loss": 1.0307376384735107,
      "eval_runtime": 0.9331,
      "eval_samples_per_second": 26.794,
      "eval_steps_per_second": 13.933,
      "num_input_tokens_seen": 2330072,
      "step": 4200
    },
    {
      "epoch": 73.77876106194691,
      "grad_norm": 11.25,
      "learning_rate": 4.8649588762091016e-05,
      "loss": 1.0285,
      "num_input_tokens_seen": 2332936,
      "step": 4205
    },
    {
      "epoch": 73.86725663716814,
      "grad_norm": 8.5,
      "learning_rate": 4.8646403969016016e-05,
      "loss": 1.0419,
      "num_input_tokens_seen": 2335720,
      "step": 4210
    },
    {
      "epoch": 73.95575221238938,
      "grad_norm": 9.5,
      "learning_rate": 4.864321552936838e-05,
      "loss": 1.1534,
      "num_input_tokens_seen": 2338216,
      "step": 4215
    },
    {
      "epoch": 74.03539823008849,
      "grad_norm": 8.1875,
      "learning_rate": 4.864002344363978e-05,
      "loss": 1.0307,
      "num_input_tokens_seen": 2340392,
      "step": 4220
    },
    {
      "epoch": 74.12389380530973,
      "grad_norm": 6.53125,
      "learning_rate": 4.863682771232248e-05,
      "loss": 0.8574,
      "num_input_tokens_seen": 2343400,
      "step": 4225
    },
    {
      "epoch": 74.21238938053098,
      "grad_norm": 10.4375,
      "learning_rate": 4.8633628335909324e-05,
      "loss": 1.1959,
      "num_input_tokens_seen": 2345992,
      "step": 4230
    },
    {
      "epoch": 74.30088495575221,
      "grad_norm": 8.3125,
      "learning_rate": 4.8630425314893676e-05,
      "loss": 0.9981,
      "num_input_tokens_seen": 2349112,
      "step": 4235
    },
    {
      "epoch": 74.38938053097345,
      "grad_norm": 6.875,
      "learning_rate": 4.862721864976948e-05,
      "loss": 0.9741,
      "num_input_tokens_seen": 2352184,
      "step": 4240
    },
    {
      "epoch": 74.47787610619469,
      "grad_norm": 12.375,
      "learning_rate": 4.862400834103125e-05,
      "loss": 1.0721,
      "num_input_tokens_seen": 2354616,
      "step": 4245
    },
    {
      "epoch": 74.56637168141593,
      "grad_norm": 9.75,
      "learning_rate": 4.862079438917406e-05,
      "loss": 0.9288,
      "num_input_tokens_seen": 2357480,
      "step": 4250
    },
    {
      "epoch": 74.65486725663717,
      "grad_norm": 7.6875,
      "learning_rate": 4.8617576794693536e-05,
      "loss": 1.1928,
      "num_input_tokens_seen": 2360200,
      "step": 4255
    },
    {
      "epoch": 74.7433628318584,
      "grad_norm": 8.3125,
      "learning_rate": 4.8614355558085875e-05,
      "loss": 1.0284,
      "num_input_tokens_seen": 2362920,
      "step": 4260
    },
    {
      "epoch": 74.83185840707965,
      "grad_norm": 9.125,
      "learning_rate": 4.861113067984783e-05,
      "loss": 1.0616,
      "num_input_tokens_seen": 2365576,
      "step": 4265
    },
    {
      "epoch": 74.92035398230088,
      "grad_norm": 10.9375,
      "learning_rate": 4.860790216047671e-05,
      "loss": 1.2074,
      "num_input_tokens_seen": 2368504,
      "step": 4270
    },
    {
      "epoch": 75.0,
      "grad_norm": 32.25,
      "learning_rate": 4.860467000047041e-05,
      "loss": 1.4901,
      "num_input_tokens_seen": 2370576,
      "step": 4275
    },
    {
      "epoch": 75.08849557522124,
      "grad_norm": 7.59375,
      "learning_rate": 4.860143420032737e-05,
      "loss": 1.0099,
      "num_input_tokens_seen": 2373376,
      "step": 4280
    },
    {
      "epoch": 75.17699115044248,
      "grad_norm": 8.1875,
      "learning_rate": 4.859819476054657e-05,
      "loss": 1.1099,
      "num_input_tokens_seen": 2376240,
      "step": 4285
    },
    {
      "epoch": 75.26548672566372,
      "grad_norm": 6.65625,
      "learning_rate": 4.859495168162758e-05,
      "loss": 0.9611,
      "num_input_tokens_seen": 2378864,
      "step": 4290
    },
    {
      "epoch": 75.35398230088495,
      "grad_norm": 9.25,
      "learning_rate": 4.859170496407054e-05,
      "loss": 1.1006,
      "num_input_tokens_seen": 2381696,
      "step": 4295
    },
    {
      "epoch": 75.4424778761062,
      "grad_norm": 6.875,
      "learning_rate": 4.8588454608376114e-05,
      "loss": 1.1048,
      "num_input_tokens_seen": 2384368,
      "step": 4300
    },
    {
      "epoch": 75.53097345132744,
      "grad_norm": 8.1875,
      "learning_rate": 4.8585200615045555e-05,
      "loss": 0.913,
      "num_input_tokens_seen": 2387280,
      "step": 4305
    },
    {
      "epoch": 75.61946902654867,
      "grad_norm": 11.0625,
      "learning_rate": 4.8581942984580674e-05,
      "loss": 1.1311,
      "num_input_tokens_seen": 2389904,
      "step": 4310
    },
    {
      "epoch": 75.70796460176992,
      "grad_norm": 9.375,
      "learning_rate": 4.857868171748384e-05,
      "loss": 1.2695,
      "num_input_tokens_seen": 2393008,
      "step": 4315
    },
    {
      "epoch": 75.79646017699115,
      "grad_norm": 8.875,
      "learning_rate": 4.8575416814257976e-05,
      "loss": 1.0569,
      "num_input_tokens_seen": 2395632,
      "step": 4320
    },
    {
      "epoch": 75.88495575221239,
      "grad_norm": 7.625,
      "learning_rate": 4.857214827540657e-05,
      "loss": 0.8947,
      "num_input_tokens_seen": 2398496,
      "step": 4325
    },
    {
      "epoch": 75.97345132743362,
      "grad_norm": 7.4375,
      "learning_rate": 4.856887610143367e-05,
      "loss": 1.0112,
      "num_input_tokens_seen": 2401776,
      "step": 4330
    },
    {
      "epoch": 76.05309734513274,
      "grad_norm": 7.375,
      "learning_rate": 4.8565600292843896e-05,
      "loss": 1.0474,
      "num_input_tokens_seen": 2404304,
      "step": 4335
    },
    {
      "epoch": 76.14159292035399,
      "grad_norm": 7.78125,
      "learning_rate": 4.856232085014241e-05,
      "loss": 0.9461,
      "num_input_tokens_seen": 2406688,
      "step": 4340
    },
    {
      "epoch": 76.23008849557522,
      "grad_norm": 11.4375,
      "learning_rate": 4.855903777383495e-05,
      "loss": 0.8782,
      "num_input_tokens_seen": 2409696,
      "step": 4345
    },
    {
      "epoch": 76.31858407079646,
      "grad_norm": 8.6875,
      "learning_rate": 4.85557510644278e-05,
      "loss": 0.9893,
      "num_input_tokens_seen": 2412784,
      "step": 4350
    },
    {
      "epoch": 76.40707964601769,
      "grad_norm": 7.65625,
      "learning_rate": 4.855246072242782e-05,
      "loss": 1.1651,
      "num_input_tokens_seen": 2415664,
      "step": 4355
    },
    {
      "epoch": 76.49557522123894,
      "grad_norm": 9.5625,
      "learning_rate": 4.8549166748342414e-05,
      "loss": 1.354,
      "num_input_tokens_seen": 2418352,
      "step": 4360
    },
    {
      "epoch": 76.58407079646018,
      "grad_norm": 10.875,
      "learning_rate": 4.8545869142679556e-05,
      "loss": 1.0407,
      "num_input_tokens_seen": 2420848,
      "step": 4365
    },
    {
      "epoch": 76.67256637168141,
      "grad_norm": 7.5,
      "learning_rate": 4.8542567905947776e-05,
      "loss": 1.2074,
      "num_input_tokens_seen": 2423728,
      "step": 4370
    },
    {
      "epoch": 76.76106194690266,
      "grad_norm": 7.40625,
      "learning_rate": 4.853926303865618e-05,
      "loss": 1.0713,
      "num_input_tokens_seen": 2426768,
      "step": 4375
    },
    {
      "epoch": 76.84955752212389,
      "grad_norm": 7.65625,
      "learning_rate": 4.853595454131441e-05,
      "loss": 0.9709,
      "num_input_tokens_seen": 2429696,
      "step": 4380
    },
    {
      "epoch": 76.93805309734513,
      "grad_norm": 7.875,
      "learning_rate": 4.8532642414432674e-05,
      "loss": 1.1244,
      "num_input_tokens_seen": 2432400,
      "step": 4385
    },
    {
      "epoch": 77.01769911504425,
      "grad_norm": 11.4375,
      "learning_rate": 4.8529326658521754e-05,
      "loss": 1.0042,
      "num_input_tokens_seen": 2434448,
      "step": 4390
    },
    {
      "epoch": 77.10619469026548,
      "grad_norm": 7.625,
      "learning_rate": 4.8526007274092965e-05,
      "loss": 1.0752,
      "num_input_tokens_seen": 2437456,
      "step": 4395
    },
    {
      "epoch": 77.19469026548673,
      "grad_norm": 8.0625,
      "learning_rate": 4.852268426165822e-05,
      "loss": 1.2125,
      "num_input_tokens_seen": 2440176,
      "step": 4400
    },
    {
      "epoch": 77.19469026548673,
      "eval_loss": 1.0342280864715576,
      "eval_runtime": 0.9308,
      "eval_samples_per_second": 26.858,
      "eval_steps_per_second": 13.966,
      "num_input_tokens_seen": 2440176,
      "step": 4400
    },
    {
      "epoch": 77.28318584070796,
      "grad_norm": 9.1875,
      "learning_rate": 4.851935762172995e-05,
      "loss": 1.0244,
      "num_input_tokens_seen": 2443136,
      "step": 4405
    },
    {
      "epoch": 77.3716814159292,
      "grad_norm": 6.21875,
      "learning_rate": 4.8516027354821175e-05,
      "loss": 1.0819,
      "num_input_tokens_seen": 2445680,
      "step": 4410
    },
    {
      "epoch": 77.46017699115045,
      "grad_norm": 9.0625,
      "learning_rate": 4.851269346144546e-05,
      "loss": 1.1346,
      "num_input_tokens_seen": 2447952,
      "step": 4415
    },
    {
      "epoch": 77.54867256637168,
      "grad_norm": 8.5625,
      "learning_rate": 4.850935594211693e-05,
      "loss": 0.9837,
      "num_input_tokens_seen": 2450672,
      "step": 4420
    },
    {
      "epoch": 77.63716814159292,
      "grad_norm": 8.8125,
      "learning_rate": 4.850601479735029e-05,
      "loss": 1.0729,
      "num_input_tokens_seen": 2453552,
      "step": 4425
    },
    {
      "epoch": 77.72566371681415,
      "grad_norm": 8.25,
      "learning_rate": 4.850267002766076e-05,
      "loss": 1.1657,
      "num_input_tokens_seen": 2456192,
      "step": 4430
    },
    {
      "epoch": 77.8141592920354,
      "grad_norm": 10.875,
      "learning_rate": 4.849932163356417e-05,
      "loss": 0.9076,
      "num_input_tokens_seen": 2460016,
      "step": 4435
    },
    {
      "epoch": 77.90265486725664,
      "grad_norm": 7.59375,
      "learning_rate": 4.8495969615576864e-05,
      "loss": 0.9228,
      "num_input_tokens_seen": 2462736,
      "step": 4440
    },
    {
      "epoch": 77.99115044247787,
      "grad_norm": 9.75,
      "learning_rate": 4.849261397421577e-05,
      "loss": 1.1188,
      "num_input_tokens_seen": 2465312,
      "step": 4445
    },
    {
      "epoch": 78.070796460177,
      "grad_norm": 8.625,
      "learning_rate": 4.848925470999839e-05,
      "loss": 0.8862,
      "num_input_tokens_seen": 2468016,
      "step": 4450
    },
    {
      "epoch": 78.15929203539822,
      "grad_norm": 10.5,
      "learning_rate": 4.848589182344273e-05,
      "loss": 1.1276,
      "num_input_tokens_seen": 2470752,
      "step": 4455
    },
    {
      "epoch": 78.24778761061947,
      "grad_norm": 8.125,
      "learning_rate": 4.848252531506742e-05,
      "loss": 1.0926,
      "num_input_tokens_seen": 2473200,
      "step": 4460
    },
    {
      "epoch": 78.33628318584071,
      "grad_norm": 8.625,
      "learning_rate": 4.847915518539161e-05,
      "loss": 1.0926,
      "num_input_tokens_seen": 2475648,
      "step": 4465
    },
    {
      "epoch": 78.42477876106194,
      "grad_norm": 9.0,
      "learning_rate": 4.847578143493501e-05,
      "loss": 0.9726,
      "num_input_tokens_seen": 2478112,
      "step": 4470
    },
    {
      "epoch": 78.51327433628319,
      "grad_norm": 9.75,
      "learning_rate": 4.847240406421789e-05,
      "loss": 0.8999,
      "num_input_tokens_seen": 2481392,
      "step": 4475
    },
    {
      "epoch": 78.60176991150442,
      "grad_norm": 7.75,
      "learning_rate": 4.84690230737611e-05,
      "loss": 1.2127,
      "num_input_tokens_seen": 2484128,
      "step": 4480
    },
    {
      "epoch": 78.69026548672566,
      "grad_norm": 11.25,
      "learning_rate": 4.846563846408602e-05,
      "loss": 0.9943,
      "num_input_tokens_seen": 2487008,
      "step": 4485
    },
    {
      "epoch": 78.77876106194691,
      "grad_norm": 13.0,
      "learning_rate": 4.84622502357146e-05,
      "loss": 1.0937,
      "num_input_tokens_seen": 2489760,
      "step": 4490
    },
    {
      "epoch": 78.86725663716814,
      "grad_norm": 7.5625,
      "learning_rate": 4.8458858389169345e-05,
      "loss": 1.0136,
      "num_input_tokens_seen": 2492704,
      "step": 4495
    },
    {
      "epoch": 78.95575221238938,
      "grad_norm": 8.25,
      "learning_rate": 4.8455462924973334e-05,
      "loss": 1.0878,
      "num_input_tokens_seen": 2495600,
      "step": 4500
    },
    {
      "epoch": 79.03539823008849,
      "grad_norm": 7.59375,
      "learning_rate": 4.845206384365018e-05,
      "loss": 1.1666,
      "num_input_tokens_seen": 2497816,
      "step": 4505
    },
    {
      "epoch": 79.12389380530973,
      "grad_norm": 9.75,
      "learning_rate": 4.844866114572405e-05,
      "loss": 1.1535,
      "num_input_tokens_seen": 2500568,
      "step": 4510
    },
    {
      "epoch": 79.21238938053098,
      "grad_norm": 8.5625,
      "learning_rate": 4.8445254831719706e-05,
      "loss": 1.0959,
      "num_input_tokens_seen": 2503176,
      "step": 4515
    },
    {
      "epoch": 79.30088495575221,
      "grad_norm": 10.25,
      "learning_rate": 4.8441844902162434e-05,
      "loss": 1.0669,
      "num_input_tokens_seen": 2506424,
      "step": 4520
    },
    {
      "epoch": 79.38938053097345,
      "grad_norm": 7.40625,
      "learning_rate": 4.843843135757809e-05,
      "loss": 1.1299,
      "num_input_tokens_seen": 2509064,
      "step": 4525
    },
    {
      "epoch": 79.47787610619469,
      "grad_norm": 11.0,
      "learning_rate": 4.843501419849308e-05,
      "loss": 1.2051,
      "num_input_tokens_seen": 2512136,
      "step": 4530
    },
    {
      "epoch": 79.56637168141593,
      "grad_norm": 8.5625,
      "learning_rate": 4.8431593425434386e-05,
      "loss": 0.9508,
      "num_input_tokens_seen": 2514840,
      "step": 4535
    },
    {
      "epoch": 79.65486725663717,
      "grad_norm": 9.4375,
      "learning_rate": 4.8428169038929526e-05,
      "loss": 0.9957,
      "num_input_tokens_seen": 2517832,
      "step": 4540
    },
    {
      "epoch": 79.7433628318584,
      "grad_norm": 6.0625,
      "learning_rate": 4.8424741039506575e-05,
      "loss": 1.0463,
      "num_input_tokens_seen": 2520536,
      "step": 4545
    },
    {
      "epoch": 79.83185840707965,
      "grad_norm": 6.9375,
      "learning_rate": 4.842130942769419e-05,
      "loss": 0.9185,
      "num_input_tokens_seen": 2523576,
      "step": 4550
    },
    {
      "epoch": 79.92035398230088,
      "grad_norm": 11.4375,
      "learning_rate": 4.841787420402156e-05,
      "loss": 0.9422,
      "num_input_tokens_seen": 2526312,
      "step": 4555
    },
    {
      "epoch": 80.0,
      "grad_norm": 9.5625,
      "learning_rate": 4.841443536901844e-05,
      "loss": 1.169,
      "num_input_tokens_seen": 2528720,
      "step": 4560
    },
    {
      "epoch": 80.08849557522124,
      "grad_norm": 7.3125,
      "learning_rate": 4.841099292321514e-05,
      "loss": 1.1365,
      "num_input_tokens_seen": 2531568,
      "step": 4565
    },
    {
      "epoch": 80.17699115044248,
      "grad_norm": 13.125,
      "learning_rate": 4.8407546867142525e-05,
      "loss": 1.0092,
      "num_input_tokens_seen": 2534432,
      "step": 4570
    },
    {
      "epoch": 80.26548672566372,
      "grad_norm": 8.25,
      "learning_rate": 4.840409720133203e-05,
      "loss": 0.9128,
      "num_input_tokens_seen": 2537184,
      "step": 4575
    },
    {
      "epoch": 80.35398230088495,
      "grad_norm": 6.15625,
      "learning_rate": 4.8400643926315634e-05,
      "loss": 1.0098,
      "num_input_tokens_seen": 2539952,
      "step": 4580
    },
    {
      "epoch": 80.4424778761062,
      "grad_norm": 5.59375,
      "learning_rate": 4.839718704262587e-05,
      "loss": 0.8995,
      "num_input_tokens_seen": 2542560,
      "step": 4585
    },
    {
      "epoch": 80.53097345132744,
      "grad_norm": 10.6875,
      "learning_rate": 4.839372655079585e-05,
      "loss": 1.1203,
      "num_input_tokens_seen": 2545328,
      "step": 4590
    },
    {
      "epoch": 80.61946902654867,
      "grad_norm": 6.375,
      "learning_rate": 4.83902624513592e-05,
      "loss": 1.0966,
      "num_input_tokens_seen": 2548112,
      "step": 4595
    },
    {
      "epoch": 80.70796460176992,
      "grad_norm": 11.75,
      "learning_rate": 4.838679474485014e-05,
      "loss": 1.0047,
      "num_input_tokens_seen": 2551216,
      "step": 4600
    },
    {
      "epoch": 80.70796460176992,
      "eval_loss": 1.0329639911651611,
      "eval_runtime": 0.9324,
      "eval_samples_per_second": 26.812,
      "eval_steps_per_second": 13.942,
      "num_input_tokens_seen": 2551216,
      "step": 4600
    },
    {
      "epoch": 80.79646017699115,
      "grad_norm": 7.96875,
      "learning_rate": 4.838332343180343e-05,
      "loss": 1.0389,
      "num_input_tokens_seen": 2553952,
      "step": 4605
    },
    {
      "epoch": 80.88495575221239,
      "grad_norm": 8.8125,
      "learning_rate": 4.83798485127544e-05,
      "loss": 1.1833,
      "num_input_tokens_seen": 2556688,
      "step": 4610
    },
    {
      "epoch": 80.97345132743362,
      "grad_norm": 9.125,
      "learning_rate": 4.837636998823892e-05,
      "loss": 1.142,
      "num_input_tokens_seen": 2559808,
      "step": 4615
    },
    {
      "epoch": 81.05309734513274,
      "grad_norm": 7.4375,
      "learning_rate": 4.8372887858793414e-05,
      "loss": 1.2727,
      "num_input_tokens_seen": 2562104,
      "step": 4620
    },
    {
      "epoch": 81.14159292035399,
      "grad_norm": 6.46875,
      "learning_rate": 4.836940212495489e-05,
      "loss": 1.13,
      "num_input_tokens_seen": 2565080,
      "step": 4625
    },
    {
      "epoch": 81.23008849557522,
      "grad_norm": 6.625,
      "learning_rate": 4.836591278726087e-05,
      "loss": 0.8753,
      "num_input_tokens_seen": 2567704,
      "step": 4630
    },
    {
      "epoch": 81.31858407079646,
      "grad_norm": 9.0625,
      "learning_rate": 4.836241984624947e-05,
      "loss": 1.0638,
      "num_input_tokens_seen": 2570408,
      "step": 4635
    },
    {
      "epoch": 81.40707964601769,
      "grad_norm": 9.3125,
      "learning_rate": 4.8358923302459336e-05,
      "loss": 1.3096,
      "num_input_tokens_seen": 2572856,
      "step": 4640
    },
    {
      "epoch": 81.49557522123894,
      "grad_norm": 9.0,
      "learning_rate": 4.835542315642968e-05,
      "loss": 1.0253,
      "num_input_tokens_seen": 2576024,
      "step": 4645
    },
    {
      "epoch": 81.58407079646018,
      "grad_norm": 8.0,
      "learning_rate": 4.8351919408700274e-05,
      "loss": 0.9927,
      "num_input_tokens_seen": 2578648,
      "step": 4650
    },
    {
      "epoch": 81.67256637168141,
      "grad_norm": 11.25,
      "learning_rate": 4.834841205981144e-05,
      "loss": 1.1445,
      "num_input_tokens_seen": 2581624,
      "step": 4655
    },
    {
      "epoch": 81.76106194690266,
      "grad_norm": 8.3125,
      "learning_rate": 4.8344901110304054e-05,
      "loss": 1.0044,
      "num_input_tokens_seen": 2584856,
      "step": 4660
    },
    {
      "epoch": 81.84955752212389,
      "grad_norm": 9.25,
      "learning_rate": 4.8341386560719534e-05,
      "loss": 1.0003,
      "num_input_tokens_seen": 2587976,
      "step": 4665
    },
    {
      "epoch": 81.93805309734513,
      "grad_norm": 11.0,
      "learning_rate": 4.833786841159989e-05,
      "loss": 1.1092,
      "num_input_tokens_seen": 2590456,
      "step": 4670
    },
    {
      "epoch": 82.01769911504425,
      "grad_norm": 9.8125,
      "learning_rate": 4.833434666348765e-05,
      "loss": 1.0408,
      "num_input_tokens_seen": 2593056,
      "step": 4675
    },
    {
      "epoch": 82.10619469026548,
      "grad_norm": 9.5,
      "learning_rate": 4.833082131692592e-05,
      "loss": 0.8716,
      "num_input_tokens_seen": 2595744,
      "step": 4680
    },
    {
      "epoch": 82.19469026548673,
      "grad_norm": 8.75,
      "learning_rate": 4.832729237245835e-05,
      "loss": 1.0153,
      "num_input_tokens_seen": 2598352,
      "step": 4685
    },
    {
      "epoch": 82.28318584070796,
      "grad_norm": 7.5625,
      "learning_rate": 4.8323759830629145e-05,
      "loss": 1.2748,
      "num_input_tokens_seen": 2600720,
      "step": 4690
    },
    {
      "epoch": 82.3716814159292,
      "grad_norm": 7.25,
      "learning_rate": 4.8320223691983066e-05,
      "loss": 0.9397,
      "num_input_tokens_seen": 2603376,
      "step": 4695
    },
    {
      "epoch": 82.46017699115045,
      "grad_norm": 7.40625,
      "learning_rate": 4.831668395706544e-05,
      "loss": 1.054,
      "num_input_tokens_seen": 2605968,
      "step": 4700
    },
    {
      "epoch": 82.54867256637168,
      "grad_norm": 7.21875,
      "learning_rate": 4.8313140626422125e-05,
      "loss": 1.2583,
      "num_input_tokens_seen": 2609232,
      "step": 4705
    },
    {
      "epoch": 82.63716814159292,
      "grad_norm": 10.0625,
      "learning_rate": 4.830959370059956e-05,
      "loss": 1.1316,
      "num_input_tokens_seen": 2612112,
      "step": 4710
    },
    {
      "epoch": 82.72566371681415,
      "grad_norm": 9.6875,
      "learning_rate": 4.830604318014472e-05,
      "loss": 1.0984,
      "num_input_tokens_seen": 2615264,
      "step": 4715
    },
    {
      "epoch": 82.8141592920354,
      "grad_norm": 8.125,
      "learning_rate": 4.830248906560514e-05,
      "loss": 0.9323,
      "num_input_tokens_seen": 2617904,
      "step": 4720
    },
    {
      "epoch": 82.90265486725664,
      "grad_norm": 8.625,
      "learning_rate": 4.829893135752891e-05,
      "loss": 0.9594,
      "num_input_tokens_seen": 2621216,
      "step": 4725
    },
    {
      "epoch": 82.99115044247787,
      "grad_norm": 12.75,
      "learning_rate": 4.829537005646466e-05,
      "loss": 1.0353,
      "num_input_tokens_seen": 2624048,
      "step": 4730
    },
    {
      "epoch": 83.070796460177,
      "grad_norm": 8.4375,
      "learning_rate": 4.8291805162961615e-05,
      "loss": 0.9786,
      "num_input_tokens_seen": 2626256,
      "step": 4735
    },
    {
      "epoch": 83.15929203539822,
      "grad_norm": 9.3125,
      "learning_rate": 4.82882366775695e-05,
      "loss": 0.8802,
      "num_input_tokens_seen": 2628832,
      "step": 4740
    },
    {
      "epoch": 83.24778761061947,
      "grad_norm": 9.1875,
      "learning_rate": 4.828466460083864e-05,
      "loss": 1.2609,
      "num_input_tokens_seen": 2631360,
      "step": 4745
    },
    {
      "epoch": 83.33628318584071,
      "grad_norm": 7.65625,
      "learning_rate": 4.8281088933319877e-05,
      "loss": 1.0545,
      "num_input_tokens_seen": 2634688,
      "step": 4750
    },
    {
      "epoch": 83.42477876106194,
      "grad_norm": 8.3125,
      "learning_rate": 4.827750967556464e-05,
      "loss": 1.2021,
      "num_input_tokens_seen": 2637344,
      "step": 4755
    },
    {
      "epoch": 83.51327433628319,
      "grad_norm": 8.75,
      "learning_rate": 4.827392682812488e-05,
      "loss": 1.1432,
      "num_input_tokens_seen": 2640256,
      "step": 4760
    },
    {
      "epoch": 83.60176991150442,
      "grad_norm": 8.625,
      "learning_rate": 4.827034039155312e-05,
      "loss": 1.1484,
      "num_input_tokens_seen": 2643248,
      "step": 4765
    },
    {
      "epoch": 83.69026548672566,
      "grad_norm": 8.5625,
      "learning_rate": 4.8266750366402445e-05,
      "loss": 1.0406,
      "num_input_tokens_seen": 2646128,
      "step": 4770
    },
    {
      "epoch": 83.77876106194691,
      "grad_norm": 7.3125,
      "learning_rate": 4.8263156753226476e-05,
      "loss": 0.9232,
      "num_input_tokens_seen": 2649040,
      "step": 4775
    },
    {
      "epoch": 83.86725663716814,
      "grad_norm": 7.3125,
      "learning_rate": 4.8259559552579394e-05,
      "loss": 1.0066,
      "num_input_tokens_seen": 2652160,
      "step": 4780
    },
    {
      "epoch": 83.95575221238938,
      "grad_norm": 7.46875,
      "learning_rate": 4.825595876501593e-05,
      "loss": 1.0671,
      "num_input_tokens_seen": 2654912,
      "step": 4785
    },
    {
      "epoch": 84.03539823008849,
      "grad_norm": 8.8125,
      "learning_rate": 4.825235439109137e-05,
      "loss": 1.2851,
      "num_input_tokens_seen": 2657328,
      "step": 4790
    },
    {
      "epoch": 84.12389380530973,
      "grad_norm": 8.8125,
      "learning_rate": 4.824874643136156e-05,
      "loss": 1.1723,
      "num_input_tokens_seen": 2660080,
      "step": 4795
    },
    {
      "epoch": 84.21238938053098,
      "grad_norm": 6.90625,
      "learning_rate": 4.824513488638288e-05,
      "loss": 1.0866,
      "num_input_tokens_seen": 2662848,
      "step": 4800
    },
    {
      "epoch": 84.21238938053098,
      "eval_loss": 1.0257511138916016,
      "eval_runtime": 0.9319,
      "eval_samples_per_second": 26.826,
      "eval_steps_per_second": 13.95,
      "num_input_tokens_seen": 2662848,
      "step": 4800
    },
    {
      "epoch": 84.30088495575221,
      "grad_norm": 12.125,
      "learning_rate": 4.8241519756712293e-05,
      "loss": 1.1786,
      "num_input_tokens_seen": 2665376,
      "step": 4805
    },
    {
      "epoch": 84.38938053097345,
      "grad_norm": 8.8125,
      "learning_rate": 4.8237901042907285e-05,
      "loss": 0.9217,
      "num_input_tokens_seen": 2668496,
      "step": 4810
    },
    {
      "epoch": 84.47787610619469,
      "grad_norm": 8.125,
      "learning_rate": 4.823427874552591e-05,
      "loss": 1.1439,
      "num_input_tokens_seen": 2671344,
      "step": 4815
    },
    {
      "epoch": 84.56637168141593,
      "grad_norm": 7.46875,
      "learning_rate": 4.823065286512677e-05,
      "loss": 0.8704,
      "num_input_tokens_seen": 2674064,
      "step": 4820
    },
    {
      "epoch": 84.65486725663717,
      "grad_norm": 9.625,
      "learning_rate": 4.8227023402269025e-05,
      "loss": 1.2408,
      "num_input_tokens_seen": 2677072,
      "step": 4825
    },
    {
      "epoch": 84.7433628318584,
      "grad_norm": 6.40625,
      "learning_rate": 4.822339035751239e-05,
      "loss": 0.9874,
      "num_input_tokens_seen": 2679568,
      "step": 4830
    },
    {
      "epoch": 84.83185840707965,
      "grad_norm": 9.0,
      "learning_rate": 4.8219753731417104e-05,
      "loss": 1.1749,
      "num_input_tokens_seen": 2681984,
      "step": 4835
    },
    {
      "epoch": 84.92035398230088,
      "grad_norm": 5.15625,
      "learning_rate": 4.821611352454401e-05,
      "loss": 0.9841,
      "num_input_tokens_seen": 2685296,
      "step": 4840
    },
    {
      "epoch": 85.0,
      "grad_norm": 7.96875,
      "learning_rate": 4.8212469737454444e-05,
      "loss": 0.9246,
      "num_input_tokens_seen": 2687704,
      "step": 4845
    },
    {
      "epoch": 85.08849557522124,
      "grad_norm": 8.0625,
      "learning_rate": 4.820882237071035e-05,
      "loss": 1.0019,
      "num_input_tokens_seen": 2690616,
      "step": 4850
    },
    {
      "epoch": 85.17699115044248,
      "grad_norm": 10.875,
      "learning_rate": 4.820517142487417e-05,
      "loss": 1.1614,
      "num_input_tokens_seen": 2693528,
      "step": 4855
    },
    {
      "epoch": 85.26548672566372,
      "grad_norm": 7.21875,
      "learning_rate": 4.8201516900508956e-05,
      "loss": 1.0614,
      "num_input_tokens_seen": 2696184,
      "step": 4860
    },
    {
      "epoch": 85.35398230088495,
      "grad_norm": 9.1875,
      "learning_rate": 4.819785879817827e-05,
      "loss": 1.2948,
      "num_input_tokens_seen": 2699000,
      "step": 4865
    },
    {
      "epoch": 85.4424778761062,
      "grad_norm": 8.9375,
      "learning_rate": 4.8194197118446226e-05,
      "loss": 1.1429,
      "num_input_tokens_seen": 2701944,
      "step": 4870
    },
    {
      "epoch": 85.53097345132744,
      "grad_norm": 7.8125,
      "learning_rate": 4.819053186187752e-05,
      "loss": 0.9268,
      "num_input_tokens_seen": 2704776,
      "step": 4875
    },
    {
      "epoch": 85.61946902654867,
      "grad_norm": 9.8125,
      "learning_rate": 4.818686302903736e-05,
      "loss": 0.9275,
      "num_input_tokens_seen": 2707816,
      "step": 4880
    },
    {
      "epoch": 85.70796460176992,
      "grad_norm": 8.0,
      "learning_rate": 4.818319062049154e-05,
      "loss": 1.0834,
      "num_input_tokens_seen": 2710728,
      "step": 4885
    },
    {
      "epoch": 85.79646017699115,
      "grad_norm": 7.53125,
      "learning_rate": 4.817951463680639e-05,
      "loss": 1.1133,
      "num_input_tokens_seen": 2713400,
      "step": 4890
    },
    {
      "epoch": 85.88495575221239,
      "grad_norm": 8.1875,
      "learning_rate": 4.817583507854879e-05,
      "loss": 1.0582,
      "num_input_tokens_seen": 2716088,
      "step": 4895
    },
    {
      "epoch": 85.97345132743362,
      "grad_norm": 9.5,
      "learning_rate": 4.817215194628617e-05,
      "loss": 0.9874,
      "num_input_tokens_seen": 2718888,
      "step": 4900
    },
    {
      "epoch": 86.05309734513274,
      "grad_norm": 8.5,
      "learning_rate": 4.816846524058653e-05,
      "loss": 0.9804,
      "num_input_tokens_seen": 2721088,
      "step": 4905
    },
    {
      "epoch": 86.14159292035399,
      "grad_norm": 8.0,
      "learning_rate": 4.816477496201839e-05,
      "loss": 1.0055,
      "num_input_tokens_seen": 2724176,
      "step": 4910
    },
    {
      "epoch": 86.23008849557522,
      "grad_norm": 8.3125,
      "learning_rate": 4.8161081111150845e-05,
      "loss": 0.9172,
      "num_input_tokens_seen": 2726544,
      "step": 4915
    },
    {
      "epoch": 86.31858407079646,
      "grad_norm": 8.625,
      "learning_rate": 4.815738368855354e-05,
      "loss": 0.9988,
      "num_input_tokens_seen": 2729008,
      "step": 4920
    },
    {
      "epoch": 86.40707964601769,
      "grad_norm": 9.0625,
      "learning_rate": 4.815368269479664e-05,
      "loss": 1.1071,
      "num_input_tokens_seen": 2731952,
      "step": 4925
    },
    {
      "epoch": 86.49557522123894,
      "grad_norm": 7.6875,
      "learning_rate": 4.814997813045092e-05,
      "loss": 1.1259,
      "num_input_tokens_seen": 2734336,
      "step": 4930
    },
    {
      "epoch": 86.58407079646018,
      "grad_norm": 8.25,
      "learning_rate": 4.814626999608764e-05,
      "loss": 1.0702,
      "num_input_tokens_seen": 2737120,
      "step": 4935
    },
    {
      "epoch": 86.67256637168141,
      "grad_norm": 8.0625,
      "learning_rate": 4.814255829227865e-05,
      "loss": 0.9583,
      "num_input_tokens_seen": 2739824,
      "step": 4940
    },
    {
      "epoch": 86.76106194690266,
      "grad_norm": 10.5,
      "learning_rate": 4.813884301959635e-05,
      "loss": 1.0925,
      "num_input_tokens_seen": 2742672,
      "step": 4945
    },
    {
      "epoch": 86.84955752212389,
      "grad_norm": 8.75,
      "learning_rate": 4.813512417861368e-05,
      "loss": 1.1307,
      "num_input_tokens_seen": 2745616,
      "step": 4950
    },
    {
      "epoch": 86.93805309734513,
      "grad_norm": 7.21875,
      "learning_rate": 4.813140176990411e-05,
      "loss": 1.0027,
      "num_input_tokens_seen": 2748656,
      "step": 4955
    },
    {
      "epoch": 87.01769911504425,
      "grad_norm": 6.9375,
      "learning_rate": 4.8127675794041714e-05,
      "loss": 0.9581,
      "num_input_tokens_seen": 2751376,
      "step": 4960
    },
    {
      "epoch": 87.10619469026548,
      "grad_norm": 8.6875,
      "learning_rate": 4.812394625160107e-05,
      "loss": 1.274,
      "num_input_tokens_seen": 2754064,
      "step": 4965
    },
    {
      "epoch": 87.19469026548673,
      "grad_norm": 8.4375,
      "learning_rate": 4.812021314315732e-05,
      "loss": 1.1429,
      "num_input_tokens_seen": 2756768,
      "step": 4970
    },
    {
      "epoch": 87.28318584070796,
      "grad_norm": 8.0625,
      "learning_rate": 4.811647646928616e-05,
      "loss": 1.1099,
      "num_input_tokens_seen": 2759344,
      "step": 4975
    },
    {
      "epoch": 87.3716814159292,
      "grad_norm": 11.3125,
      "learning_rate": 4.8112736230563814e-05,
      "loss": 1.1401,
      "num_input_tokens_seen": 2762416,
      "step": 4980
    },
    {
      "epoch": 87.46017699115045,
      "grad_norm": 6.5625,
      "learning_rate": 4.81089924275671e-05,
      "loss": 1.0319,
      "num_input_tokens_seen": 2765120,
      "step": 4985
    },
    {
      "epoch": 87.54867256637168,
      "grad_norm": 8.25,
      "learning_rate": 4.810524506087335e-05,
      "loss": 0.9897,
      "num_input_tokens_seen": 2768416,
      "step": 4990
    },
    {
      "epoch": 87.63716814159292,
      "grad_norm": 8.4375,
      "learning_rate": 4.810149413106044e-05,
      "loss": 1.0935,
      "num_input_tokens_seen": 2771568,
      "step": 4995
    },
    {
      "epoch": 87.72566371681415,
      "grad_norm": 8.75,
      "learning_rate": 4.809773963870684e-05,
      "loss": 0.9433,
      "num_input_tokens_seen": 2774160,
      "step": 5000
    },
    {
      "epoch": 87.72566371681415,
      "eval_loss": 1.038374423980713,
      "eval_runtime": 0.9345,
      "eval_samples_per_second": 26.751,
      "eval_steps_per_second": 13.911,
      "num_input_tokens_seen": 2774160,
      "step": 5000
    },
    {
      "epoch": 87.8141592920354,
      "grad_norm": 8.5625,
      "learning_rate": 4.809398158439151e-05,
      "loss": 0.9934,
      "num_input_tokens_seen": 2776768,
      "step": 5005
    },
    {
      "epoch": 87.90265486725664,
      "grad_norm": 7.5625,
      "learning_rate": 4.8090219968694005e-05,
      "loss": 0.9934,
      "num_input_tokens_seen": 2779920,
      "step": 5010
    },
    {
      "epoch": 87.99115044247787,
      "grad_norm": 9.625,
      "learning_rate": 4.808645479219442e-05,
      "loss": 0.9775,
      "num_input_tokens_seen": 2782480,
      "step": 5015
    },
    {
      "epoch": 88.070796460177,
      "grad_norm": 13.1875,
      "learning_rate": 4.8082686055473375e-05,
      "loss": 1.2003,
      "num_input_tokens_seen": 2784608,
      "step": 5020
    },
    {
      "epoch": 88.15929203539822,
      "grad_norm": 9.125,
      "learning_rate": 4.8078913759112066e-05,
      "loss": 0.8818,
      "num_input_tokens_seen": 2787200,
      "step": 5025
    },
    {
      "epoch": 88.24778761061947,
      "grad_norm": 10.8125,
      "learning_rate": 4.807513790369223e-05,
      "loss": 1.0892,
      "num_input_tokens_seen": 2790224,
      "step": 5030
    },
    {
      "epoch": 88.33628318584071,
      "grad_norm": 7.15625,
      "learning_rate": 4.8071358489796145e-05,
      "loss": 1.0931,
      "num_input_tokens_seen": 2793056,
      "step": 5035
    },
    {
      "epoch": 88.42477876106194,
      "grad_norm": 7.09375,
      "learning_rate": 4.806757551800665e-05,
      "loss": 1.1039,
      "num_input_tokens_seen": 2795824,
      "step": 5040
    },
    {
      "epoch": 88.51327433628319,
      "grad_norm": 11.875,
      "learning_rate": 4.806378898890713e-05,
      "loss": 0.9459,
      "num_input_tokens_seen": 2798256,
      "step": 5045
    },
    {
      "epoch": 88.60176991150442,
      "grad_norm": 6.90625,
      "learning_rate": 4.80599989030815e-05,
      "loss": 0.913,
      "num_input_tokens_seen": 2801456,
      "step": 5050
    },
    {
      "epoch": 88.69026548672566,
      "grad_norm": 8.375,
      "learning_rate": 4.805620526111426e-05,
      "loss": 0.9863,
      "num_input_tokens_seen": 2804224,
      "step": 5055
    },
    {
      "epoch": 88.77876106194691,
      "grad_norm": 7.875,
      "learning_rate": 4.805240806359042e-05,
      "loss": 1.015,
      "num_input_tokens_seen": 2807312,
      "step": 5060
    },
    {
      "epoch": 88.86725663716814,
      "grad_norm": 7.84375,
      "learning_rate": 4.804860731109557e-05,
      "loss": 1.2167,
      "num_input_tokens_seen": 2810048,
      "step": 5065
    },
    {
      "epoch": 88.95575221238938,
      "grad_norm": 10.3125,
      "learning_rate": 4.804480300421581e-05,
      "loss": 1.2164,
      "num_input_tokens_seen": 2813040,
      "step": 5070
    },
    {
      "epoch": 89.03539823008849,
      "grad_norm": 9.5625,
      "learning_rate": 4.804099514353784e-05,
      "loss": 0.9727,
      "num_input_tokens_seen": 2815272,
      "step": 5075
    },
    {
      "epoch": 89.12389380530973,
      "grad_norm": 5.84375,
      "learning_rate": 4.8037183729648867e-05,
      "loss": 0.8596,
      "num_input_tokens_seen": 2818072,
      "step": 5080
    },
    {
      "epoch": 89.21238938053098,
      "grad_norm": 11.0625,
      "learning_rate": 4.803336876313666e-05,
      "loss": 1.2128,
      "num_input_tokens_seen": 2820904,
      "step": 5085
    },
    {
      "epoch": 89.30088495575221,
      "grad_norm": 6.0625,
      "learning_rate": 4.802955024458953e-05,
      "loss": 1.2212,
      "num_input_tokens_seen": 2823576,
      "step": 5090
    },
    {
      "epoch": 89.38938053097345,
      "grad_norm": 13.3125,
      "learning_rate": 4.802572817459634e-05,
      "loss": 1.0162,
      "num_input_tokens_seen": 2826248,
      "step": 5095
    },
    {
      "epoch": 89.47787610619469,
      "grad_norm": 8.0625,
      "learning_rate": 4.802190255374651e-05,
      "loss": 1.0655,
      "num_input_tokens_seen": 2829224,
      "step": 5100
    },
    {
      "epoch": 89.56637168141593,
      "grad_norm": 9.0625,
      "learning_rate": 4.801807338263e-05,
      "loss": 1.0996,
      "num_input_tokens_seen": 2831608,
      "step": 5105
    },
    {
      "epoch": 89.65486725663717,
      "grad_norm": 12.0625,
      "learning_rate": 4.8014240661837306e-05,
      "loss": 1.0135,
      "num_input_tokens_seen": 2834568,
      "step": 5110
    },
    {
      "epoch": 89.7433628318584,
      "grad_norm": 7.3125,
      "learning_rate": 4.80104043919595e-05,
      "loss": 0.9564,
      "num_input_tokens_seen": 2837832,
      "step": 5115
    },
    {
      "epoch": 89.83185840707965,
      "grad_norm": 7.9375,
      "learning_rate": 4.800656457358815e-05,
      "loss": 1.0163,
      "num_input_tokens_seen": 2840984,
      "step": 5120
    },
    {
      "epoch": 89.92035398230088,
      "grad_norm": 8.25,
      "learning_rate": 4.800272120731544e-05,
      "loss": 1.1182,
      "num_input_tokens_seen": 2843608,
      "step": 5125
    },
    {
      "epoch": 90.0,
      "grad_norm": 8.75,
      "learning_rate": 4.799887429373404e-05,
      "loss": 0.9897,
      "num_input_tokens_seen": 2846040,
      "step": 5130
    },
    {
      "epoch": 90.08849557522124,
      "grad_norm": 12.9375,
      "learning_rate": 4.79950238334372e-05,
      "loss": 1.1596,
      "num_input_tokens_seen": 2848552,
      "step": 5135
    },
    {
      "epoch": 90.17699115044248,
      "grad_norm": 9.4375,
      "learning_rate": 4.799116982701872e-05,
      "loss": 1.0602,
      "num_input_tokens_seen": 2852024,
      "step": 5140
    },
    {
      "epoch": 90.26548672566372,
      "grad_norm": 9.875,
      "learning_rate": 4.7987312275072926e-05,
      "loss": 1.1652,
      "num_input_tokens_seen": 2854568,
      "step": 5145
    },
    {
      "epoch": 90.35398230088495,
      "grad_norm": 7.71875,
      "learning_rate": 4.79834511781947e-05,
      "loss": 0.962,
      "num_input_tokens_seen": 2857560,
      "step": 5150
    },
    {
      "epoch": 90.4424778761062,
      "grad_norm": 8.5625,
      "learning_rate": 4.797958653697947e-05,
      "loss": 0.968,
      "num_input_tokens_seen": 2860696,
      "step": 5155
    },
    {
      "epoch": 90.53097345132744,
      "grad_norm": 8.3125,
      "learning_rate": 4.7975718352023225e-05,
      "loss": 1.2909,
      "num_input_tokens_seen": 2863544,
      "step": 5160
    },
    {
      "epoch": 90.61946902654867,
      "grad_norm": 11.0625,
      "learning_rate": 4.7971846623922476e-05,
      "loss": 0.935,
      "num_input_tokens_seen": 2866168,
      "step": 5165
    },
    {
      "epoch": 90.70796460176992,
      "grad_norm": 7.1875,
      "learning_rate": 4.7967971353274294e-05,
      "loss": 1.1473,
      "num_input_tokens_seen": 2868728,
      "step": 5170
    },
    {
      "epoch": 90.79646017699115,
      "grad_norm": 7.28125,
      "learning_rate": 4.79640925406763e-05,
      "loss": 1.0378,
      "num_input_tokens_seen": 2871608,
      "step": 5175
    },
    {
      "epoch": 90.88495575221239,
      "grad_norm": 7.40625,
      "learning_rate": 4.796021018672664e-05,
      "loss": 1.0036,
      "num_input_tokens_seen": 2873944,
      "step": 5180
    },
    {
      "epoch": 90.97345132743362,
      "grad_norm": 5.90625,
      "learning_rate": 4.795632429202405e-05,
      "loss": 0.902,
      "num_input_tokens_seen": 2876968,
      "step": 5185
    },
    {
      "epoch": 91.05309734513274,
      "grad_norm": 7.21875,
      "learning_rate": 4.795243485716775e-05,
      "loss": 0.8835,
      "num_input_tokens_seen": 2879416,
      "step": 5190
    },
    {
      "epoch": 91.14159292035399,
      "grad_norm": 9.625,
      "learning_rate": 4.794854188275757e-05,
      "loss": 1.1886,
      "num_input_tokens_seen": 2882760,
      "step": 5195
    },
    {
      "epoch": 91.23008849557522,
      "grad_norm": 7.5,
      "learning_rate": 4.794464536939384e-05,
      "loss": 1.1497,
      "num_input_tokens_seen": 2885448,
      "step": 5200
    },
    {
      "epoch": 91.23008849557522,
      "eval_loss": 1.0298749208450317,
      "eval_runtime": 0.9378,
      "eval_samples_per_second": 26.657,
      "eval_steps_per_second": 13.862,
      "num_input_tokens_seen": 2885448,
      "step": 5200
    },
    {
      "epoch": 91.31858407079646,
      "grad_norm": 11.9375,
      "learning_rate": 4.794074531767745e-05,
      "loss": 1.1789,
      "num_input_tokens_seen": 2888040,
      "step": 5205
    },
    {
      "epoch": 91.40707964601769,
      "grad_norm": 9.875,
      "learning_rate": 4.7936841728209834e-05,
      "loss": 1.0798,
      "num_input_tokens_seen": 2891192,
      "step": 5210
    },
    {
      "epoch": 91.49557522123894,
      "grad_norm": 6.25,
      "learning_rate": 4.7932934601593e-05,
      "loss": 0.9841,
      "num_input_tokens_seen": 2893928,
      "step": 5215
    },
    {
      "epoch": 91.58407079646018,
      "grad_norm": 7.28125,
      "learning_rate": 4.792902393842943e-05,
      "loss": 0.9159,
      "num_input_tokens_seen": 2896552,
      "step": 5220
    },
    {
      "epoch": 91.67256637168141,
      "grad_norm": 10.8125,
      "learning_rate": 4.792510973932225e-05,
      "loss": 0.8343,
      "num_input_tokens_seen": 2899208,
      "step": 5225
    },
    {
      "epoch": 91.76106194690266,
      "grad_norm": 8.0625,
      "learning_rate": 4.7921192004875036e-05,
      "loss": 1.1917,
      "num_input_tokens_seen": 2901768,
      "step": 5230
    },
    {
      "epoch": 91.84955752212389,
      "grad_norm": 8.9375,
      "learning_rate": 4.791727073569198e-05,
      "loss": 1.1622,
      "num_input_tokens_seen": 2904312,
      "step": 5235
    },
    {
      "epoch": 91.93805309734513,
      "grad_norm": 8.5625,
      "learning_rate": 4.7913345932377775e-05,
      "loss": 1.0301,
      "num_input_tokens_seen": 2907288,
      "step": 5240
    },
    {
      "epoch": 92.01769911504425,
      "grad_norm": 8.4375,
      "learning_rate": 4.790941759553769e-05,
      "loss": 1.0976,
      "num_input_tokens_seen": 2909848,
      "step": 5245
    },
    {
      "epoch": 92.10619469026548,
      "grad_norm": 7.78125,
      "learning_rate": 4.79054857257775e-05,
      "loss": 1.1615,
      "num_input_tokens_seen": 2912392,
      "step": 5250
    },
    {
      "epoch": 92.19469026548673,
      "grad_norm": 7.78125,
      "learning_rate": 4.790155032370357e-05,
      "loss": 0.9685,
      "num_input_tokens_seen": 2915224,
      "step": 5255
    },
    {
      "epoch": 92.28318584070796,
      "grad_norm": 8.3125,
      "learning_rate": 4.789761138992278e-05,
      "loss": 0.8849,
      "num_input_tokens_seen": 2917864,
      "step": 5260
    },
    {
      "epoch": 92.3716814159292,
      "grad_norm": 8.5625,
      "learning_rate": 4.7893668925042565e-05,
      "loss": 1.0897,
      "num_input_tokens_seen": 2920792,
      "step": 5265
    },
    {
      "epoch": 92.46017699115045,
      "grad_norm": 6.90625,
      "learning_rate": 4.78897229296709e-05,
      "loss": 1.046,
      "num_input_tokens_seen": 2923544,
      "step": 5270
    },
    {
      "epoch": 92.54867256637168,
      "grad_norm": 11.0,
      "learning_rate": 4.7885773404416315e-05,
      "loss": 1.0624,
      "num_input_tokens_seen": 2926312,
      "step": 5275
    },
    {
      "epoch": 92.63716814159292,
      "grad_norm": 9.375,
      "learning_rate": 4.788182034988786e-05,
      "loss": 1.1537,
      "num_input_tokens_seen": 2928792,
      "step": 5280
    },
    {
      "epoch": 92.72566371681415,
      "grad_norm": 11.375,
      "learning_rate": 4.787786376669516e-05,
      "loss": 1.1181,
      "num_input_tokens_seen": 2931384,
      "step": 5285
    },
    {
      "epoch": 92.8141592920354,
      "grad_norm": 7.5625,
      "learning_rate": 4.787390365544837e-05,
      "loss": 1.0462,
      "num_input_tokens_seen": 2934600,
      "step": 5290
    },
    {
      "epoch": 92.90265486725664,
      "grad_norm": 7.3125,
      "learning_rate": 4.786994001675818e-05,
      "loss": 1.0243,
      "num_input_tokens_seen": 2937720,
      "step": 5295
    },
    {
      "epoch": 92.99115044247787,
      "grad_norm": 8.375,
      "learning_rate": 4.786597285123584e-05,
      "loss": 0.9944,
      "num_input_tokens_seen": 2940248,
      "step": 5300
    },
    {
      "epoch": 93.070796460177,
      "grad_norm": 9.0,
      "learning_rate": 4.7862002159493135e-05,
      "loss": 1.0674,
      "num_input_tokens_seen": 2942808,
      "step": 5305
    },
    {
      "epoch": 93.15929203539822,
      "grad_norm": 8.8125,
      "learning_rate": 4.785802794214239e-05,
      "loss": 1.1913,
      "num_input_tokens_seen": 2945544,
      "step": 5310
    },
    {
      "epoch": 93.24778761061947,
      "grad_norm": 5.34375,
      "learning_rate": 4.7854050199796495e-05,
      "loss": 1.0275,
      "num_input_tokens_seen": 2948344,
      "step": 5315
    },
    {
      "epoch": 93.33628318584071,
      "grad_norm": 6.8125,
      "learning_rate": 4.7850068933068845e-05,
      "loss": 1.1048,
      "num_input_tokens_seen": 2951048,
      "step": 5320
    },
    {
      "epoch": 93.42477876106194,
      "grad_norm": 7.28125,
      "learning_rate": 4.7846084142573425e-05,
      "loss": 1.132,
      "num_input_tokens_seen": 2953656,
      "step": 5325
    },
    {
      "epoch": 93.51327433628319,
      "grad_norm": 6.6875,
      "learning_rate": 4.7842095828924725e-05,
      "loss": 1.0355,
      "num_input_tokens_seen": 2956584,
      "step": 5330
    },
    {
      "epoch": 93.60176991150442,
      "grad_norm": 7.28125,
      "learning_rate": 4.783810399273779e-05,
      "loss": 0.9561,
      "num_input_tokens_seen": 2959528,
      "step": 5335
    },
    {
      "epoch": 93.69026548672566,
      "grad_norm": 7.96875,
      "learning_rate": 4.7834108634628226e-05,
      "loss": 1.3051,
      "num_input_tokens_seen": 2962184,
      "step": 5340
    },
    {
      "epoch": 93.77876106194691,
      "grad_norm": 7.4375,
      "learning_rate": 4.783010975521216e-05,
      "loss": 1.1078,
      "num_input_tokens_seen": 2965032,
      "step": 5345
    },
    {
      "epoch": 93.86725663716814,
      "grad_norm": 6.8125,
      "learning_rate": 4.782610735510626e-05,
      "loss": 0.7892,
      "num_input_tokens_seen": 2967352,
      "step": 5350
    },
    {
      "epoch": 93.95575221238938,
      "grad_norm": 8.875,
      "learning_rate": 4.782210143492776e-05,
      "loss": 1.0503,
      "num_input_tokens_seen": 2970104,
      "step": 5355
    },
    {
      "epoch": 94.03539823008849,
      "grad_norm": 8.6875,
      "learning_rate": 4.781809199529442e-05,
      "loss": 0.8238,
      "num_input_tokens_seen": 2972608,
      "step": 5360
    },
    {
      "epoch": 94.12389380530973,
      "grad_norm": 8.25,
      "learning_rate": 4.781407903682454e-05,
      "loss": 1.1627,
      "num_input_tokens_seen": 2975392,
      "step": 5365
    },
    {
      "epoch": 94.21238938053098,
      "grad_norm": 9.4375,
      "learning_rate": 4.781006256013698e-05,
      "loss": 1.0542,
      "num_input_tokens_seen": 2978240,
      "step": 5370
    },
    {
      "epoch": 94.30088495575221,
      "grad_norm": 8.3125,
      "learning_rate": 4.7806042565851115e-05,
      "loss": 0.8805,
      "num_input_tokens_seen": 2980912,
      "step": 5375
    },
    {
      "epoch": 94.38938053097345,
      "grad_norm": 8.375,
      "learning_rate": 4.7802019054586895e-05,
      "loss": 1.1609,
      "num_input_tokens_seen": 2983920,
      "step": 5380
    },
    {
      "epoch": 94.47787610619469,
      "grad_norm": 11.125,
      "learning_rate": 4.779799202696479e-05,
      "loss": 1.1192,
      "num_input_tokens_seen": 2986992,
      "step": 5385
    },
    {
      "epoch": 94.56637168141593,
      "grad_norm": 6.625,
      "learning_rate": 4.779396148360581e-05,
      "loss": 0.9493,
      "num_input_tokens_seen": 2990128,
      "step": 5390
    },
    {
      "epoch": 94.65486725663717,
      "grad_norm": 7.9375,
      "learning_rate": 4.7789927425131517e-05,
      "loss": 1.2147,
      "num_input_tokens_seen": 2992832,
      "step": 5395
    },
    {
      "epoch": 94.7433628318584,
      "grad_norm": 9.0,
      "learning_rate": 4.778588985216403e-05,
      "loss": 1.0695,
      "num_input_tokens_seen": 2995680,
      "step": 5400
    },
    {
      "epoch": 94.7433628318584,
      "eval_loss": 1.0259016752243042,
      "eval_runtime": 0.9332,
      "eval_samples_per_second": 26.789,
      "eval_steps_per_second": 13.93,
      "num_input_tokens_seen": 2995680,
      "step": 5400
    },
    {
      "epoch": 94.83185840707965,
      "grad_norm": 9.4375,
      "learning_rate": 4.778184876532598e-05,
      "loss": 1.0506,
      "num_input_tokens_seen": 2998208,
      "step": 5405
    },
    {
      "epoch": 94.92035398230088,
      "grad_norm": 6.5,
      "learning_rate": 4.7777804165240556e-05,
      "loss": 0.9845,
      "num_input_tokens_seen": 3000864,
      "step": 5410
    },
    {
      "epoch": 95.0,
      "grad_norm": 8.1875,
      "learning_rate": 4.7773756052531485e-05,
      "loss": 1.0536,
      "num_input_tokens_seen": 3003272,
      "step": 5415
    },
    {
      "epoch": 95.08849557522124,
      "grad_norm": 6.875,
      "learning_rate": 4.7769704427823035e-05,
      "loss": 1.0529,
      "num_input_tokens_seen": 3005896,
      "step": 5420
    },
    {
      "epoch": 95.17699115044248,
      "grad_norm": 11.1875,
      "learning_rate": 4.776564929174003e-05,
      "loss": 1.2617,
      "num_input_tokens_seen": 3008664,
      "step": 5425
    },
    {
      "epoch": 95.26548672566372,
      "grad_norm": 7.25,
      "learning_rate": 4.7761590644907806e-05,
      "loss": 0.8811,
      "num_input_tokens_seen": 3012024,
      "step": 5430
    },
    {
      "epoch": 95.35398230088495,
      "grad_norm": 9.125,
      "learning_rate": 4.7757528487952263e-05,
      "loss": 0.7775,
      "num_input_tokens_seen": 3015096,
      "step": 5435
    },
    {
      "epoch": 95.4424778761062,
      "grad_norm": 9.3125,
      "learning_rate": 4.7753462821499836e-05,
      "loss": 1.1538,
      "num_input_tokens_seen": 3017672,
      "step": 5440
    },
    {
      "epoch": 95.53097345132744,
      "grad_norm": 7.15625,
      "learning_rate": 4.774939364617751e-05,
      "loss": 0.8952,
      "num_input_tokens_seen": 3020488,
      "step": 5445
    },
    {
      "epoch": 95.61946902654867,
      "grad_norm": 7.71875,
      "learning_rate": 4.7745320962612795e-05,
      "loss": 1.1756,
      "num_input_tokens_seen": 3023432,
      "step": 5450
    },
    {
      "epoch": 95.70796460176992,
      "grad_norm": 11.375,
      "learning_rate": 4.7741244771433756e-05,
      "loss": 1.0846,
      "num_input_tokens_seen": 3025992,
      "step": 5455
    },
    {
      "epoch": 95.79646017699115,
      "grad_norm": 8.4375,
      "learning_rate": 4.7737165073268985e-05,
      "loss": 1.1312,
      "num_input_tokens_seen": 3028904,
      "step": 5460
    },
    {
      "epoch": 95.88495575221239,
      "grad_norm": 7.65625,
      "learning_rate": 4.7733081868747626e-05,
      "loss": 0.9929,
      "num_input_tokens_seen": 3031928,
      "step": 5465
    },
    {
      "epoch": 95.97345132743362,
      "grad_norm": 7.96875,
      "learning_rate": 4.772899515849936e-05,
      "loss": 1.1495,
      "num_input_tokens_seen": 3034504,
      "step": 5470
    },
    {
      "epoch": 96.05309734513274,
      "grad_norm": 8.8125,
      "learning_rate": 4.7724904943154414e-05,
      "loss": 1.0981,
      "num_input_tokens_seen": 3036680,
      "step": 5475
    },
    {
      "epoch": 96.14159292035399,
      "grad_norm": 11.1875,
      "learning_rate": 4.772081122334354e-05,
      "loss": 1.0595,
      "num_input_tokens_seen": 3039416,
      "step": 5480
    },
    {
      "epoch": 96.23008849557522,
      "grad_norm": 6.0625,
      "learning_rate": 4.771671399969806e-05,
      "loss": 0.865,
      "num_input_tokens_seen": 3042616,
      "step": 5485
    },
    {
      "epoch": 96.31858407079646,
      "grad_norm": 12.25,
      "learning_rate": 4.7712613272849794e-05,
      "loss": 1.1527,
      "num_input_tokens_seen": 3045416,
      "step": 5490
    },
    {
      "epoch": 96.40707964601769,
      "grad_norm": 6.90625,
      "learning_rate": 4.770850904343114e-05,
      "loss": 1.0275,
      "num_input_tokens_seen": 3048024,
      "step": 5495
    },
    {
      "epoch": 96.49557522123894,
      "grad_norm": 10.0625,
      "learning_rate": 4.770440131207502e-05,
      "loss": 1.0899,
      "num_input_tokens_seen": 3050616,
      "step": 5500
    },
    {
      "epoch": 96.58407079646018,
      "grad_norm": 8.5,
      "learning_rate": 4.7700290079414896e-05,
      "loss": 1.1428,
      "num_input_tokens_seen": 3053688,
      "step": 5505
    },
    {
      "epoch": 96.67256637168141,
      "grad_norm": 8.75,
      "learning_rate": 4.769617534608477e-05,
      "loss": 1.0128,
      "num_input_tokens_seen": 3056536,
      "step": 5510
    },
    {
      "epoch": 96.76106194690266,
      "grad_norm": 7.6875,
      "learning_rate": 4.7692057112719193e-05,
      "loss": 0.9689,
      "num_input_tokens_seen": 3059352,
      "step": 5515
    },
    {
      "epoch": 96.84955752212389,
      "grad_norm": 6.59375,
      "learning_rate": 4.7687935379953234e-05,
      "loss": 0.9795,
      "num_input_tokens_seen": 3062040,
      "step": 5520
    },
    {
      "epoch": 96.93805309734513,
      "grad_norm": 8.6875,
      "learning_rate": 4.7683810148422534e-05,
      "loss": 1.1524,
      "num_input_tokens_seen": 3064920,
      "step": 5525
    },
    {
      "epoch": 97.01769911504425,
      "grad_norm": 6.875,
      "learning_rate": 4.767968141876324e-05,
      "loss": 0.9622,
      "num_input_tokens_seen": 3067568,
      "step": 5530
    },
    {
      "epoch": 97.10619469026548,
      "grad_norm": 5.96875,
      "learning_rate": 4.767554919161207e-05,
      "loss": 0.8749,
      "num_input_tokens_seen": 3070304,
      "step": 5535
    },
    {
      "epoch": 97.19469026548673,
      "grad_norm": 8.9375,
      "learning_rate": 4.767141346760624e-05,
      "loss": 1.2704,
      "num_input_tokens_seen": 3072944,
      "step": 5540
    },
    {
      "epoch": 97.28318584070796,
      "grad_norm": 6.40625,
      "learning_rate": 4.766727424738356e-05,
      "loss": 1.01,
      "num_input_tokens_seen": 3076080,
      "step": 5545
    },
    {
      "epoch": 97.3716814159292,
      "grad_norm": 8.6875,
      "learning_rate": 4.7663131531582325e-05,
      "loss": 1.1337,
      "num_input_tokens_seen": 3079024,
      "step": 5550
    },
    {
      "epoch": 97.46017699115045,
      "grad_norm": 11.25,
      "learning_rate": 4.765898532084142e-05,
      "loss": 1.0415,
      "num_input_tokens_seen": 3081744,
      "step": 5555
    },
    {
      "epoch": 97.54867256637168,
      "grad_norm": 10.5625,
      "learning_rate": 4.765483561580022e-05,
      "loss": 1.1127,
      "num_input_tokens_seen": 3084608,
      "step": 5560
    },
    {
      "epoch": 97.63716814159292,
      "grad_norm": 7.96875,
      "learning_rate": 4.7650682417098666e-05,
      "loss": 0.9418,
      "num_input_tokens_seen": 3087536,
      "step": 5565
    },
    {
      "epoch": 97.72566371681415,
      "grad_norm": 7.5,
      "learning_rate": 4.7646525725377244e-05,
      "loss": 1.0537,
      "num_input_tokens_seen": 3090000,
      "step": 5570
    },
    {
      "epoch": 97.8141592920354,
      "grad_norm": 11.4375,
      "learning_rate": 4.764236554127696e-05,
      "loss": 1.0426,
      "num_input_tokens_seen": 3092992,
      "step": 5575
    },
    {
      "epoch": 97.90265486725664,
      "grad_norm": 7.03125,
      "learning_rate": 4.7638201865439356e-05,
      "loss": 1.1653,
      "num_input_tokens_seen": 3095648,
      "step": 5580
    },
    {
      "epoch": 97.99115044247787,
      "grad_norm": 7.1875,
      "learning_rate": 4.7634034698506545e-05,
      "loss": 1.1255,
      "num_input_tokens_seen": 3098896,
      "step": 5585
    },
    {
      "epoch": 98.070796460177,
      "grad_norm": 9.125,
      "learning_rate": 4.762986404112115e-05,
      "loss": 0.8592,
      "num_input_tokens_seen": 3101392,
      "step": 5590
    },
    {
      "epoch": 98.15929203539822,
      "grad_norm": 10.625,
      "learning_rate": 4.762568989392633e-05,
      "loss": 0.9776,
      "num_input_tokens_seen": 3104384,
      "step": 5595
    },
    {
      "epoch": 98.24778761061947,
      "grad_norm": 12.125,
      "learning_rate": 4.76215122575658e-05,
      "loss": 1.1854,
      "num_input_tokens_seen": 3106768,
      "step": 5600
    },
    {
      "epoch": 98.24778761061947,
      "eval_loss": 1.019257664680481,
      "eval_runtime": 0.9332,
      "eval_samples_per_second": 26.789,
      "eval_steps_per_second": 13.93,
      "num_input_tokens_seen": 3106768,
      "step": 5600
    },
    {
      "epoch": 98.33628318584071,
      "grad_norm": 9.1875,
      "learning_rate": 4.7617331132683795e-05,
      "loss": 1.1102,
      "num_input_tokens_seen": 3109616,
      "step": 5605
    },
    {
      "epoch": 98.42477876106194,
      "grad_norm": 10.875,
      "learning_rate": 4.7613146519925105e-05,
      "loss": 0.9172,
      "num_input_tokens_seen": 3112832,
      "step": 5610
    },
    {
      "epoch": 98.51327433628319,
      "grad_norm": 9.625,
      "learning_rate": 4.7608958419935045e-05,
      "loss": 1.0699,
      "num_input_tokens_seen": 3115440,
      "step": 5615
    },
    {
      "epoch": 98.60176991150442,
      "grad_norm": 7.75,
      "learning_rate": 4.760476683335948e-05,
      "loss": 0.8182,
      "num_input_tokens_seen": 3118528,
      "step": 5620
    },
    {
      "epoch": 98.69026548672566,
      "grad_norm": 7.21875,
      "learning_rate": 4.760057176084479e-05,
      "loss": 1.0637,
      "num_input_tokens_seen": 3120960,
      "step": 5625
    },
    {
      "epoch": 98.77876106194691,
      "grad_norm": 9.3125,
      "learning_rate": 4.759637320303793e-05,
      "loss": 1.2406,
      "num_input_tokens_seen": 3123920,
      "step": 5630
    },
    {
      "epoch": 98.86725663716814,
      "grad_norm": 8.6875,
      "learning_rate": 4.759217116058635e-05,
      "loss": 1.0577,
      "num_input_tokens_seen": 3126688,
      "step": 5635
    },
    {
      "epoch": 98.95575221238938,
      "grad_norm": 14.625,
      "learning_rate": 4.758796563413807e-05,
      "loss": 1.1905,
      "num_input_tokens_seen": 3129232,
      "step": 5640
    },
    {
      "epoch": 99.03539823008849,
      "grad_norm": 7.125,
      "learning_rate": 4.758375662434163e-05,
      "loss": 0.9684,
      "num_input_tokens_seen": 3131408,
      "step": 5645
    },
    {
      "epoch": 99.12389380530973,
      "grad_norm": 7.71875,
      "learning_rate": 4.7579544131846114e-05,
      "loss": 1.0592,
      "num_input_tokens_seen": 3134448,
      "step": 5650
    },
    {
      "epoch": 99.21238938053098,
      "grad_norm": 7.59375,
      "learning_rate": 4.757532815730114e-05,
      "loss": 0.8947,
      "num_input_tokens_seen": 3137488,
      "step": 5655
    },
    {
      "epoch": 99.30088495575221,
      "grad_norm": 7.53125,
      "learning_rate": 4.7571108701356865e-05,
      "loss": 1.1357,
      "num_input_tokens_seen": 3140144,
      "step": 5660
    },
    {
      "epoch": 99.38938053097345,
      "grad_norm": 9.8125,
      "learning_rate": 4.756688576466398e-05,
      "loss": 1.1111,
      "num_input_tokens_seen": 3142928,
      "step": 5665
    },
    {
      "epoch": 99.47787610619469,
      "grad_norm": 7.4375,
      "learning_rate": 4.756265934787372e-05,
      "loss": 0.9855,
      "num_input_tokens_seen": 3145856,
      "step": 5670
    },
    {
      "epoch": 99.56637168141593,
      "grad_norm": 9.0625,
      "learning_rate": 4.755842945163785e-05,
      "loss": 1.2347,
      "num_input_tokens_seen": 3148432,
      "step": 5675
    },
    {
      "epoch": 99.65486725663717,
      "grad_norm": 8.125,
      "learning_rate": 4.755419607660867e-05,
      "loss": 1.0742,
      "num_input_tokens_seen": 3151552,
      "step": 5680
    },
    {
      "epoch": 99.7433628318584,
      "grad_norm": 6.875,
      "learning_rate": 4.7549959223439016e-05,
      "loss": 0.9215,
      "num_input_tokens_seen": 3153952,
      "step": 5685
    },
    {
      "epoch": 99.83185840707965,
      "grad_norm": 8.0625,
      "learning_rate": 4.754571889278228e-05,
      "loss": 1.0489,
      "num_input_tokens_seen": 3157008,
      "step": 5690
    },
    {
      "epoch": 99.92035398230088,
      "grad_norm": 7.78125,
      "learning_rate": 4.754147508529235e-05,
      "loss": 1.2397,
      "num_input_tokens_seen": 3159696,
      "step": 5695
    },
    {
      "epoch": 100.0,
      "grad_norm": 3.609375,
      "learning_rate": 4.75372278016237e-05,
      "loss": 0.8564,
      "num_input_tokens_seen": 3161808,
      "step": 5700
    },
    {
      "epoch": 100.08849557522124,
      "grad_norm": 7.59375,
      "learning_rate": 4.753297704243129e-05,
      "loss": 1.1367,
      "num_input_tokens_seen": 3164528,
      "step": 5705
    },
    {
      "epoch": 100.17699115044248,
      "grad_norm": 10.1875,
      "learning_rate": 4.752872280837066e-05,
      "loss": 1.1092,
      "num_input_tokens_seen": 3167248,
      "step": 5710
    },
    {
      "epoch": 100.26548672566372,
      "grad_norm": 7.40625,
      "learning_rate": 4.752446510009786e-05,
      "loss": 1.0621,
      "num_input_tokens_seen": 3170336,
      "step": 5715
    },
    {
      "epoch": 100.35398230088495,
      "grad_norm": 7.9375,
      "learning_rate": 4.7520203918269476e-05,
      "loss": 0.9918,
      "num_input_tokens_seen": 3172768,
      "step": 5720
    },
    {
      "epoch": 100.4424778761062,
      "grad_norm": 9.125,
      "learning_rate": 4.751593926354265e-05,
      "loss": 1.0159,
      "num_input_tokens_seen": 3175472,
      "step": 5725
    },
    {
      "epoch": 100.53097345132744,
      "grad_norm": 10.4375,
      "learning_rate": 4.751167113657503e-05,
      "loss": 0.9545,
      "num_input_tokens_seen": 3178224,
      "step": 5730
    },
    {
      "epoch": 100.61946902654867,
      "grad_norm": 7.71875,
      "learning_rate": 4.7507399538024834e-05,
      "loss": 1.1403,
      "num_input_tokens_seen": 3180944,
      "step": 5735
    },
    {
      "epoch": 100.70796460176992,
      "grad_norm": 6.6875,
      "learning_rate": 4.750312446855077e-05,
      "loss": 0.8629,
      "num_input_tokens_seen": 3184368,
      "step": 5740
    },
    {
      "epoch": 100.79646017699115,
      "grad_norm": 8.0,
      "learning_rate": 4.749884592881212e-05,
      "loss": 1.1983,
      "num_input_tokens_seen": 3187248,
      "step": 5745
    },
    {
      "epoch": 100.88495575221239,
      "grad_norm": 7.84375,
      "learning_rate": 4.74945639194687e-05,
      "loss": 1.0369,
      "num_input_tokens_seen": 3189872,
      "step": 5750
    },
    {
      "epoch": 100.97345132743362,
      "grad_norm": 7.90625,
      "learning_rate": 4.749027844118083e-05,
      "loss": 1.0201,
      "num_input_tokens_seen": 3193088,
      "step": 5755
    },
    {
      "epoch": 101.05309734513274,
      "grad_norm": 8.9375,
      "learning_rate": 4.7485989494609395e-05,
      "loss": 1.47,
      "num_input_tokens_seen": 3195352,
      "step": 5760
    },
    {
      "epoch": 101.14159292035399,
      "grad_norm": 10.375,
      "learning_rate": 4.748169708041581e-05,
      "loss": 1.1092,
      "num_input_tokens_seen": 3197848,
      "step": 5765
    },
    {
      "epoch": 101.23008849557522,
      "grad_norm": 7.125,
      "learning_rate": 4.7477401199262004e-05,
      "loss": 0.7453,
      "num_input_tokens_seen": 3200936,
      "step": 5770
    },
    {
      "epoch": 101.31858407079646,
      "grad_norm": 7.59375,
      "learning_rate": 4.747310185181048e-05,
      "loss": 1.189,
      "num_input_tokens_seen": 3203640,
      "step": 5775
    },
    {
      "epoch": 101.40707964601769,
      "grad_norm": 7.84375,
      "learning_rate": 4.746879903872422e-05,
      "loss": 0.9157,
      "num_input_tokens_seen": 3206376,
      "step": 5780
    },
    {
      "epoch": 101.49557522123894,
      "grad_norm": 7.15625,
      "learning_rate": 4.746449276066679e-05,
      "loss": 1.1142,
      "num_input_tokens_seen": 3209480,
      "step": 5785
    },
    {
      "epoch": 101.58407079646018,
      "grad_norm": 8.0625,
      "learning_rate": 4.746018301830227e-05,
      "loss": 1.0823,
      "num_input_tokens_seen": 3212776,
      "step": 5790
    },
    {
      "epoch": 101.67256637168141,
      "grad_norm": 7.875,
      "learning_rate": 4.7455869812295275e-05,
      "loss": 0.9929,
      "num_input_tokens_seen": 3215256,
      "step": 5795
    },
    {
      "epoch": 101.76106194690266,
      "grad_norm": 9.75,
      "learning_rate": 4.7451553143310964e-05,
      "loss": 1.1601,
      "num_input_tokens_seen": 3218248,
      "step": 5800
    },
    {
      "epoch": 101.76106194690266,
      "eval_loss": 1.0228207111358643,
      "eval_runtime": 0.935,
      "eval_samples_per_second": 26.737,
      "eval_steps_per_second": 13.903,
      "num_input_tokens_seen": 3218248,
      "step": 5800
    },
    {
      "epoch": 101.84955752212389,
      "grad_norm": 9.6875,
      "learning_rate": 4.744723301201501e-05,
      "loss": 1.1425,
      "num_input_tokens_seen": 3220952,
      "step": 5805
    },
    {
      "epoch": 101.93805309734513,
      "grad_norm": 8.25,
      "learning_rate": 4.744290941907364e-05,
      "loss": 1.05,
      "num_input_tokens_seen": 3223640,
      "step": 5810
    },
    {
      "epoch": 102.01769911504425,
      "grad_norm": 7.5,
      "learning_rate": 4.7438582365153594e-05,
      "loss": 1.0527,
      "num_input_tokens_seen": 3225920,
      "step": 5815
    },
    {
      "epoch": 102.10619469026548,
      "grad_norm": 7.84375,
      "learning_rate": 4.743425185092217e-05,
      "loss": 1.1461,
      "num_input_tokens_seen": 3228960,
      "step": 5820
    },
    {
      "epoch": 102.19469026548673,
      "grad_norm": 6.46875,
      "learning_rate": 4.742991787704719e-05,
      "loss": 0.9355,
      "num_input_tokens_seen": 3232336,
      "step": 5825
    },
    {
      "epoch": 102.28318584070796,
      "grad_norm": 9.75,
      "learning_rate": 4.7425580444196994e-05,
      "loss": 1.0628,
      "num_input_tokens_seen": 3235008,
      "step": 5830
    },
    {
      "epoch": 102.3716814159292,
      "grad_norm": 14.75,
      "learning_rate": 4.742123955304048e-05,
      "loss": 1.1339,
      "num_input_tokens_seen": 3237632,
      "step": 5835
    },
    {
      "epoch": 102.46017699115045,
      "grad_norm": 10.3125,
      "learning_rate": 4.741689520424706e-05,
      "loss": 0.9782,
      "num_input_tokens_seen": 3240448,
      "step": 5840
    },
    {
      "epoch": 102.54867256637168,
      "grad_norm": 9.1875,
      "learning_rate": 4.741254739848669e-05,
      "loss": 1.1324,
      "num_input_tokens_seen": 3242752,
      "step": 5845
    },
    {
      "epoch": 102.63716814159292,
      "grad_norm": 8.5,
      "learning_rate": 4.740819613642987e-05,
      "loss": 1.1729,
      "num_input_tokens_seen": 3245552,
      "step": 5850
    },
    {
      "epoch": 102.72566371681415,
      "grad_norm": 8.1875,
      "learning_rate": 4.74038414187476e-05,
      "loss": 1.0076,
      "num_input_tokens_seen": 3248080,
      "step": 5855
    },
    {
      "epoch": 102.8141592920354,
      "grad_norm": 10.75,
      "learning_rate": 4.739948324611144e-05,
      "loss": 1.0125,
      "num_input_tokens_seen": 3251200,
      "step": 5860
    },
    {
      "epoch": 102.90265486725664,
      "grad_norm": 8.75,
      "learning_rate": 4.7395121619193465e-05,
      "loss": 1.0011,
      "num_input_tokens_seen": 3254512,
      "step": 5865
    },
    {
      "epoch": 102.99115044247787,
      "grad_norm": 8.9375,
      "learning_rate": 4.7390756538666313e-05,
      "loss": 0.9989,
      "num_input_tokens_seen": 3257280,
      "step": 5870
    },
    {
      "epoch": 103.070796460177,
      "grad_norm": 9.6875,
      "learning_rate": 4.738638800520311e-05,
      "loss": 1.0581,
      "num_input_tokens_seen": 3259864,
      "step": 5875
    },
    {
      "epoch": 103.15929203539822,
      "grad_norm": 6.78125,
      "learning_rate": 4.738201601947757e-05,
      "loss": 0.895,
      "num_input_tokens_seen": 3262200,
      "step": 5880
    },
    {
      "epoch": 103.24778761061947,
      "grad_norm": 9.0625,
      "learning_rate": 4.7377640582163876e-05,
      "loss": 1.088,
      "num_input_tokens_seen": 3264712,
      "step": 5885
    },
    {
      "epoch": 103.33628318584071,
      "grad_norm": 9.5625,
      "learning_rate": 4.7373261693936786e-05,
      "loss": 1.0946,
      "num_input_tokens_seen": 3267592,
      "step": 5890
    },
    {
      "epoch": 103.42477876106194,
      "grad_norm": 6.96875,
      "learning_rate": 4.7368879355471595e-05,
      "loss": 0.9568,
      "num_input_tokens_seen": 3270088,
      "step": 5895
    },
    {
      "epoch": 103.51327433628319,
      "grad_norm": 9.25,
      "learning_rate": 4.736449356744409e-05,
      "loss": 1.1026,
      "num_input_tokens_seen": 3272632,
      "step": 5900
    },
    {
      "epoch": 103.60176991150442,
      "grad_norm": 7.0625,
      "learning_rate": 4.736010433053064e-05,
      "loss": 1.2395,
      "num_input_tokens_seen": 3275480,
      "step": 5905
    },
    {
      "epoch": 103.69026548672566,
      "grad_norm": 8.5,
      "learning_rate": 4.73557116454081e-05,
      "loss": 1.01,
      "num_input_tokens_seen": 3278008,
      "step": 5910
    },
    {
      "epoch": 103.77876106194691,
      "grad_norm": 11.25,
      "learning_rate": 4.735131551275389e-05,
      "loss": 1.1543,
      "num_input_tokens_seen": 3281048,
      "step": 5915
    },
    {
      "epoch": 103.86725663716814,
      "grad_norm": 9.0,
      "learning_rate": 4.734691593324594e-05,
      "loss": 1.0168,
      "num_input_tokens_seen": 3284664,
      "step": 5920
    },
    {
      "epoch": 103.95575221238938,
      "grad_norm": 9.0625,
      "learning_rate": 4.734251290756272e-05,
      "loss": 1.0296,
      "num_input_tokens_seen": 3287768,
      "step": 5925
    },
    {
      "epoch": 104.03539823008849,
      "grad_norm": 7.5625,
      "learning_rate": 4.7338106436383246e-05,
      "loss": 1.0519,
      "num_input_tokens_seen": 3290344,
      "step": 5930
    },
    {
      "epoch": 104.12389380530973,
      "grad_norm": 7.9375,
      "learning_rate": 4.733369652038703e-05,
      "loss": 0.9289,
      "num_input_tokens_seen": 3293224,
      "step": 5935
    },
    {
      "epoch": 104.21238938053098,
      "grad_norm": 12.0,
      "learning_rate": 4.7329283160254156e-05,
      "loss": 1.0997,
      "num_input_tokens_seen": 3296152,
      "step": 5940
    },
    {
      "epoch": 104.30088495575221,
      "grad_norm": 7.75,
      "learning_rate": 4.732486635666521e-05,
      "loss": 1.2407,
      "num_input_tokens_seen": 3299048,
      "step": 5945
    },
    {
      "epoch": 104.38938053097345,
      "grad_norm": 7.65625,
      "learning_rate": 4.732044611030132e-05,
      "loss": 0.9287,
      "num_input_tokens_seen": 3302184,
      "step": 5950
    },
    {
      "epoch": 104.47787610619469,
      "grad_norm": 7.46875,
      "learning_rate": 4.731602242184414e-05,
      "loss": 1.0728,
      "num_input_tokens_seen": 3304616,
      "step": 5955
    },
    {
      "epoch": 104.56637168141593,
      "grad_norm": 9.5,
      "learning_rate": 4.7311595291975864e-05,
      "loss": 1.138,
      "num_input_tokens_seen": 3307352,
      "step": 5960
    },
    {
      "epoch": 104.65486725663717,
      "grad_norm": 8.75,
      "learning_rate": 4.7307164721379216e-05,
      "loss": 1.2208,
      "num_input_tokens_seen": 3310344,
      "step": 5965
    },
    {
      "epoch": 104.7433628318584,
      "grad_norm": 6.96875,
      "learning_rate": 4.730273071073743e-05,
      "loss": 0.9729,
      "num_input_tokens_seen": 3313224,
      "step": 5970
    },
    {
      "epoch": 104.83185840707965,
      "grad_norm": 7.78125,
      "learning_rate": 4.729829326073429e-05,
      "loss": 1.0318,
      "num_input_tokens_seen": 3315768,
      "step": 5975
    },
    {
      "epoch": 104.92035398230088,
      "grad_norm": 7.1875,
      "learning_rate": 4.7293852372054126e-05,
      "loss": 0.8811,
      "num_input_tokens_seen": 3318376,
      "step": 5980
    },
    {
      "epoch": 105.0,
      "grad_norm": 7.3125,
      "learning_rate": 4.728940804538176e-05,
      "loss": 1.044,
      "num_input_tokens_seen": 3320728,
      "step": 5985
    },
    {
      "epoch": 105.08849557522124,
      "grad_norm": 7.21875,
      "learning_rate": 4.7284960281402556e-05,
      "loss": 0.979,
      "num_input_tokens_seen": 3323384,
      "step": 5990
    },
    {
      "epoch": 105.17699115044248,
      "grad_norm": 7.65625,
      "learning_rate": 4.728050908080244e-05,
      "loss": 1.0674,
      "num_input_tokens_seen": 3325912,
      "step": 5995
    },
    {
      "epoch": 105.26548672566372,
      "grad_norm": 8.9375,
      "learning_rate": 4.727605444426782e-05,
      "loss": 0.9279,
      "num_input_tokens_seen": 3329176,
      "step": 6000
    },
    {
      "epoch": 105.26548672566372,
      "eval_loss": 1.0300800800323486,
      "eval_runtime": 0.9362,
      "eval_samples_per_second": 26.703,
      "eval_steps_per_second": 13.885,
      "num_input_tokens_seen": 3329176,
      "step": 6000
    },
    {
      "epoch": 105.35398230088495,
      "grad_norm": 8.4375,
      "learning_rate": 4.727159637248567e-05,
      "loss": 1.1109,
      "num_input_tokens_seen": 3332152,
      "step": 6005
    },
    {
      "epoch": 105.4424778761062,
      "grad_norm": 6.90625,
      "learning_rate": 4.7267134866143474e-05,
      "loss": 0.9545,
      "num_input_tokens_seen": 3334968,
      "step": 6010
    },
    {
      "epoch": 105.53097345132744,
      "grad_norm": 7.75,
      "learning_rate": 4.726266992592926e-05,
      "loss": 1.1268,
      "num_input_tokens_seen": 3337816,
      "step": 6015
    },
    {
      "epoch": 105.61946902654867,
      "grad_norm": 9.5625,
      "learning_rate": 4.725820155253157e-05,
      "loss": 0.9525,
      "num_input_tokens_seen": 3340584,
      "step": 6020
    },
    {
      "epoch": 105.70796460176992,
      "grad_norm": 7.5,
      "learning_rate": 4.725372974663948e-05,
      "loss": 1.1398,
      "num_input_tokens_seen": 3343480,
      "step": 6025
    },
    {
      "epoch": 105.79646017699115,
      "grad_norm": 10.75,
      "learning_rate": 4.724925450894262e-05,
      "loss": 1.002,
      "num_input_tokens_seen": 3346440,
      "step": 6030
    },
    {
      "epoch": 105.88495575221239,
      "grad_norm": 9.3125,
      "learning_rate": 4.72447758401311e-05,
      "loss": 1.1274,
      "num_input_tokens_seen": 3349176,
      "step": 6035
    },
    {
      "epoch": 105.97345132743362,
      "grad_norm": 9.125,
      "learning_rate": 4.7240293740895616e-05,
      "loss": 1.1966,
      "num_input_tokens_seen": 3351528,
      "step": 6040
    },
    {
      "epoch": 106.05309734513274,
      "grad_norm": 8.4375,
      "learning_rate": 4.723580821192733e-05,
      "loss": 1.4509,
      "num_input_tokens_seen": 3353480,
      "step": 6045
    },
    {
      "epoch": 106.14159292035399,
      "grad_norm": 8.0625,
      "learning_rate": 4.7231319253917996e-05,
      "loss": 1.0211,
      "num_input_tokens_seen": 3356856,
      "step": 6050
    },
    {
      "epoch": 106.23008849557522,
      "grad_norm": 9.4375,
      "learning_rate": 4.722682686755986e-05,
      "loss": 1.0725,
      "num_input_tokens_seen": 3360120,
      "step": 6055
    },
    {
      "epoch": 106.31858407079646,
      "grad_norm": 8.3125,
      "learning_rate": 4.722233105354569e-05,
      "loss": 0.9553,
      "num_input_tokens_seen": 3362792,
      "step": 6060
    },
    {
      "epoch": 106.40707964601769,
      "grad_norm": 5.34375,
      "learning_rate": 4.7217831812568815e-05,
      "loss": 1.1979,
      "num_input_tokens_seen": 3365144,
      "step": 6065
    },
    {
      "epoch": 106.49557522123894,
      "grad_norm": 9.3125,
      "learning_rate": 4.721332914532307e-05,
      "loss": 1.2194,
      "num_input_tokens_seen": 3367768,
      "step": 6070
    },
    {
      "epoch": 106.58407079646018,
      "grad_norm": 6.875,
      "learning_rate": 4.720882305250281e-05,
      "loss": 0.9653,
      "num_input_tokens_seen": 3370888,
      "step": 6075
    },
    {
      "epoch": 106.67256637168141,
      "grad_norm": 6.34375,
      "learning_rate": 4.720431353480295e-05,
      "loss": 1.0189,
      "num_input_tokens_seen": 3373336,
      "step": 6080
    },
    {
      "epoch": 106.76106194690266,
      "grad_norm": 7.75,
      "learning_rate": 4.719980059291891e-05,
      "loss": 1.0233,
      "num_input_tokens_seen": 3375960,
      "step": 6085
    },
    {
      "epoch": 106.84955752212389,
      "grad_norm": 9.8125,
      "learning_rate": 4.7195284227546634e-05,
      "loss": 1.0522,
      "num_input_tokens_seen": 3378840,
      "step": 6090
    },
    {
      "epoch": 106.93805309734513,
      "grad_norm": 11.4375,
      "learning_rate": 4.7190764439382604e-05,
      "loss": 0.9941,
      "num_input_tokens_seen": 3381384,
      "step": 6095
    },
    {
      "epoch": 107.01769911504425,
      "grad_norm": 8.375,
      "learning_rate": 4.7186241229123826e-05,
      "loss": 1.0605,
      "num_input_tokens_seen": 3383784,
      "step": 6100
    },
    {
      "epoch": 107.10619469026548,
      "grad_norm": 7.78125,
      "learning_rate": 4.718171459746785e-05,
      "loss": 1.0333,
      "num_input_tokens_seen": 3386968,
      "step": 6105
    },
    {
      "epoch": 107.19469026548673,
      "grad_norm": 7.59375,
      "learning_rate": 4.717718454511273e-05,
      "loss": 1.1866,
      "num_input_tokens_seen": 3390088,
      "step": 6110
    },
    {
      "epoch": 107.28318584070796,
      "grad_norm": 9.25,
      "learning_rate": 4.7172651072757056e-05,
      "loss": 0.9134,
      "num_input_tokens_seen": 3392888,
      "step": 6115
    },
    {
      "epoch": 107.3716814159292,
      "grad_norm": 7.03125,
      "learning_rate": 4.7168114181099945e-05,
      "loss": 0.9621,
      "num_input_tokens_seen": 3395752,
      "step": 6120
    },
    {
      "epoch": 107.46017699115045,
      "grad_norm": 6.78125,
      "learning_rate": 4.716357387084105e-05,
      "loss": 1.0015,
      "num_input_tokens_seen": 3398776,
      "step": 6125
    },
    {
      "epoch": 107.54867256637168,
      "grad_norm": 8.4375,
      "learning_rate": 4.715903014268054e-05,
      "loss": 1.1458,
      "num_input_tokens_seen": 3401048,
      "step": 6130
    },
    {
      "epoch": 107.63716814159292,
      "grad_norm": 7.78125,
      "learning_rate": 4.715448299731911e-05,
      "loss": 0.9843,
      "num_input_tokens_seen": 3403976,
      "step": 6135
    },
    {
      "epoch": 107.72566371681415,
      "grad_norm": 7.59375,
      "learning_rate": 4.7149932435457986e-05,
      "loss": 1.1281,
      "num_input_tokens_seen": 3406840,
      "step": 6140
    },
    {
      "epoch": 107.8141592920354,
      "grad_norm": 7.75,
      "learning_rate": 4.714537845779894e-05,
      "loss": 0.9256,
      "num_input_tokens_seen": 3409928,
      "step": 6145
    },
    {
      "epoch": 107.90265486725664,
      "grad_norm": 9.125,
      "learning_rate": 4.714082106504423e-05,
      "loss": 1.1713,
      "num_input_tokens_seen": 3412440,
      "step": 6150
    },
    {
      "epoch": 107.99115044247787,
      "grad_norm": 9.4375,
      "learning_rate": 4.713626025789667e-05,
      "loss": 1.1941,
      "num_input_tokens_seen": 3415112,
      "step": 6155
    },
    {
      "epoch": 108.070796460177,
      "grad_norm": 8.875,
      "learning_rate": 4.7131696037059606e-05,
      "loss": 1.1144,
      "num_input_tokens_seen": 3417752,
      "step": 6160
    },
    {
      "epoch": 108.15929203539822,
      "grad_norm": 7.8125,
      "learning_rate": 4.712712840323689e-05,
      "loss": 1.0929,
      "num_input_tokens_seen": 3420472,
      "step": 6165
    },
    {
      "epoch": 108.24778761061947,
      "grad_norm": 8.0,
      "learning_rate": 4.71225573571329e-05,
      "loss": 0.8761,
      "num_input_tokens_seen": 3423016,
      "step": 6170
    },
    {
      "epoch": 108.33628318584071,
      "grad_norm": 8.25,
      "learning_rate": 4.711798289945256e-05,
      "loss": 1.1071,
      "num_input_tokens_seen": 3426088,
      "step": 6175
    },
    {
      "epoch": 108.42477876106194,
      "grad_norm": 11.875,
      "learning_rate": 4.71134050309013e-05,
      "loss": 1.1153,
      "num_input_tokens_seen": 3429112,
      "step": 6180
    },
    {
      "epoch": 108.51327433628319,
      "grad_norm": 11.375,
      "learning_rate": 4.710882375218509e-05,
      "loss": 0.9475,
      "num_input_tokens_seen": 3432040,
      "step": 6185
    },
    {
      "epoch": 108.60176991150442,
      "grad_norm": 9.4375,
      "learning_rate": 4.7104239064010424e-05,
      "loss": 0.9718,
      "num_input_tokens_seen": 3434984,
      "step": 6190
    },
    {
      "epoch": 108.69026548672566,
      "grad_norm": 8.25,
      "learning_rate": 4.709965096708432e-05,
      "loss": 0.9989,
      "num_input_tokens_seen": 3437624,
      "step": 6195
    },
    {
      "epoch": 108.77876106194691,
      "grad_norm": 8.4375,
      "learning_rate": 4.709505946211431e-05,
      "loss": 1.193,
      "num_input_tokens_seen": 3440344,
      "step": 6200
    },
    {
      "epoch": 108.77876106194691,
      "eval_loss": 1.031175971031189,
      "eval_runtime": 0.9383,
      "eval_samples_per_second": 26.644,
      "eval_steps_per_second": 13.855,
      "num_input_tokens_seen": 3440344,
      "step": 6200
    },
    {
      "epoch": 108.86725663716814,
      "grad_norm": 8.25,
      "learning_rate": 4.709046454980846e-05,
      "loss": 0.9957,
      "num_input_tokens_seen": 3443448,
      "step": 6205
    },
    {
      "epoch": 108.95575221238938,
      "grad_norm": 9.375,
      "learning_rate": 4.708586623087538e-05,
      "loss": 1.1367,
      "num_input_tokens_seen": 3445976,
      "step": 6210
    },
    {
      "epoch": 109.03539823008849,
      "grad_norm": 6.71875,
      "learning_rate": 4.708126450602418e-05,
      "loss": 1.0494,
      "num_input_tokens_seen": 3448136,
      "step": 6215
    },
    {
      "epoch": 109.12389380530973,
      "grad_norm": 8.75,
      "learning_rate": 4.7076659375964495e-05,
      "loss": 1.0218,
      "num_input_tokens_seen": 3451224,
      "step": 6220
    },
    {
      "epoch": 109.21238938053098,
      "grad_norm": 12.0625,
      "learning_rate": 4.707205084140651e-05,
      "loss": 1.1065,
      "num_input_tokens_seen": 3454072,
      "step": 6225
    },
    {
      "epoch": 109.30088495575221,
      "grad_norm": 7.4375,
      "learning_rate": 4.7067438903060904e-05,
      "loss": 1.2253,
      "num_input_tokens_seen": 3456680,
      "step": 6230
    },
    {
      "epoch": 109.38938053097345,
      "grad_norm": 6.8125,
      "learning_rate": 4.70628235616389e-05,
      "loss": 0.9484,
      "num_input_tokens_seen": 3459400,
      "step": 6235
    },
    {
      "epoch": 109.47787610619469,
      "grad_norm": 14.4375,
      "learning_rate": 4.7058204817852256e-05,
      "loss": 1.1939,
      "num_input_tokens_seen": 3461832,
      "step": 6240
    },
    {
      "epoch": 109.56637168141593,
      "grad_norm": 8.125,
      "learning_rate": 4.705358267241322e-05,
      "loss": 0.9745,
      "num_input_tokens_seen": 3464760,
      "step": 6245
    },
    {
      "epoch": 109.65486725663717,
      "grad_norm": 7.875,
      "learning_rate": 4.704895712603459e-05,
      "loss": 0.947,
      "num_input_tokens_seen": 3467960,
      "step": 6250
    },
    {
      "epoch": 109.7433628318584,
      "grad_norm": 8.0625,
      "learning_rate": 4.704432817942969e-05,
      "loss": 1.0811,
      "num_input_tokens_seen": 3471112,
      "step": 6255
    },
    {
      "epoch": 109.83185840707965,
      "grad_norm": 8.25,
      "learning_rate": 4.703969583331236e-05,
      "loss": 1.0756,
      "num_input_tokens_seen": 3473800,
      "step": 6260
    },
    {
      "epoch": 109.92035398230088,
      "grad_norm": 10.25,
      "learning_rate": 4.7035060088396965e-05,
      "loss": 1.083,
      "num_input_tokens_seen": 3476552,
      "step": 6265
    },
    {
      "epoch": 110.0,
      "grad_norm": 11.5625,
      "learning_rate": 4.703042094539839e-05,
      "loss": 0.9721,
      "num_input_tokens_seen": 3478936,
      "step": 6270
    },
    {
      "epoch": 110.08849557522124,
      "grad_norm": 9.25,
      "learning_rate": 4.702577840503206e-05,
      "loss": 1.2338,
      "num_input_tokens_seen": 3481480,
      "step": 6275
    },
    {
      "epoch": 110.17699115044248,
      "grad_norm": 7.40625,
      "learning_rate": 4.70211324680139e-05,
      "loss": 1.0958,
      "num_input_tokens_seen": 3484552,
      "step": 6280
    },
    {
      "epoch": 110.26548672566372,
      "grad_norm": 8.375,
      "learning_rate": 4.7016483135060386e-05,
      "loss": 0.7816,
      "num_input_tokens_seen": 3487064,
      "step": 6285
    },
    {
      "epoch": 110.35398230088495,
      "grad_norm": 11.25,
      "learning_rate": 4.701183040688849e-05,
      "loss": 1.2178,
      "num_input_tokens_seen": 3489784,
      "step": 6290
    },
    {
      "epoch": 110.4424778761062,
      "grad_norm": 9.875,
      "learning_rate": 4.700717428421573e-05,
      "loss": 1.1093,
      "num_input_tokens_seen": 3492424,
      "step": 6295
    },
    {
      "epoch": 110.53097345132744,
      "grad_norm": 7.65625,
      "learning_rate": 4.700251476776014e-05,
      "loss": 1.0516,
      "num_input_tokens_seen": 3495112,
      "step": 6300
    },
    {
      "epoch": 110.61946902654867,
      "grad_norm": 7.4375,
      "learning_rate": 4.699785185824026e-05,
      "loss": 0.9869,
      "num_input_tokens_seen": 3498216,
      "step": 6305
    },
    {
      "epoch": 110.70796460176992,
      "grad_norm": 5.8125,
      "learning_rate": 4.699318555637519e-05,
      "loss": 1.1085,
      "num_input_tokens_seen": 3501000,
      "step": 6310
    },
    {
      "epoch": 110.79646017699115,
      "grad_norm": 7.3125,
      "learning_rate": 4.6988515862884525e-05,
      "loss": 0.9519,
      "num_input_tokens_seen": 3503944,
      "step": 6315
    },
    {
      "epoch": 110.88495575221239,
      "grad_norm": 11.75,
      "learning_rate": 4.698384277848838e-05,
      "loss": 1.0203,
      "num_input_tokens_seen": 3506776,
      "step": 6320
    },
    {
      "epoch": 110.97345132743362,
      "grad_norm": 9.125,
      "learning_rate": 4.6979166303907425e-05,
      "loss": 1.095,
      "num_input_tokens_seen": 3509480,
      "step": 6325
    },
    {
      "epoch": 111.05309734513274,
      "grad_norm": 7.6875,
      "learning_rate": 4.697448643986281e-05,
      "loss": 1.0729,
      "num_input_tokens_seen": 3511680,
      "step": 6330
    },
    {
      "epoch": 111.14159292035399,
      "grad_norm": 7.71875,
      "learning_rate": 4.696980318707624e-05,
      "loss": 1.0503,
      "num_input_tokens_seen": 3514272,
      "step": 6335
    },
    {
      "epoch": 111.23008849557522,
      "grad_norm": 7.875,
      "learning_rate": 4.6965116546269924e-05,
      "loss": 0.9484,
      "num_input_tokens_seen": 3517056,
      "step": 6340
    },
    {
      "epoch": 111.31858407079646,
      "grad_norm": 8.25,
      "learning_rate": 4.6960426518166615e-05,
      "loss": 1.2367,
      "num_input_tokens_seen": 3519568,
      "step": 6345
    },
    {
      "epoch": 111.40707964601769,
      "grad_norm": 7.0625,
      "learning_rate": 4.6955733103489556e-05,
      "loss": 0.9601,
      "num_input_tokens_seen": 3522432,
      "step": 6350
    },
    {
      "epoch": 111.49557522123894,
      "grad_norm": 7.75,
      "learning_rate": 4.695103630296255e-05,
      "loss": 1.2897,
      "num_input_tokens_seen": 3525120,
      "step": 6355
    },
    {
      "epoch": 111.58407079646018,
      "grad_norm": 9.625,
      "learning_rate": 4.694633611730988e-05,
      "loss": 0.9677,
      "num_input_tokens_seen": 3528336,
      "step": 6360
    },
    {
      "epoch": 111.67256637168141,
      "grad_norm": 6.78125,
      "learning_rate": 4.694163254725639e-05,
      "loss": 0.9397,
      "num_input_tokens_seen": 3531184,
      "step": 6365
    },
    {
      "epoch": 111.76106194690266,
      "grad_norm": 8.4375,
      "learning_rate": 4.693692559352743e-05,
      "loss": 1.0102,
      "num_input_tokens_seen": 3533680,
      "step": 6370
    },
    {
      "epoch": 111.84955752212389,
      "grad_norm": 11.5,
      "learning_rate": 4.693221525684886e-05,
      "loss": 1.1314,
      "num_input_tokens_seen": 3536384,
      "step": 6375
    },
    {
      "epoch": 111.93805309734513,
      "grad_norm": 7.34375,
      "learning_rate": 4.6927501537947084e-05,
      "loss": 1.169,
      "num_input_tokens_seen": 3539408,
      "step": 6380
    },
    {
      "epoch": 112.01769911504425,
      "grad_norm": 8.4375,
      "learning_rate": 4.692278443754901e-05,
      "loss": 0.8387,
      "num_input_tokens_seen": 3542128,
      "step": 6385
    },
    {
      "epoch": 112.10619469026548,
      "grad_norm": 8.5625,
      "learning_rate": 4.691806395638208e-05,
      "loss": 1.1158,
      "num_input_tokens_seen": 3544720,
      "step": 6390
    },
    {
      "epoch": 112.19469026548673,
      "grad_norm": 11.75,
      "learning_rate": 4.6913340095174255e-05,
      "loss": 1.0795,
      "num_input_tokens_seen": 3547536,
      "step": 6395
    },
    {
      "epoch": 112.28318584070796,
      "grad_norm": 8.1875,
      "learning_rate": 4.690861285465399e-05,
      "loss": 1.0355,
      "num_input_tokens_seen": 3550560,
      "step": 6400
    },
    {
      "epoch": 112.28318584070796,
      "eval_loss": 1.03902006149292,
      "eval_runtime": 0.9324,
      "eval_samples_per_second": 26.812,
      "eval_steps_per_second": 13.942,
      "num_input_tokens_seen": 3550560,
      "step": 6400
    },
    {
      "epoch": 112.3716814159292,
      "grad_norm": 7.03125,
      "learning_rate": 4.690388223555031e-05,
      "loss": 0.9762,
      "num_input_tokens_seen": 3553168,
      "step": 6405
    },
    {
      "epoch": 112.46017699115045,
      "grad_norm": 9.125,
      "learning_rate": 4.689914823859273e-05,
      "loss": 0.9462,
      "num_input_tokens_seen": 3555632,
      "step": 6410
    },
    {
      "epoch": 112.54867256637168,
      "grad_norm": 10.6875,
      "learning_rate": 4.689441086451129e-05,
      "loss": 0.9967,
      "num_input_tokens_seen": 3558560,
      "step": 6415
    },
    {
      "epoch": 112.63716814159292,
      "grad_norm": 8.625,
      "learning_rate": 4.688967011403655e-05,
      "loss": 0.9813,
      "num_input_tokens_seen": 3561632,
      "step": 6420
    },
    {
      "epoch": 112.72566371681415,
      "grad_norm": 7.6875,
      "learning_rate": 4.68849259878996e-05,
      "loss": 0.9073,
      "num_input_tokens_seen": 3564688,
      "step": 6425
    },
    {
      "epoch": 112.8141592920354,
      "grad_norm": 10.625,
      "learning_rate": 4.6880178486832036e-05,
      "loss": 1.3103,
      "num_input_tokens_seen": 3567040,
      "step": 6430
    },
    {
      "epoch": 112.90265486725664,
      "grad_norm": 8.875,
      "learning_rate": 4.687542761156598e-05,
      "loss": 1.2245,
      "num_input_tokens_seen": 3569856,
      "step": 6435
    },
    {
      "epoch": 112.99115044247787,
      "grad_norm": 6.96875,
      "learning_rate": 4.6870673362834096e-05,
      "loss": 0.9716,
      "num_input_tokens_seen": 3572528,
      "step": 6440
    },
    {
      "epoch": 113.070796460177,
      "grad_norm": 9.6875,
      "learning_rate": 4.6865915741369526e-05,
      "loss": 1.4848,
      "num_input_tokens_seen": 3575184,
      "step": 6445
    },
    {
      "epoch": 113.15929203539822,
      "grad_norm": 9.375,
      "learning_rate": 4.686115474790597e-05,
      "loss": 1.2735,
      "num_input_tokens_seen": 3577872,
      "step": 6450
    },
    {
      "epoch": 113.24778761061947,
      "grad_norm": 8.0,
      "learning_rate": 4.685639038317762e-05,
      "loss": 0.9349,
      "num_input_tokens_seen": 3580768,
      "step": 6455
    },
    {
      "epoch": 113.33628318584071,
      "grad_norm": 6.90625,
      "learning_rate": 4.685162264791921e-05,
      "loss": 1.0358,
      "num_input_tokens_seen": 3583792,
      "step": 6460
    },
    {
      "epoch": 113.42477876106194,
      "grad_norm": 11.125,
      "learning_rate": 4.684685154286599e-05,
      "loss": 1.0331,
      "num_input_tokens_seen": 3586560,
      "step": 6465
    },
    {
      "epoch": 113.51327433628319,
      "grad_norm": 7.875,
      "learning_rate": 4.684207706875371e-05,
      "loss": 1.0793,
      "num_input_tokens_seen": 3589536,
      "step": 6470
    },
    {
      "epoch": 113.60176991150442,
      "grad_norm": 8.5,
      "learning_rate": 4.683729922631866e-05,
      "loss": 0.9169,
      "num_input_tokens_seen": 3592224,
      "step": 6475
    },
    {
      "epoch": 113.69026548672566,
      "grad_norm": 6.78125,
      "learning_rate": 4.683251801629765e-05,
      "loss": 0.9105,
      "num_input_tokens_seen": 3595392,
      "step": 6480
    },
    {
      "epoch": 113.77876106194691,
      "grad_norm": 9.3125,
      "learning_rate": 4.6827733439428e-05,
      "loss": 1.1296,
      "num_input_tokens_seen": 3598112,
      "step": 6485
    },
    {
      "epoch": 113.86725663716814,
      "grad_norm": 7.65625,
      "learning_rate": 4.682294549644754e-05,
      "loss": 1.0788,
      "num_input_tokens_seen": 3600560,
      "step": 6490
    },
    {
      "epoch": 113.95575221238938,
      "grad_norm": 8.25,
      "learning_rate": 4.681815418809464e-05,
      "loss": 1.0763,
      "num_input_tokens_seen": 3603296,
      "step": 6495
    },
    {
      "epoch": 114.03539823008849,
      "grad_norm": 5.59375,
      "learning_rate": 4.681335951510819e-05,
      "loss": 0.8629,
      "num_input_tokens_seen": 3605256,
      "step": 6500
    },
    {
      "epoch": 114.12389380530973,
      "grad_norm": 9.5,
      "learning_rate": 4.6808561478227576e-05,
      "loss": 1.0293,
      "num_input_tokens_seen": 3608136,
      "step": 6505
    },
    {
      "epoch": 114.21238938053098,
      "grad_norm": 9.875,
      "learning_rate": 4.680376007819271e-05,
      "loss": 1.2437,
      "num_input_tokens_seen": 3610744,
      "step": 6510
    },
    {
      "epoch": 114.30088495575221,
      "grad_norm": 7.625,
      "learning_rate": 4.679895531574405e-05,
      "loss": 1.1161,
      "num_input_tokens_seen": 3613656,
      "step": 6515
    },
    {
      "epoch": 114.38938053097345,
      "grad_norm": 6.375,
      "learning_rate": 4.679414719162253e-05,
      "loss": 1.0006,
      "num_input_tokens_seen": 3616696,
      "step": 6520
    },
    {
      "epoch": 114.47787610619469,
      "grad_norm": 8.3125,
      "learning_rate": 4.6789335706569635e-05,
      "loss": 1.184,
      "num_input_tokens_seen": 3619240,
      "step": 6525
    },
    {
      "epoch": 114.56637168141593,
      "grad_norm": 11.9375,
      "learning_rate": 4.678452086132734e-05,
      "loss": 1.3086,
      "num_input_tokens_seen": 3622248,
      "step": 6530
    },
    {
      "epoch": 114.65486725663717,
      "grad_norm": 7.90625,
      "learning_rate": 4.677970265663818e-05,
      "loss": 0.9414,
      "num_input_tokens_seen": 3625176,
      "step": 6535
    },
    {
      "epoch": 114.7433628318584,
      "grad_norm": 9.4375,
      "learning_rate": 4.677488109324517e-05,
      "loss": 0.9636,
      "num_input_tokens_seen": 3628280,
      "step": 6540
    },
    {
      "epoch": 114.83185840707965,
      "grad_norm": 7.3125,
      "learning_rate": 4.6770056171891846e-05,
      "loss": 1.0507,
      "num_input_tokens_seen": 3630824,
      "step": 6545
    },
    {
      "epoch": 114.92035398230088,
      "grad_norm": 6.4375,
      "learning_rate": 4.6765227893322286e-05,
      "loss": 0.9374,
      "num_input_tokens_seen": 3633640,
      "step": 6550
    },
    {
      "epoch": 115.0,
      "grad_norm": 13.5,
      "learning_rate": 4.676039625828107e-05,
      "loss": 0.867,
      "num_input_tokens_seen": 3636064,
      "step": 6555
    },
    {
      "epoch": 115.08849557522124,
      "grad_norm": 11.9375,
      "learning_rate": 4.675556126751328e-05,
      "loss": 1.0287,
      "num_input_tokens_seen": 3638960,
      "step": 6560
    },
    {
      "epoch": 115.17699115044248,
      "grad_norm": 8.375,
      "learning_rate": 4.6750722921764556e-05,
      "loss": 1.1876,
      "num_input_tokens_seen": 3642128,
      "step": 6565
    },
    {
      "epoch": 115.26548672566372,
      "grad_norm": 8.625,
      "learning_rate": 4.674588122178102e-05,
      "loss": 1.1233,
      "num_input_tokens_seen": 3645008,
      "step": 6570
    },
    {
      "epoch": 115.35398230088495,
      "grad_norm": 7.8125,
      "learning_rate": 4.674103616830931e-05,
      "loss": 0.9472,
      "num_input_tokens_seen": 3647808,
      "step": 6575
    },
    {
      "epoch": 115.4424778761062,
      "grad_norm": 8.3125,
      "learning_rate": 4.673618776209663e-05,
      "loss": 0.8882,
      "num_input_tokens_seen": 3650688,
      "step": 6580
    },
    {
      "epoch": 115.53097345132744,
      "grad_norm": 10.5,
      "learning_rate": 4.673133600389063e-05,
      "loss": 0.9716,
      "num_input_tokens_seen": 3653088,
      "step": 6585
    },
    {
      "epoch": 115.61946902654867,
      "grad_norm": 6.625,
      "learning_rate": 4.672648089443953e-05,
      "loss": 1.0548,
      "num_input_tokens_seen": 3655936,
      "step": 6590
    },
    {
      "epoch": 115.70796460176992,
      "grad_norm": 7.3125,
      "learning_rate": 4.672162243449204e-05,
      "loss": 1.0833,
      "num_input_tokens_seen": 3658352,
      "step": 6595
    },
    {
      "epoch": 115.79646017699115,
      "grad_norm": 13.5625,
      "learning_rate": 4.67167606247974e-05,
      "loss": 0.9656,
      "num_input_tokens_seen": 3661824,
      "step": 6600
    },
    {
      "epoch": 115.79646017699115,
      "eval_loss": 1.0291366577148438,
      "eval_runtime": 0.9331,
      "eval_samples_per_second": 26.793,
      "eval_steps_per_second": 13.932,
      "num_input_tokens_seen": 3661824,
      "step": 6600
    },
    {
      "epoch": 115.88495575221239,
      "grad_norm": 9.0,
      "learning_rate": 4.671189546610536e-05,
      "loss": 1.2862,
      "num_input_tokens_seen": 3664320,
      "step": 6605
    },
    {
      "epoch": 115.97345132743362,
      "grad_norm": 8.0625,
      "learning_rate": 4.67070269591662e-05,
      "loss": 1.0544,
      "num_input_tokens_seen": 3667136,
      "step": 6610
    },
    {
      "epoch": 116.05309734513274,
      "grad_norm": 11.1875,
      "learning_rate": 4.670215510473068e-05,
      "loss": 1.1881,
      "num_input_tokens_seen": 3669760,
      "step": 6615
    },
    {
      "epoch": 116.14159292035399,
      "grad_norm": 7.90625,
      "learning_rate": 4.669727990355013e-05,
      "loss": 1.0476,
      "num_input_tokens_seen": 3672224,
      "step": 6620
    },
    {
      "epoch": 116.23008849557522,
      "grad_norm": 6.875,
      "learning_rate": 4.669240135637635e-05,
      "loss": 0.9718,
      "num_input_tokens_seen": 3675488,
      "step": 6625
    },
    {
      "epoch": 116.31858407079646,
      "grad_norm": 8.4375,
      "learning_rate": 4.6687519463961675e-05,
      "loss": 1.2731,
      "num_input_tokens_seen": 3678352,
      "step": 6630
    },
    {
      "epoch": 116.40707964601769,
      "grad_norm": 7.625,
      "learning_rate": 4.668263422705896e-05,
      "loss": 0.9356,
      "num_input_tokens_seen": 3681216,
      "step": 6635
    },
    {
      "epoch": 116.49557522123894,
      "grad_norm": 8.75,
      "learning_rate": 4.667774564642156e-05,
      "loss": 1.0166,
      "num_input_tokens_seen": 3684032,
      "step": 6640
    },
    {
      "epoch": 116.58407079646018,
      "grad_norm": 9.125,
      "learning_rate": 4.6672853722803365e-05,
      "loss": 0.8796,
      "num_input_tokens_seen": 3686960,
      "step": 6645
    },
    {
      "epoch": 116.67256637168141,
      "grad_norm": 8.25,
      "learning_rate": 4.666795845695877e-05,
      "loss": 0.9983,
      "num_input_tokens_seen": 3689744,
      "step": 6650
    },
    {
      "epoch": 116.76106194690266,
      "grad_norm": 9.375,
      "learning_rate": 4.666305984964269e-05,
      "loss": 1.1184,
      "num_input_tokens_seen": 3692096,
      "step": 6655
    },
    {
      "epoch": 116.84955752212389,
      "grad_norm": 8.625,
      "learning_rate": 4.6658157901610535e-05,
      "loss": 1.0095,
      "num_input_tokens_seen": 3694928,
      "step": 6660
    },
    {
      "epoch": 116.93805309734513,
      "grad_norm": 9.5625,
      "learning_rate": 4.665325261361826e-05,
      "loss": 1.1492,
      "num_input_tokens_seen": 3697840,
      "step": 6665
    },
    {
      "epoch": 117.01769911504425,
      "grad_norm": 8.0625,
      "learning_rate": 4.664834398642232e-05,
      "loss": 1.1894,
      "num_input_tokens_seen": 3700072,
      "step": 6670
    },
    {
      "epoch": 117.10619469026548,
      "grad_norm": 9.3125,
      "learning_rate": 4.6643432020779686e-05,
      "loss": 0.9905,
      "num_input_tokens_seen": 3702728,
      "step": 6675
    },
    {
      "epoch": 117.19469026548673,
      "grad_norm": 7.34375,
      "learning_rate": 4.663851671744786e-05,
      "loss": 1.1434,
      "num_input_tokens_seen": 3705224,
      "step": 6680
    },
    {
      "epoch": 117.28318584070796,
      "grad_norm": 5.5,
      "learning_rate": 4.6633598077184815e-05,
      "loss": 0.9611,
      "num_input_tokens_seen": 3707640,
      "step": 6685
    },
    {
      "epoch": 117.3716814159292,
      "grad_norm": 10.5,
      "learning_rate": 4.662867610074908e-05,
      "loss": 1.1321,
      "num_input_tokens_seen": 3710632,
      "step": 6690
    },
    {
      "epoch": 117.46017699115045,
      "grad_norm": 8.5625,
      "learning_rate": 4.6623750788899696e-05,
      "loss": 1.0283,
      "num_input_tokens_seen": 3713784,
      "step": 6695
    },
    {
      "epoch": 117.54867256637168,
      "grad_norm": 7.5,
      "learning_rate": 4.6618822142396195e-05,
      "loss": 0.8718,
      "num_input_tokens_seen": 3716920,
      "step": 6700
    },
    {
      "epoch": 117.63716814159292,
      "grad_norm": 6.875,
      "learning_rate": 4.661389016199864e-05,
      "loss": 1.1911,
      "num_input_tokens_seen": 3719672,
      "step": 6705
    },
    {
      "epoch": 117.72566371681415,
      "grad_norm": 6.65625,
      "learning_rate": 4.660895484846761e-05,
      "loss": 0.9079,
      "num_input_tokens_seen": 3722184,
      "step": 6710
    },
    {
      "epoch": 117.8141592920354,
      "grad_norm": 6.59375,
      "learning_rate": 4.660401620256418e-05,
      "loss": 1.092,
      "num_input_tokens_seen": 3725176,
      "step": 6715
    },
    {
      "epoch": 117.90265486725664,
      "grad_norm": 9.8125,
      "learning_rate": 4.659907422504997e-05,
      "loss": 1.221,
      "num_input_tokens_seen": 3727832,
      "step": 6720
    },
    {
      "epoch": 117.99115044247787,
      "grad_norm": 7.53125,
      "learning_rate": 4.6594128916687074e-05,
      "loss": 1.0488,
      "num_input_tokens_seen": 3730968,
      "step": 6725
    },
    {
      "epoch": 118.070796460177,
      "grad_norm": 9.1875,
      "learning_rate": 4.658918027823813e-05,
      "loss": 1.175,
      "num_input_tokens_seen": 3733488,
      "step": 6730
    },
    {
      "epoch": 118.15929203539822,
      "grad_norm": 6.8125,
      "learning_rate": 4.658422831046628e-05,
      "loss": 1.0751,
      "num_input_tokens_seen": 3736368,
      "step": 6735
    },
    {
      "epoch": 118.24778761061947,
      "grad_norm": 8.6875,
      "learning_rate": 4.657927301413518e-05,
      "loss": 1.0516,
      "num_input_tokens_seen": 3739216,
      "step": 6740
    },
    {
      "epoch": 118.33628318584071,
      "grad_norm": 9.25,
      "learning_rate": 4.657431439000901e-05,
      "loss": 1.018,
      "num_input_tokens_seen": 3741968,
      "step": 6745
    },
    {
      "epoch": 118.42477876106194,
      "grad_norm": 8.25,
      "learning_rate": 4.656935243885243e-05,
      "loss": 0.8658,
      "num_input_tokens_seen": 3744880,
      "step": 6750
    },
    {
      "epoch": 118.51327433628319,
      "grad_norm": 8.375,
      "learning_rate": 4.656438716143066e-05,
      "loss": 1.2342,
      "num_input_tokens_seen": 3747296,
      "step": 6755
    },
    {
      "epoch": 118.60176991150442,
      "grad_norm": 7.5,
      "learning_rate": 4.6559418558509384e-05,
      "loss": 0.8262,
      "num_input_tokens_seen": 3750576,
      "step": 6760
    },
    {
      "epoch": 118.69026548672566,
      "grad_norm": 9.75,
      "learning_rate": 4.6554446630854833e-05,
      "loss": 1.3456,
      "num_input_tokens_seen": 3753488,
      "step": 6765
    },
    {
      "epoch": 118.77876106194691,
      "grad_norm": 6.3125,
      "learning_rate": 4.654947137923374e-05,
      "loss": 0.9925,
      "num_input_tokens_seen": 3756128,
      "step": 6770
    },
    {
      "epoch": 118.86725663716814,
      "grad_norm": 6.625,
      "learning_rate": 4.654449280441335e-05,
      "loss": 0.9571,
      "num_input_tokens_seen": 3758912,
      "step": 6775
    },
    {
      "epoch": 118.95575221238938,
      "grad_norm": 11.3125,
      "learning_rate": 4.653951090716143e-05,
      "loss": 0.9265,
      "num_input_tokens_seen": 3761472,
      "step": 6780
    },
    {
      "epoch": 119.03539823008849,
      "grad_norm": 7.65625,
      "learning_rate": 4.653452568824625e-05,
      "loss": 1.1192,
      "num_input_tokens_seen": 3763616,
      "step": 6785
    },
    {
      "epoch": 119.12389380530973,
      "grad_norm": 7.53125,
      "learning_rate": 4.6529537148436585e-05,
      "loss": 1.1671,
      "num_input_tokens_seen": 3765776,
      "step": 6790
    },
    {
      "epoch": 119.21238938053098,
      "grad_norm": 9.5625,
      "learning_rate": 4.6524545288501734e-05,
      "loss": 1.1212,
      "num_input_tokens_seen": 3768768,
      "step": 6795
    },
    {
      "epoch": 119.30088495575221,
      "grad_norm": 7.09375,
      "learning_rate": 4.6519550109211506e-05,
      "loss": 0.8823,
      "num_input_tokens_seen": 3771856,
      "step": 6800
    },
    {
      "epoch": 119.30088495575221,
      "eval_loss": 1.0279802083969116,
      "eval_runtime": 0.9341,
      "eval_samples_per_second": 26.763,
      "eval_steps_per_second": 13.917,
      "num_input_tokens_seen": 3771856,
      "step": 6800
    },
    {
      "epoch": 119.38938053097345,
      "grad_norm": 7.4375,
      "learning_rate": 4.651455161133622e-05,
      "loss": 0.9444,
      "num_input_tokens_seen": 3774448,
      "step": 6805
    },
    {
      "epoch": 119.47787610619469,
      "grad_norm": 7.65625,
      "learning_rate": 4.6509549795646704e-05,
      "loss": 1.1816,
      "num_input_tokens_seen": 3777152,
      "step": 6810
    },
    {
      "epoch": 119.56637168141593,
      "grad_norm": 9.0625,
      "learning_rate": 4.6504544662914306e-05,
      "loss": 1.0044,
      "num_input_tokens_seen": 3780304,
      "step": 6815
    },
    {
      "epoch": 119.65486725663717,
      "grad_norm": 8.8125,
      "learning_rate": 4.6499536213910876e-05,
      "loss": 1.1449,
      "num_input_tokens_seen": 3783008,
      "step": 6820
    },
    {
      "epoch": 119.7433628318584,
      "grad_norm": 8.875,
      "learning_rate": 4.6494524449408786e-05,
      "loss": 1.1562,
      "num_input_tokens_seen": 3785376,
      "step": 6825
    },
    {
      "epoch": 119.83185840707965,
      "grad_norm": 7.0,
      "learning_rate": 4.6489509370180903e-05,
      "loss": 1.0546,
      "num_input_tokens_seen": 3788576,
      "step": 6830
    },
    {
      "epoch": 119.92035398230088,
      "grad_norm": 7.75,
      "learning_rate": 4.648449097700063e-05,
      "loss": 1.1555,
      "num_input_tokens_seen": 3791488,
      "step": 6835
    },
    {
      "epoch": 120.0,
      "grad_norm": 9.5625,
      "learning_rate": 4.647946927064185e-05,
      "loss": 0.9387,
      "num_input_tokens_seen": 3793928,
      "step": 6840
    },
    {
      "epoch": 120.08849557522124,
      "grad_norm": 7.78125,
      "learning_rate": 4.647444425187898e-05,
      "loss": 0.9683,
      "num_input_tokens_seen": 3796536,
      "step": 6845
    },
    {
      "epoch": 120.17699115044248,
      "grad_norm": 10.3125,
      "learning_rate": 4.646941592148695e-05,
      "loss": 1.0333,
      "num_input_tokens_seen": 3799384,
      "step": 6850
    },
    {
      "epoch": 120.26548672566372,
      "grad_norm": 8.125,
      "learning_rate": 4.646438428024117e-05,
      "loss": 1.2244,
      "num_input_tokens_seen": 3801944,
      "step": 6855
    },
    {
      "epoch": 120.35398230088495,
      "grad_norm": 6.65625,
      "learning_rate": 4.64593493289176e-05,
      "loss": 1.0762,
      "num_input_tokens_seen": 3804808,
      "step": 6860
    },
    {
      "epoch": 120.4424778761062,
      "grad_norm": 8.75,
      "learning_rate": 4.64543110682927e-05,
      "loss": 1.0394,
      "num_input_tokens_seen": 3807400,
      "step": 6865
    },
    {
      "epoch": 120.53097345132744,
      "grad_norm": 8.0,
      "learning_rate": 4.644926949914341e-05,
      "loss": 1.0321,
      "num_input_tokens_seen": 3810264,
      "step": 6870
    },
    {
      "epoch": 120.61946902654867,
      "grad_norm": 7.375,
      "learning_rate": 4.644422462224722e-05,
      "loss": 0.9681,
      "num_input_tokens_seen": 3813160,
      "step": 6875
    },
    {
      "epoch": 120.70796460176992,
      "grad_norm": 7.28125,
      "learning_rate": 4.643917643838211e-05,
      "loss": 0.8487,
      "num_input_tokens_seen": 3816424,
      "step": 6880
    },
    {
      "epoch": 120.79646017699115,
      "grad_norm": 6.9375,
      "learning_rate": 4.6434124948326564e-05,
      "loss": 1.3025,
      "num_input_tokens_seen": 3819080,
      "step": 6885
    },
    {
      "epoch": 120.88495575221239,
      "grad_norm": 6.6875,
      "learning_rate": 4.6429070152859594e-05,
      "loss": 1.0931,
      "num_input_tokens_seen": 3821784,
      "step": 6890
    },
    {
      "epoch": 120.97345132743362,
      "grad_norm": 6.8125,
      "learning_rate": 4.6424012052760714e-05,
      "loss": 1.0471,
      "num_input_tokens_seen": 3824648,
      "step": 6895
    },
    {
      "epoch": 121.05309734513274,
      "grad_norm": 9.75,
      "learning_rate": 4.6418950648809945e-05,
      "loss": 0.9972,
      "num_input_tokens_seen": 3827112,
      "step": 6900
    },
    {
      "epoch": 121.14159292035399,
      "grad_norm": 10.5,
      "learning_rate": 4.641388594178782e-05,
      "loss": 1.1057,
      "num_input_tokens_seen": 3830296,
      "step": 6905
    },
    {
      "epoch": 121.23008849557522,
      "grad_norm": 8.6875,
      "learning_rate": 4.640881793247538e-05,
      "loss": 1.0337,
      "num_input_tokens_seen": 3833304,
      "step": 6910
    },
    {
      "epoch": 121.31858407079646,
      "grad_norm": 8.5,
      "learning_rate": 4.6403746621654173e-05,
      "loss": 0.9856,
      "num_input_tokens_seen": 3835784,
      "step": 6915
    },
    {
      "epoch": 121.40707964601769,
      "grad_norm": 7.375,
      "learning_rate": 4.639867201010626e-05,
      "loss": 1.0357,
      "num_input_tokens_seen": 3838712,
      "step": 6920
    },
    {
      "epoch": 121.49557522123894,
      "grad_norm": 7.71875,
      "learning_rate": 4.6393594098614204e-05,
      "loss": 1.1043,
      "num_input_tokens_seen": 3841272,
      "step": 6925
    },
    {
      "epoch": 121.58407079646018,
      "grad_norm": 6.21875,
      "learning_rate": 4.63885128879611e-05,
      "loss": 1.0653,
      "num_input_tokens_seen": 3844472,
      "step": 6930
    },
    {
      "epoch": 121.67256637168141,
      "grad_norm": 8.5625,
      "learning_rate": 4.638342837893052e-05,
      "loss": 1.0052,
      "num_input_tokens_seen": 3847512,
      "step": 6935
    },
    {
      "epoch": 121.76106194690266,
      "grad_norm": 9.0625,
      "learning_rate": 4.6378340572306565e-05,
      "loss": 0.9743,
      "num_input_tokens_seen": 3849992,
      "step": 6940
    },
    {
      "epoch": 121.84955752212389,
      "grad_norm": 7.65625,
      "learning_rate": 4.6373249468873833e-05,
      "loss": 0.9829,
      "num_input_tokens_seen": 3852680,
      "step": 6945
    },
    {
      "epoch": 121.93805309734513,
      "grad_norm": 9.6875,
      "learning_rate": 4.636815506941744e-05,
      "loss": 1.1533,
      "num_input_tokens_seen": 3855672,
      "step": 6950
    },
    {
      "epoch": 122.01769911504425,
      "grad_norm": 6.3125,
      "learning_rate": 4.6363057374723004e-05,
      "loss": 1.0407,
      "num_input_tokens_seen": 3857960,
      "step": 6955
    },
    {
      "epoch": 122.10619469026548,
      "grad_norm": 10.625,
      "learning_rate": 4.635795638557666e-05,
      "loss": 1.0744,
      "num_input_tokens_seen": 3860312,
      "step": 6960
    },
    {
      "epoch": 122.19469026548673,
      "grad_norm": 7.53125,
      "learning_rate": 4.635285210276504e-05,
      "loss": 1.1606,
      "num_input_tokens_seen": 3863304,
      "step": 6965
    },
    {
      "epoch": 122.28318584070796,
      "grad_norm": 7.78125,
      "learning_rate": 4.6347744527075295e-05,
      "loss": 0.9263,
      "num_input_tokens_seen": 3866392,
      "step": 6970
    },
    {
      "epoch": 122.3716814159292,
      "grad_norm": 7.59375,
      "learning_rate": 4.634263365929506e-05,
      "loss": 1.0484,
      "num_input_tokens_seen": 3868792,
      "step": 6975
    },
    {
      "epoch": 122.46017699115045,
      "grad_norm": 9.4375,
      "learning_rate": 4.6337519500212515e-05,
      "loss": 1.0128,
      "num_input_tokens_seen": 3871816,
      "step": 6980
    },
    {
      "epoch": 122.54867256637168,
      "grad_norm": 8.625,
      "learning_rate": 4.633240205061632e-05,
      "loss": 1.0007,
      "num_input_tokens_seen": 3874664,
      "step": 6985
    },
    {
      "epoch": 122.63716814159292,
      "grad_norm": 9.625,
      "learning_rate": 4.632728131129565e-05,
      "loss": 1.062,
      "num_input_tokens_seen": 3877880,
      "step": 6990
    },
    {
      "epoch": 122.72566371681415,
      "grad_norm": 10.125,
      "learning_rate": 4.632215728304018e-05,
      "loss": 1.0484,
      "num_input_tokens_seen": 3880280,
      "step": 6995
    },
    {
      "epoch": 122.8141592920354,
      "grad_norm": 10.5,
      "learning_rate": 4.63170299666401e-05,
      "loss": 1.2007,
      "num_input_tokens_seen": 3883176,
      "step": 7000
    },
    {
      "epoch": 122.8141592920354,
      "eval_loss": 1.0319406986236572,
      "eval_runtime": 0.9365,
      "eval_samples_per_second": 26.694,
      "eval_steps_per_second": 13.881,
      "num_input_tokens_seen": 3883176,
      "step": 7000
    },
    {
      "epoch": 122.90265486725664,
      "grad_norm": 6.125,
      "learning_rate": 4.631189936288612e-05,
      "loss": 1.0308,
      "num_input_tokens_seen": 3886104,
      "step": 7005
    },
    {
      "epoch": 122.99115044247787,
      "grad_norm": 7.6875,
      "learning_rate": 4.630676547256944e-05,
      "loss": 1.0313,
      "num_input_tokens_seen": 3888744,
      "step": 7010
    },
    {
      "epoch": 123.070796460177,
      "grad_norm": 9.25,
      "learning_rate": 4.630162829648176e-05,
      "loss": 1.0259,
      "num_input_tokens_seen": 3891928,
      "step": 7015
    },
    {
      "epoch": 123.15929203539822,
      "grad_norm": 8.75,
      "learning_rate": 4.629648783541531e-05,
      "loss": 1.0975,
      "num_input_tokens_seen": 3894808,
      "step": 7020
    },
    {
      "epoch": 123.24778761061947,
      "grad_norm": 8.5625,
      "learning_rate": 4.6291344090162804e-05,
      "loss": 1.0785,
      "num_input_tokens_seen": 3897544,
      "step": 7025
    },
    {
      "epoch": 123.33628318584071,
      "grad_norm": 6.375,
      "learning_rate": 4.628619706151748e-05,
      "loss": 1.0576,
      "num_input_tokens_seen": 3900136,
      "step": 7030
    },
    {
      "epoch": 123.42477876106194,
      "grad_norm": 10.6875,
      "learning_rate": 4.628104675027306e-05,
      "loss": 1.023,
      "num_input_tokens_seen": 3902808,
      "step": 7035
    },
    {
      "epoch": 123.51327433628319,
      "grad_norm": 6.1875,
      "learning_rate": 4.6275893157223805e-05,
      "loss": 0.8962,
      "num_input_tokens_seen": 3905480,
      "step": 7040
    },
    {
      "epoch": 123.60176991150442,
      "grad_norm": 8.6875,
      "learning_rate": 4.627073628316445e-05,
      "loss": 0.9585,
      "num_input_tokens_seen": 3908408,
      "step": 7045
    },
    {
      "epoch": 123.69026548672566,
      "grad_norm": 8.25,
      "learning_rate": 4.626557612889026e-05,
      "loss": 1.1271,
      "num_input_tokens_seen": 3911416,
      "step": 7050
    },
    {
      "epoch": 123.77876106194691,
      "grad_norm": 11.0625,
      "learning_rate": 4.626041269519699e-05,
      "loss": 1.1275,
      "num_input_tokens_seen": 3913864,
      "step": 7055
    },
    {
      "epoch": 123.86725663716814,
      "grad_norm": 6.875,
      "learning_rate": 4.6255245982880905e-05,
      "loss": 1.0125,
      "num_input_tokens_seen": 3916552,
      "step": 7060
    },
    {
      "epoch": 123.95575221238938,
      "grad_norm": 12.0625,
      "learning_rate": 4.625007599273879e-05,
      "loss": 0.9727,
      "num_input_tokens_seen": 3919288,
      "step": 7065
    },
    {
      "epoch": 124.03539823008849,
      "grad_norm": 7.34375,
      "learning_rate": 4.6244902725567895e-05,
      "loss": 1.1571,
      "num_input_tokens_seen": 3921896,
      "step": 7070
    },
    {
      "epoch": 124.12389380530973,
      "grad_norm": 9.1875,
      "learning_rate": 4.6239726182166024e-05,
      "loss": 0.9505,
      "num_input_tokens_seen": 3925208,
      "step": 7075
    },
    {
      "epoch": 124.21238938053098,
      "grad_norm": 7.3125,
      "learning_rate": 4.623454636333147e-05,
      "loss": 0.8296,
      "num_input_tokens_seen": 3927768,
      "step": 7080
    },
    {
      "epoch": 124.30088495575221,
      "grad_norm": 9.1875,
      "learning_rate": 4.622936326986301e-05,
      "loss": 0.9104,
      "num_input_tokens_seen": 3930536,
      "step": 7085
    },
    {
      "epoch": 124.38938053097345,
      "grad_norm": 9.0,
      "learning_rate": 4.6224176902559946e-05,
      "loss": 1.1612,
      "num_input_tokens_seen": 3933992,
      "step": 7090
    },
    {
      "epoch": 124.47787610619469,
      "grad_norm": 9.3125,
      "learning_rate": 4.621898726222209e-05,
      "loss": 0.9508,
      "num_input_tokens_seen": 3936952,
      "step": 7095
    },
    {
      "epoch": 124.56637168141593,
      "grad_norm": 8.6875,
      "learning_rate": 4.6213794349649744e-05,
      "loss": 1.2232,
      "num_input_tokens_seen": 3939464,
      "step": 7100
    },
    {
      "epoch": 124.65486725663717,
      "grad_norm": 12.8125,
      "learning_rate": 4.6208598165643715e-05,
      "loss": 1.2332,
      "num_input_tokens_seen": 3941784,
      "step": 7105
    },
    {
      "epoch": 124.7433628318584,
      "grad_norm": 6.4375,
      "learning_rate": 4.620339871100533e-05,
      "loss": 0.9312,
      "num_input_tokens_seen": 3944968,
      "step": 7110
    },
    {
      "epoch": 124.83185840707965,
      "grad_norm": 7.5625,
      "learning_rate": 4.6198195986536394e-05,
      "loss": 1.3147,
      "num_input_tokens_seen": 3947336,
      "step": 7115
    },
    {
      "epoch": 124.92035398230088,
      "grad_norm": 6.09375,
      "learning_rate": 4.619298999303926e-05,
      "loss": 1.0474,
      "num_input_tokens_seen": 3949944,
      "step": 7120
    },
    {
      "epoch": 125.0,
      "grad_norm": 15.4375,
      "learning_rate": 4.618778073131673e-05,
      "loss": 1.0407,
      "num_input_tokens_seen": 3952280,
      "step": 7125
    },
    {
      "epoch": 125.08849557522124,
      "grad_norm": 9.9375,
      "learning_rate": 4.618256820217215e-05,
      "loss": 1.1008,
      "num_input_tokens_seen": 3955288,
      "step": 7130
    },
    {
      "epoch": 125.17699115044248,
      "grad_norm": 9.5625,
      "learning_rate": 4.617735240640936e-05,
      "loss": 1.1395,
      "num_input_tokens_seen": 3957992,
      "step": 7135
    },
    {
      "epoch": 125.26548672566372,
      "grad_norm": 8.9375,
      "learning_rate": 4.6172133344832705e-05,
      "loss": 1.0462,
      "num_input_tokens_seen": 3960840,
      "step": 7140
    },
    {
      "epoch": 125.35398230088495,
      "grad_norm": 7.0,
      "learning_rate": 4.6166911018247004e-05,
      "loss": 0.9666,
      "num_input_tokens_seen": 3963736,
      "step": 7145
    },
    {
      "epoch": 125.4424778761062,
      "grad_norm": 13.0,
      "learning_rate": 4.616168542745764e-05,
      "loss": 1.2583,
      "num_input_tokens_seen": 3966360,
      "step": 7150
    },
    {
      "epoch": 125.53097345132744,
      "grad_norm": 9.25,
      "learning_rate": 4.6156456573270446e-05,
      "loss": 1.1458,
      "num_input_tokens_seen": 3969224,
      "step": 7155
    },
    {
      "epoch": 125.61946902654867,
      "grad_norm": 9.8125,
      "learning_rate": 4.615122445649177e-05,
      "loss": 0.9272,
      "num_input_tokens_seen": 3972200,
      "step": 7160
    },
    {
      "epoch": 125.70796460176992,
      "grad_norm": 8.5625,
      "learning_rate": 4.6145989077928486e-05,
      "loss": 1.0762,
      "num_input_tokens_seen": 3974536,
      "step": 7165
    },
    {
      "epoch": 125.79646017699115,
      "grad_norm": 7.78125,
      "learning_rate": 4.6140750438387953e-05,
      "loss": 0.9552,
      "num_input_tokens_seen": 3977448,
      "step": 7170
    },
    {
      "epoch": 125.88495575221239,
      "grad_norm": 5.96875,
      "learning_rate": 4.613550853867803e-05,
      "loss": 0.9849,
      "num_input_tokens_seen": 3980232,
      "step": 7175
    },
    {
      "epoch": 125.97345132743362,
      "grad_norm": 6.78125,
      "learning_rate": 4.613026337960708e-05,
      "loss": 0.94,
      "num_input_tokens_seen": 3983384,
      "step": 7180
    },
    {
      "epoch": 126.05309734513274,
      "grad_norm": 7.40625,
      "learning_rate": 4.612501496198398e-05,
      "loss": 1.1081,
      "num_input_tokens_seen": 3985720,
      "step": 7185
    },
    {
      "epoch": 126.14159292035399,
      "grad_norm": 7.6875,
      "learning_rate": 4.61197632866181e-05,
      "loss": 1.0473,
      "num_input_tokens_seen": 3988824,
      "step": 7190
    },
    {
      "epoch": 126.23008849557522,
      "grad_norm": 7.3125,
      "learning_rate": 4.611450835431931e-05,
      "loss": 0.9633,
      "num_input_tokens_seen": 3991768,
      "step": 7195
    },
    {
      "epoch": 126.31858407079646,
      "grad_norm": 10.3125,
      "learning_rate": 4.6109250165898e-05,
      "loss": 1.1485,
      "num_input_tokens_seen": 3994264,
      "step": 7200
    },
    {
      "epoch": 126.31858407079646,
      "eval_loss": 1.029668927192688,
      "eval_runtime": 0.9363,
      "eval_samples_per_second": 26.7,
      "eval_steps_per_second": 13.884,
      "num_input_tokens_seen": 3994264,
      "step": 7200
    },
    {
      "epoch": 126.40707964601769,
      "grad_norm": 8.8125,
      "learning_rate": 4.610398872216503e-05,
      "loss": 0.8854,
      "num_input_tokens_seen": 3997016,
      "step": 7205
    },
    {
      "epoch": 126.49557522123894,
      "grad_norm": 7.3125,
      "learning_rate": 4.6098724023931796e-05,
      "loss": 1.0883,
      "num_input_tokens_seen": 3999400,
      "step": 7210
    },
    {
      "epoch": 126.58407079646018,
      "grad_norm": 9.3125,
      "learning_rate": 4.609345607201017e-05,
      "loss": 1.0792,
      "num_input_tokens_seen": 4002552,
      "step": 7215
    },
    {
      "epoch": 126.67256637168141,
      "grad_norm": 7.6875,
      "learning_rate": 4.608818486721254e-05,
      "loss": 0.9385,
      "num_input_tokens_seen": 4005880,
      "step": 7220
    },
    {
      "epoch": 126.76106194690266,
      "grad_norm": 7.9375,
      "learning_rate": 4.608291041035179e-05,
      "loss": 1.1743,
      "num_input_tokens_seen": 4008616,
      "step": 7225
    },
    {
      "epoch": 126.84955752212389,
      "grad_norm": 7.6875,
      "learning_rate": 4.607763270224132e-05,
      "loss": 1.0551,
      "num_input_tokens_seen": 4011128,
      "step": 7230
    },
    {
      "epoch": 126.93805309734513,
      "grad_norm": 7.21875,
      "learning_rate": 4.6072351743695e-05,
      "loss": 1.0884,
      "num_input_tokens_seen": 4013832,
      "step": 7235
    },
    {
      "epoch": 127.01769911504425,
      "grad_norm": 8.375,
      "learning_rate": 4.606706753552723e-05,
      "loss": 1.1672,
      "num_input_tokens_seen": 4016376,
      "step": 7240
    },
    {
      "epoch": 127.10619469026548,
      "grad_norm": 6.28125,
      "learning_rate": 4.6061780078552906e-05,
      "loss": 0.8395,
      "num_input_tokens_seen": 4019224,
      "step": 7245
    },
    {
      "epoch": 127.19469026548673,
      "grad_norm": 7.71875,
      "learning_rate": 4.605648937358742e-05,
      "loss": 0.9121,
      "num_input_tokens_seen": 4021928,
      "step": 7250
    },
    {
      "epoch": 127.28318584070796,
      "grad_norm": 8.9375,
      "learning_rate": 4.605119542144665e-05,
      "loss": 1.1691,
      "num_input_tokens_seen": 4024424,
      "step": 7255
    },
    {
      "epoch": 127.3716814159292,
      "grad_norm": 6.96875,
      "learning_rate": 4.604589822294701e-05,
      "loss": 0.9015,
      "num_input_tokens_seen": 4027272,
      "step": 7260
    },
    {
      "epoch": 127.46017699115045,
      "grad_norm": 10.3125,
      "learning_rate": 4.604059777890537e-05,
      "loss": 1.2098,
      "num_input_tokens_seen": 4029768,
      "step": 7265
    },
    {
      "epoch": 127.54867256637168,
      "grad_norm": 7.6875,
      "learning_rate": 4.6035294090139145e-05,
      "loss": 0.9675,
      "num_input_tokens_seen": 4032984,
      "step": 7270
    },
    {
      "epoch": 127.63716814159292,
      "grad_norm": 10.3125,
      "learning_rate": 4.6029987157466226e-05,
      "loss": 1.0308,
      "num_input_tokens_seen": 4036024,
      "step": 7275
    },
    {
      "epoch": 127.72566371681415,
      "grad_norm": 11.875,
      "learning_rate": 4.602467698170502e-05,
      "loss": 0.9512,
      "num_input_tokens_seen": 4038776,
      "step": 7280
    },
    {
      "epoch": 127.8141592920354,
      "grad_norm": 7.875,
      "learning_rate": 4.601936356367439e-05,
      "loss": 0.925,
      "num_input_tokens_seen": 4042024,
      "step": 7285
    },
    {
      "epoch": 127.90265486725664,
      "grad_norm": 10.25,
      "learning_rate": 4.601404690419377e-05,
      "loss": 1.377,
      "num_input_tokens_seen": 4044648,
      "step": 7290
    },
    {
      "epoch": 127.99115044247787,
      "grad_norm": 11.8125,
      "learning_rate": 4.600872700408303e-05,
      "loss": 1.3573,
      "num_input_tokens_seen": 4047816,
      "step": 7295
    },
    {
      "epoch": 128.07079646017698,
      "grad_norm": 7.375,
      "learning_rate": 4.600340386416258e-05,
      "loss": 1.0464,
      "num_input_tokens_seen": 4050328,
      "step": 7300
    },
    {
      "epoch": 128.15929203539824,
      "grad_norm": 8.6875,
      "learning_rate": 4.5998077485253296e-05,
      "loss": 0.9211,
      "num_input_tokens_seen": 4053112,
      "step": 7305
    },
    {
      "epoch": 128.24778761061947,
      "grad_norm": 8.75,
      "learning_rate": 4.59927478681766e-05,
      "loss": 1.1149,
      "num_input_tokens_seen": 4055624,
      "step": 7310
    },
    {
      "epoch": 128.3362831858407,
      "grad_norm": 6.5,
      "learning_rate": 4.5987415013754366e-05,
      "loss": 0.9875,
      "num_input_tokens_seen": 4058200,
      "step": 7315
    },
    {
      "epoch": 128.42477876106196,
      "grad_norm": 8.5625,
      "learning_rate": 4.598207892280899e-05,
      "loss": 0.9645,
      "num_input_tokens_seen": 4061064,
      "step": 7320
    },
    {
      "epoch": 128.5132743362832,
      "grad_norm": 7.4375,
      "learning_rate": 4.597673959616337e-05,
      "loss": 1.0827,
      "num_input_tokens_seen": 4063992,
      "step": 7325
    },
    {
      "epoch": 128.60176991150442,
      "grad_norm": 8.75,
      "learning_rate": 4.597139703464089e-05,
      "loss": 1.1338,
      "num_input_tokens_seen": 4067080,
      "step": 7330
    },
    {
      "epoch": 128.69026548672565,
      "grad_norm": 5.78125,
      "learning_rate": 4.596605123906545e-05,
      "loss": 0.9344,
      "num_input_tokens_seen": 4069864,
      "step": 7335
    },
    {
      "epoch": 128.7787610619469,
      "grad_norm": 12.125,
      "learning_rate": 4.596070221026143e-05,
      "loss": 1.2131,
      "num_input_tokens_seen": 4072504,
      "step": 7340
    },
    {
      "epoch": 128.86725663716814,
      "grad_norm": 8.0625,
      "learning_rate": 4.595534994905372e-05,
      "loss": 1.1686,
      "num_input_tokens_seen": 4075320,
      "step": 7345
    },
    {
      "epoch": 128.95575221238937,
      "grad_norm": 9.6875,
      "learning_rate": 4.594999445626771e-05,
      "loss": 1.1458,
      "num_input_tokens_seen": 4077784,
      "step": 7350
    },
    {
      "epoch": 129.0353982300885,
      "grad_norm": 8.75,
      "learning_rate": 4.5944635732729276e-05,
      "loss": 0.9473,
      "num_input_tokens_seen": 4080128,
      "step": 7355
    },
    {
      "epoch": 129.12389380530973,
      "grad_norm": 7.5625,
      "learning_rate": 4.5939273779264804e-05,
      "loss": 1.1109,
      "num_input_tokens_seen": 4083104,
      "step": 7360
    },
    {
      "epoch": 129.21238938053096,
      "grad_norm": 9.5625,
      "learning_rate": 4.593390859670118e-05,
      "loss": 1.0002,
      "num_input_tokens_seen": 4086512,
      "step": 7365
    },
    {
      "epoch": 129.30088495575222,
      "grad_norm": 8.3125,
      "learning_rate": 4.5928540185865776e-05,
      "loss": 0.9739,
      "num_input_tokens_seen": 4089152,
      "step": 7370
    },
    {
      "epoch": 129.38938053097345,
      "grad_norm": 9.5,
      "learning_rate": 4.592316854758648e-05,
      "loss": 1.1298,
      "num_input_tokens_seen": 4091392,
      "step": 7375
    },
    {
      "epoch": 129.47787610619469,
      "grad_norm": 8.3125,
      "learning_rate": 4.5917793682691646e-05,
      "loss": 0.9689,
      "num_input_tokens_seen": 4094192,
      "step": 7380
    },
    {
      "epoch": 129.56637168141592,
      "grad_norm": 9.8125,
      "learning_rate": 4.5912415592010164e-05,
      "loss": 1.2017,
      "num_input_tokens_seen": 4096624,
      "step": 7385
    },
    {
      "epoch": 129.65486725663717,
      "grad_norm": 9.0625,
      "learning_rate": 4.5907034276371386e-05,
      "loss": 0.8317,
      "num_input_tokens_seen": 4099664,
      "step": 7390
    },
    {
      "epoch": 129.7433628318584,
      "grad_norm": 9.375,
      "learning_rate": 4.5901649736605196e-05,
      "loss": 1.2179,
      "num_input_tokens_seen": 4102352,
      "step": 7395
    },
    {
      "epoch": 129.83185840707964,
      "grad_norm": 10.375,
      "learning_rate": 4.589626197354195e-05,
      "loss": 1.0637,
      "num_input_tokens_seen": 4105440,
      "step": 7400
    },
    {
      "epoch": 129.83185840707964,
      "eval_loss": 1.0335756540298462,
      "eval_runtime": 0.9336,
      "eval_samples_per_second": 26.778,
      "eval_steps_per_second": 13.924,
      "num_input_tokens_seen": 4105440,
      "step": 7400
    },
    {
      "epoch": 129.9203539823009,
      "grad_norm": 8.375,
      "learning_rate": 4.5890870988012504e-05,
      "loss": 1.1454,
      "num_input_tokens_seen": 4108528,
      "step": 7405
    },
    {
      "epoch": 130.0,
      "grad_norm": 9.3125,
      "learning_rate": 4.5885476780848226e-05,
      "loss": 0.9433,
      "num_input_tokens_seen": 4110432,
      "step": 7410
    },
    {
      "epoch": 130.08849557522123,
      "grad_norm": 8.375,
      "learning_rate": 4.5880079352880964e-05,
      "loss": 1.046,
      "num_input_tokens_seen": 4113408,
      "step": 7415
    },
    {
      "epoch": 130.1769911504425,
      "grad_norm": 11.9375,
      "learning_rate": 4.5874678704943065e-05,
      "loss": 1.2567,
      "num_input_tokens_seen": 4116288,
      "step": 7420
    },
    {
      "epoch": 130.26548672566372,
      "grad_norm": 9.5,
      "learning_rate": 4.5869274837867394e-05,
      "loss": 0.9218,
      "num_input_tokens_seen": 4119360,
      "step": 7425
    },
    {
      "epoch": 130.35398230088495,
      "grad_norm": 6.875,
      "learning_rate": 4.5863867752487275e-05,
      "loss": 1.0005,
      "num_input_tokens_seen": 4122000,
      "step": 7430
    },
    {
      "epoch": 130.44247787610618,
      "grad_norm": 8.6875,
      "learning_rate": 4.5858457449636554e-05,
      "loss": 0.989,
      "num_input_tokens_seen": 4124624,
      "step": 7435
    },
    {
      "epoch": 130.53097345132744,
      "grad_norm": 8.125,
      "learning_rate": 4.5853043930149574e-05,
      "loss": 1.0245,
      "num_input_tokens_seen": 4127600,
      "step": 7440
    },
    {
      "epoch": 130.61946902654867,
      "grad_norm": 10.25,
      "learning_rate": 4.584762719486117e-05,
      "loss": 1.3015,
      "num_input_tokens_seen": 4129904,
      "step": 7445
    },
    {
      "epoch": 130.7079646017699,
      "grad_norm": 7.875,
      "learning_rate": 4.584220724460665e-05,
      "loss": 1.1124,
      "num_input_tokens_seen": 4132640,
      "step": 7450
    },
    {
      "epoch": 130.79646017699116,
      "grad_norm": 8.9375,
      "learning_rate": 4.5836784080221865e-05,
      "loss": 0.8649,
      "num_input_tokens_seen": 4135840,
      "step": 7455
    },
    {
      "epoch": 130.8849557522124,
      "grad_norm": 6.625,
      "learning_rate": 4.583135770254312e-05,
      "loss": 1.0668,
      "num_input_tokens_seen": 4138384,
      "step": 7460
    },
    {
      "epoch": 130.97345132743362,
      "grad_norm": 8.25,
      "learning_rate": 4.5825928112407236e-05,
      "loss": 1.0897,
      "num_input_tokens_seen": 4141776,
      "step": 7465
    },
    {
      "epoch": 131.05309734513276,
      "grad_norm": 6.8125,
      "learning_rate": 4.582049531065152e-05,
      "loss": 0.9167,
      "num_input_tokens_seen": 4143800,
      "step": 7470
    },
    {
      "epoch": 131.141592920354,
      "grad_norm": 10.6875,
      "learning_rate": 4.5815059298113783e-05,
      "loss": 1.1073,
      "num_input_tokens_seen": 4146728,
      "step": 7475
    },
    {
      "epoch": 131.23008849557522,
      "grad_norm": 10.0,
      "learning_rate": 4.580962007563232e-05,
      "loss": 0.9813,
      "num_input_tokens_seen": 4149896,
      "step": 7480
    },
    {
      "epoch": 131.31858407079645,
      "grad_norm": 8.125,
      "learning_rate": 4.5804177644045935e-05,
      "loss": 1.0976,
      "num_input_tokens_seen": 4152536,
      "step": 7485
    },
    {
      "epoch": 131.4070796460177,
      "grad_norm": 7.40625,
      "learning_rate": 4.579873200419391e-05,
      "loss": 1.1692,
      "num_input_tokens_seen": 4155432,
      "step": 7490
    },
    {
      "epoch": 131.49557522123894,
      "grad_norm": 7.9375,
      "learning_rate": 4.5793283156916046e-05,
      "loss": 1.1141,
      "num_input_tokens_seen": 4158312,
      "step": 7495
    },
    {
      "epoch": 131.58407079646017,
      "grad_norm": 7.5,
      "learning_rate": 4.578783110305261e-05,
      "loss": 1.0379,
      "num_input_tokens_seen": 4160632,
      "step": 7500
    },
    {
      "epoch": 131.67256637168143,
      "grad_norm": 8.125,
      "learning_rate": 4.578237584344438e-05,
      "loss": 0.9519,
      "num_input_tokens_seen": 4163848,
      "step": 7505
    },
    {
      "epoch": 131.76106194690266,
      "grad_norm": 8.5625,
      "learning_rate": 4.577691737893263e-05,
      "loss": 1.1062,
      "num_input_tokens_seen": 4166328,
      "step": 7510
    },
    {
      "epoch": 131.8495575221239,
      "grad_norm": 8.75,
      "learning_rate": 4.577145571035912e-05,
      "loss": 1.0718,
      "num_input_tokens_seen": 4169352,
      "step": 7515
    },
    {
      "epoch": 131.93805309734512,
      "grad_norm": 8.0625,
      "learning_rate": 4.576599083856611e-05,
      "loss": 1.0497,
      "num_input_tokens_seen": 4172248,
      "step": 7520
    },
    {
      "epoch": 132.01769911504425,
      "grad_norm": 8.75,
      "learning_rate": 4.576052276439635e-05,
      "loss": 1.2098,
      "num_input_tokens_seen": 4174432,
      "step": 7525
    },
    {
      "epoch": 132.10619469026548,
      "grad_norm": 8.5625,
      "learning_rate": 4.575505148869308e-05,
      "loss": 0.8771,
      "num_input_tokens_seen": 4177232,
      "step": 7530
    },
    {
      "epoch": 132.1946902654867,
      "grad_norm": 10.1875,
      "learning_rate": 4.574957701230006e-05,
      "loss": 1.1556,
      "num_input_tokens_seen": 4179904,
      "step": 7535
    },
    {
      "epoch": 132.28318584070797,
      "grad_norm": 8.8125,
      "learning_rate": 4.57440993360615e-05,
      "loss": 1.0413,
      "num_input_tokens_seen": 4183104,
      "step": 7540
    },
    {
      "epoch": 132.3716814159292,
      "grad_norm": 8.0625,
      "learning_rate": 4.5738618460822134e-05,
      "loss": 0.8236,
      "num_input_tokens_seen": 4185888,
      "step": 7545
    },
    {
      "epoch": 132.46017699115043,
      "grad_norm": 7.625,
      "learning_rate": 4.573313438742719e-05,
      "loss": 1.1516,
      "num_input_tokens_seen": 4188816,
      "step": 7550
    },
    {
      "epoch": 132.5486725663717,
      "grad_norm": 6.5,
      "learning_rate": 4.5727647116722374e-05,
      "loss": 1.0325,
      "num_input_tokens_seen": 4191744,
      "step": 7555
    },
    {
      "epoch": 132.63716814159292,
      "grad_norm": 8.0625,
      "learning_rate": 4.5722156649553884e-05,
      "loss": 1.0873,
      "num_input_tokens_seen": 4194656,
      "step": 7560
    },
    {
      "epoch": 132.72566371681415,
      "grad_norm": 8.125,
      "learning_rate": 4.571666298676843e-05,
      "loss": 1.1609,
      "num_input_tokens_seen": 4197584,
      "step": 7565
    },
    {
      "epoch": 132.81415929203538,
      "grad_norm": 9.0625,
      "learning_rate": 4.571116612921321e-05,
      "loss": 1.1008,
      "num_input_tokens_seen": 4200288,
      "step": 7570
    },
    {
      "epoch": 132.90265486725664,
      "grad_norm": 7.59375,
      "learning_rate": 4.57056660777359e-05,
      "loss": 0.9125,
      "num_input_tokens_seen": 4202464,
      "step": 7575
    },
    {
      "epoch": 132.99115044247787,
      "grad_norm": 10.0,
      "learning_rate": 4.5700162833184666e-05,
      "loss": 1.1731,
      "num_input_tokens_seen": 4205104,
      "step": 7580
    },
    {
      "epoch": 133.07079646017698,
      "grad_norm": 9.125,
      "learning_rate": 4.5694656396408195e-05,
      "loss": 1.484,
      "num_input_tokens_seen": 4207824,
      "step": 7585
    },
    {
      "epoch": 133.15929203539824,
      "grad_norm": 8.4375,
      "learning_rate": 4.5689146768255646e-05,
      "loss": 1.0613,
      "num_input_tokens_seen": 4210496,
      "step": 7590
    },
    {
      "epoch": 133.24778761061947,
      "grad_norm": 6.625,
      "learning_rate": 4.568363394957667e-05,
      "loss": 1.094,
      "num_input_tokens_seen": 4213072,
      "step": 7595
    },
    {
      "epoch": 133.3362831858407,
      "grad_norm": 11.375,
      "learning_rate": 4.567811794122141e-05,
      "loss": 1.1949,
      "num_input_tokens_seen": 4216208,
      "step": 7600
    },
    {
      "epoch": 133.3362831858407,
      "eval_loss": 1.021116018295288,
      "eval_runtime": 0.9314,
      "eval_samples_per_second": 26.841,
      "eval_steps_per_second": 13.957,
      "num_input_tokens_seen": 4216208,
      "step": 7600
    },
    {
      "epoch": 133.42477876106196,
      "grad_norm": 8.75,
      "learning_rate": 4.56725987440405e-05,
      "loss": 1.1283,
      "num_input_tokens_seen": 4218592,
      "step": 7605
    },
    {
      "epoch": 133.5132743362832,
      "grad_norm": 8.3125,
      "learning_rate": 4.566707635888508e-05,
      "loss": 0.9514,
      "num_input_tokens_seen": 4221424,
      "step": 7610
    },
    {
      "epoch": 133.60176991150442,
      "grad_norm": 9.125,
      "learning_rate": 4.566155078660677e-05,
      "loss": 1.104,
      "num_input_tokens_seen": 4224000,
      "step": 7615
    },
    {
      "epoch": 133.69026548672565,
      "grad_norm": 8.4375,
      "learning_rate": 4.565602202805768e-05,
      "loss": 1.0116,
      "num_input_tokens_seen": 4226432,
      "step": 7620
    },
    {
      "epoch": 133.7787610619469,
      "grad_norm": 6.96875,
      "learning_rate": 4.56504900840904e-05,
      "loss": 1.0367,
      "num_input_tokens_seen": 4229264,
      "step": 7625
    },
    {
      "epoch": 133.86725663716814,
      "grad_norm": 5.40625,
      "learning_rate": 4.564495495555805e-05,
      "loss": 0.909,
      "num_input_tokens_seen": 4232128,
      "step": 7630
    },
    {
      "epoch": 133.95575221238937,
      "grad_norm": 7.5,
      "learning_rate": 4.5639416643314204e-05,
      "loss": 1.0797,
      "num_input_tokens_seen": 4235104,
      "step": 7635
    },
    {
      "epoch": 134.0353982300885,
      "grad_norm": 8.5,
      "learning_rate": 4.5633875148212946e-05,
      "loss": 0.9598,
      "num_input_tokens_seen": 4237576,
      "step": 7640
    },
    {
      "epoch": 134.12389380530973,
      "grad_norm": 6.65625,
      "learning_rate": 4.562833047110883e-05,
      "loss": 1.0131,
      "num_input_tokens_seen": 4240216,
      "step": 7645
    },
    {
      "epoch": 134.21238938053096,
      "grad_norm": 8.0625,
      "learning_rate": 4.5622782612856923e-05,
      "loss": 0.949,
      "num_input_tokens_seen": 4243944,
      "step": 7650
    },
    {
      "epoch": 134.30088495575222,
      "grad_norm": 9.875,
      "learning_rate": 4.561723157431278e-05,
      "loss": 1.0776,
      "num_input_tokens_seen": 4246728,
      "step": 7655
    },
    {
      "epoch": 134.38938053097345,
      "grad_norm": 8.125,
      "learning_rate": 4.5611677356332435e-05,
      "loss": 0.9378,
      "num_input_tokens_seen": 4249640,
      "step": 7660
    },
    {
      "epoch": 134.47787610619469,
      "grad_norm": 8.125,
      "learning_rate": 4.560611995977242e-05,
      "loss": 1.122,
      "num_input_tokens_seen": 4252120,
      "step": 7665
    },
    {
      "epoch": 134.56637168141592,
      "grad_norm": 11.1875,
      "learning_rate": 4.560055938548975e-05,
      "loss": 1.0106,
      "num_input_tokens_seen": 4254616,
      "step": 7670
    },
    {
      "epoch": 134.65486725663717,
      "grad_norm": 9.875,
      "learning_rate": 4.5594995634341944e-05,
      "loss": 1.1183,
      "num_input_tokens_seen": 4257288,
      "step": 7675
    },
    {
      "epoch": 134.7433628318584,
      "grad_norm": 8.25,
      "learning_rate": 4.5589428707187e-05,
      "loss": 0.9874,
      "num_input_tokens_seen": 4259928,
      "step": 7680
    },
    {
      "epoch": 134.83185840707964,
      "grad_norm": 7.28125,
      "learning_rate": 4.55838586048834e-05,
      "loss": 0.9778,
      "num_input_tokens_seen": 4262984,
      "step": 7685
    },
    {
      "epoch": 134.9203539823009,
      "grad_norm": 9.125,
      "learning_rate": 4.557828532829013e-05,
      "loss": 1.2286,
      "num_input_tokens_seen": 4265720,
      "step": 7690
    },
    {
      "epoch": 135.0,
      "grad_norm": 27.0,
      "learning_rate": 4.557270887826667e-05,
      "loss": 1.5944,
      "num_input_tokens_seen": 4268152,
      "step": 7695
    },
    {
      "epoch": 135.08849557522123,
      "grad_norm": 8.1875,
      "learning_rate": 4.556712925567296e-05,
      "loss": 1.0311,
      "num_input_tokens_seen": 4270696,
      "step": 7700
    },
    {
      "epoch": 135.1769911504425,
      "grad_norm": 9.1875,
      "learning_rate": 4.5561546461369454e-05,
      "loss": 1.1238,
      "num_input_tokens_seen": 4273240,
      "step": 7705
    },
    {
      "epoch": 135.26548672566372,
      "grad_norm": 8.125,
      "learning_rate": 4.55559604962171e-05,
      "loss": 1.1587,
      "num_input_tokens_seen": 4275752,
      "step": 7710
    },
    {
      "epoch": 135.35398230088495,
      "grad_norm": 5.78125,
      "learning_rate": 4.55503713610773e-05,
      "loss": 0.9144,
      "num_input_tokens_seen": 4278312,
      "step": 7715
    },
    {
      "epoch": 135.44247787610618,
      "grad_norm": 7.40625,
      "learning_rate": 4.5544779056812e-05,
      "loss": 0.9039,
      "num_input_tokens_seen": 4281400,
      "step": 7720
    },
    {
      "epoch": 135.53097345132744,
      "grad_norm": 9.0,
      "learning_rate": 4.553918358428358e-05,
      "loss": 1.0937,
      "num_input_tokens_seen": 4284136,
      "step": 7725
    },
    {
      "epoch": 135.61946902654867,
      "grad_norm": 7.875,
      "learning_rate": 4.553358494435494e-05,
      "loss": 1.1453,
      "num_input_tokens_seen": 4287416,
      "step": 7730
    },
    {
      "epoch": 135.7079646017699,
      "grad_norm": 12.875,
      "learning_rate": 4.5527983137889464e-05,
      "loss": 0.9571,
      "num_input_tokens_seen": 4290696,
      "step": 7735
    },
    {
      "epoch": 135.79646017699116,
      "grad_norm": 6.15625,
      "learning_rate": 4.5522378165751015e-05,
      "loss": 1.1462,
      "num_input_tokens_seen": 4293960,
      "step": 7740
    },
    {
      "epoch": 135.8849557522124,
      "grad_norm": 8.6875,
      "learning_rate": 4.5516770028803954e-05,
      "loss": 1.0166,
      "num_input_tokens_seen": 4296840,
      "step": 7745
    },
    {
      "epoch": 135.97345132743362,
      "grad_norm": 7.21875,
      "learning_rate": 4.5511158727913116e-05,
      "loss": 1.0119,
      "num_input_tokens_seen": 4299528,
      "step": 7750
    },
    {
      "epoch": 136.05309734513276,
      "grad_norm": 9.5,
      "learning_rate": 4.5505544263943856e-05,
      "loss": 1.1843,
      "num_input_tokens_seen": 4301648,
      "step": 7755
    },
    {
      "epoch": 136.141592920354,
      "grad_norm": 6.78125,
      "learning_rate": 4.549992663776197e-05,
      "loss": 0.9288,
      "num_input_tokens_seen": 4304096,
      "step": 7760
    },
    {
      "epoch": 136.23008849557522,
      "grad_norm": 8.9375,
      "learning_rate": 4.5494305850233786e-05,
      "loss": 1.2173,
      "num_input_tokens_seen": 4306528,
      "step": 7765
    },
    {
      "epoch": 136.31858407079645,
      "grad_norm": 8.125,
      "learning_rate": 4.5488681902226094e-05,
      "loss": 0.9613,
      "num_input_tokens_seen": 4309024,
      "step": 7770
    },
    {
      "epoch": 136.4070796460177,
      "grad_norm": 7.375,
      "learning_rate": 4.5483054794606174e-05,
      "loss": 1.0374,
      "num_input_tokens_seen": 4311312,
      "step": 7775
    },
    {
      "epoch": 136.49557522123894,
      "grad_norm": 9.625,
      "learning_rate": 4.547742452824179e-05,
      "loss": 0.9699,
      "num_input_tokens_seen": 4314480,
      "step": 7780
    },
    {
      "epoch": 136.58407079646017,
      "grad_norm": 8.5,
      "learning_rate": 4.5471791104001215e-05,
      "loss": 0.9362,
      "num_input_tokens_seen": 4317792,
      "step": 7785
    },
    {
      "epoch": 136.67256637168143,
      "grad_norm": 7.0625,
      "learning_rate": 4.546615452275319e-05,
      "loss": 1.0968,
      "num_input_tokens_seen": 4320976,
      "step": 7790
    },
    {
      "epoch": 136.76106194690266,
      "grad_norm": 6.78125,
      "learning_rate": 4.5460514785366944e-05,
      "loss": 0.9683,
      "num_input_tokens_seen": 4323888,
      "step": 7795
    },
    {
      "epoch": 136.8495575221239,
      "grad_norm": 6.84375,
      "learning_rate": 4.545487189271219e-05,
      "loss": 1.1562,
      "num_input_tokens_seen": 4326832,
      "step": 7800
    },
    {
      "epoch": 136.8495575221239,
      "eval_loss": 1.0274169445037842,
      "eval_runtime": 0.9332,
      "eval_samples_per_second": 26.79,
      "eval_steps_per_second": 13.931,
      "num_input_tokens_seen": 4326832,
      "step": 7800
    },
    {
      "epoch": 136.93805309734512,
      "grad_norm": 8.0,
      "learning_rate": 4.544922584565914e-05,
      "loss": 1.1367,
      "num_input_tokens_seen": 4329808,
      "step": 7805
    },
    {
      "epoch": 137.01769911504425,
      "grad_norm": 9.5625,
      "learning_rate": 4.544357664507848e-05,
      "loss": 1.2426,
      "num_input_tokens_seen": 4332280,
      "step": 7810
    },
    {
      "epoch": 137.10619469026548,
      "grad_norm": 9.3125,
      "learning_rate": 4.54379242918414e-05,
      "loss": 0.9969,
      "num_input_tokens_seen": 4334872,
      "step": 7815
    },
    {
      "epoch": 137.1946902654867,
      "grad_norm": 7.03125,
      "learning_rate": 4.543226878681955e-05,
      "loss": 1.0117,
      "num_input_tokens_seen": 4337672,
      "step": 7820
    },
    {
      "epoch": 137.28318584070797,
      "grad_norm": 10.125,
      "learning_rate": 4.5426610130885087e-05,
      "loss": 1.1727,
      "num_input_tokens_seen": 4340280,
      "step": 7825
    },
    {
      "epoch": 137.3716814159292,
      "grad_norm": 8.25,
      "learning_rate": 4.542094832491064e-05,
      "loss": 0.9605,
      "num_input_tokens_seen": 4342696,
      "step": 7830
    },
    {
      "epoch": 137.46017699115043,
      "grad_norm": 7.6875,
      "learning_rate": 4.541528336976934e-05,
      "loss": 1.2284,
      "num_input_tokens_seen": 4345144,
      "step": 7835
    },
    {
      "epoch": 137.5486725663717,
      "grad_norm": 9.8125,
      "learning_rate": 4.540961526633479e-05,
      "loss": 1.0733,
      "num_input_tokens_seen": 4348216,
      "step": 7840
    },
    {
      "epoch": 137.63716814159292,
      "grad_norm": 7.875,
      "learning_rate": 4.540394401548108e-05,
      "loss": 1.0107,
      "num_input_tokens_seen": 4351304,
      "step": 7845
    },
    {
      "epoch": 137.72566371681415,
      "grad_norm": 9.0,
      "learning_rate": 4.539826961808279e-05,
      "loss": 1.0143,
      "num_input_tokens_seen": 4354472,
      "step": 7850
    },
    {
      "epoch": 137.81415929203538,
      "grad_norm": 6.8125,
      "learning_rate": 4.5392592075014994e-05,
      "loss": 0.9503,
      "num_input_tokens_seen": 4357240,
      "step": 7855
    },
    {
      "epoch": 137.90265486725664,
      "grad_norm": 11.8125,
      "learning_rate": 4.538691138715322e-05,
      "loss": 1.1574,
      "num_input_tokens_seen": 4359832,
      "step": 7860
    },
    {
      "epoch": 137.99115044247787,
      "grad_norm": 8.625,
      "learning_rate": 4.5381227555373516e-05,
      "loss": 1.0927,
      "num_input_tokens_seen": 4362520,
      "step": 7865
    },
    {
      "epoch": 138.07079646017698,
      "grad_norm": 6.5,
      "learning_rate": 4.537554058055239e-05,
      "loss": 0.854,
      "num_input_tokens_seen": 4365072,
      "step": 7870
    },
    {
      "epoch": 138.15929203539824,
      "grad_norm": 8.125,
      "learning_rate": 4.5369850463566865e-05,
      "loss": 1.1269,
      "num_input_tokens_seen": 4367376,
      "step": 7875
    },
    {
      "epoch": 138.24778761061947,
      "grad_norm": 5.59375,
      "learning_rate": 4.5364157205294404e-05,
      "loss": 0.8639,
      "num_input_tokens_seen": 4370112,
      "step": 7880
    },
    {
      "epoch": 138.3362831858407,
      "grad_norm": 7.90625,
      "learning_rate": 4.5358460806612996e-05,
      "loss": 0.9562,
      "num_input_tokens_seen": 4373008,
      "step": 7885
    },
    {
      "epoch": 138.42477876106196,
      "grad_norm": 10.3125,
      "learning_rate": 4.535276126840109e-05,
      "loss": 1.0903,
      "num_input_tokens_seen": 4375824,
      "step": 7890
    },
    {
      "epoch": 138.5132743362832,
      "grad_norm": 6.625,
      "learning_rate": 4.5347058591537626e-05,
      "loss": 0.96,
      "num_input_tokens_seen": 4378656,
      "step": 7895
    },
    {
      "epoch": 138.60176991150442,
      "grad_norm": 9.1875,
      "learning_rate": 4.534135277690203e-05,
      "loss": 0.9885,
      "num_input_tokens_seen": 4381344,
      "step": 7900
    },
    {
      "epoch": 138.69026548672565,
      "grad_norm": 8.375,
      "learning_rate": 4.533564382537421e-05,
      "loss": 1.2348,
      "num_input_tokens_seen": 4384128,
      "step": 7905
    },
    {
      "epoch": 138.7787610619469,
      "grad_norm": 9.1875,
      "learning_rate": 4.532993173783456e-05,
      "loss": 1.1063,
      "num_input_tokens_seen": 4387408,
      "step": 7910
    },
    {
      "epoch": 138.86725663716814,
      "grad_norm": 9.125,
      "learning_rate": 4.5324216515163954e-05,
      "loss": 1.2386,
      "num_input_tokens_seen": 4390144,
      "step": 7915
    },
    {
      "epoch": 138.95575221238937,
      "grad_norm": 9.25,
      "learning_rate": 4.531849815824375e-05,
      "loss": 1.0609,
      "num_input_tokens_seen": 4392800,
      "step": 7920
    },
    {
      "epoch": 139.0353982300885,
      "grad_norm": 8.25,
      "learning_rate": 4.5312776667955795e-05,
      "loss": 0.8449,
      "num_input_tokens_seen": 4395144,
      "step": 7925
    },
    {
      "epoch": 139.12389380530973,
      "grad_norm": 8.25,
      "learning_rate": 4.5307052045182405e-05,
      "loss": 1.0545,
      "num_input_tokens_seen": 4398216,
      "step": 7930
    },
    {
      "epoch": 139.21238938053096,
      "grad_norm": 5.9375,
      "learning_rate": 4.53013242908064e-05,
      "loss": 0.9166,
      "num_input_tokens_seen": 4400824,
      "step": 7935
    },
    {
      "epoch": 139.30088495575222,
      "grad_norm": 7.625,
      "learning_rate": 4.529559340571107e-05,
      "loss": 0.8962,
      "num_input_tokens_seen": 4403400,
      "step": 7940
    },
    {
      "epoch": 139.38938053097345,
      "grad_norm": 8.3125,
      "learning_rate": 4.528985939078018e-05,
      "loss": 1.1276,
      "num_input_tokens_seen": 4406136,
      "step": 7945
    },
    {
      "epoch": 139.47787610619469,
      "grad_norm": 7.0625,
      "learning_rate": 4.5284122246898e-05,
      "loss": 1.1588,
      "num_input_tokens_seen": 4409368,
      "step": 7950
    },
    {
      "epoch": 139.56637168141592,
      "grad_norm": 8.5,
      "learning_rate": 4.527838197494926e-05,
      "loss": 1.0227,
      "num_input_tokens_seen": 4412264,
      "step": 7955
    },
    {
      "epoch": 139.65486725663717,
      "grad_norm": 8.5625,
      "learning_rate": 4.527263857581918e-05,
      "loss": 1.1729,
      "num_input_tokens_seen": 4415080,
      "step": 7960
    },
    {
      "epoch": 139.7433628318584,
      "grad_norm": 8.8125,
      "learning_rate": 4.526689205039347e-05,
      "loss": 1.3654,
      "num_input_tokens_seen": 4417816,
      "step": 7965
    },
    {
      "epoch": 139.83185840707964,
      "grad_norm": 8.875,
      "learning_rate": 4.5261142399558324e-05,
      "loss": 1.0194,
      "num_input_tokens_seen": 4420600,
      "step": 7970
    },
    {
      "epoch": 139.9203539823009,
      "grad_norm": 9.25,
      "learning_rate": 4.525538962420041e-05,
      "loss": 1.0164,
      "num_input_tokens_seen": 4423416,
      "step": 7975
    },
    {
      "epoch": 140.0,
      "grad_norm": 3.84375,
      "learning_rate": 4.524963372520685e-05,
      "loss": 0.7718,
      "num_input_tokens_seen": 4425808,
      "step": 7980
    },
    {
      "epoch": 140.08849557522123,
      "grad_norm": 11.5,
      "learning_rate": 4.524387470346531e-05,
      "loss": 0.9928,
      "num_input_tokens_seen": 4429056,
      "step": 7985
    },
    {
      "epoch": 140.1769911504425,
      "grad_norm": 7.65625,
      "learning_rate": 4.5238112559863885e-05,
      "loss": 0.6874,
      "num_input_tokens_seen": 4431840,
      "step": 7990
    },
    {
      "epoch": 140.26548672566372,
      "grad_norm": 7.59375,
      "learning_rate": 4.5232347295291175e-05,
      "loss": 0.9644,
      "num_input_tokens_seen": 4435072,
      "step": 7995
    },
    {
      "epoch": 140.35398230088495,
      "grad_norm": 7.96875,
      "learning_rate": 4.522657891063626e-05,
      "loss": 1.137,
      "num_input_tokens_seen": 4437792,
      "step": 8000
    },
    {
      "epoch": 140.35398230088495,
      "eval_loss": 1.0348681211471558,
      "eval_runtime": 0.9322,
      "eval_samples_per_second": 26.817,
      "eval_steps_per_second": 13.945,
      "num_input_tokens_seen": 4437792,
      "step": 8000
    },
    {
      "epoch": 140.44247787610618,
      "grad_norm": 9.3125,
      "learning_rate": 4.52208074067887e-05,
      "loss": 1.2311,
      "num_input_tokens_seen": 4440432,
      "step": 8005
    },
    {
      "epoch": 140.53097345132744,
      "grad_norm": 8.0625,
      "learning_rate": 4.5215032784638516e-05,
      "loss": 1.3774,
      "num_input_tokens_seen": 4443392,
      "step": 8010
    },
    {
      "epoch": 140.61946902654867,
      "grad_norm": 8.3125,
      "learning_rate": 4.5209255045076245e-05,
      "loss": 0.9833,
      "num_input_tokens_seen": 4445856,
      "step": 8015
    },
    {
      "epoch": 140.7079646017699,
      "grad_norm": 10.3125,
      "learning_rate": 4.5203474188992875e-05,
      "loss": 1.0781,
      "num_input_tokens_seen": 4448464,
      "step": 8020
    },
    {
      "epoch": 140.79646017699116,
      "grad_norm": 8.0625,
      "learning_rate": 4.51976902172799e-05,
      "loss": 1.0798,
      "num_input_tokens_seen": 4450864,
      "step": 8025
    },
    {
      "epoch": 140.8849557522124,
      "grad_norm": 7.46875,
      "learning_rate": 4.519190313082927e-05,
      "loss": 0.982,
      "num_input_tokens_seen": 4454080,
      "step": 8030
    },
    {
      "epoch": 140.97345132743362,
      "grad_norm": 8.9375,
      "learning_rate": 4.518611293053343e-05,
      "loss": 1.1683,
      "num_input_tokens_seen": 4456784,
      "step": 8035
    },
    {
      "epoch": 141.05309734513276,
      "grad_norm": 7.46875,
      "learning_rate": 4.51803196172853e-05,
      "loss": 0.9662,
      "num_input_tokens_seen": 4458968,
      "step": 8040
    },
    {
      "epoch": 141.141592920354,
      "grad_norm": 6.625,
      "learning_rate": 4.517452319197828e-05,
      "loss": 0.9935,
      "num_input_tokens_seen": 4461736,
      "step": 8045
    },
    {
      "epoch": 141.23008849557522,
      "grad_norm": 7.875,
      "learning_rate": 4.5168723655506265e-05,
      "loss": 1.126,
      "num_input_tokens_seen": 4464648,
      "step": 8050
    },
    {
      "epoch": 141.31858407079645,
      "grad_norm": 8.8125,
      "learning_rate": 4.51629210087636e-05,
      "loss": 1.2376,
      "num_input_tokens_seen": 4467336,
      "step": 8055
    },
    {
      "epoch": 141.4070796460177,
      "grad_norm": 7.78125,
      "learning_rate": 4.515711525264513e-05,
      "loss": 1.0131,
      "num_input_tokens_seen": 4470280,
      "step": 8060
    },
    {
      "epoch": 141.49557522123894,
      "grad_norm": 9.8125,
      "learning_rate": 4.5151306388046175e-05,
      "loss": 0.9995,
      "num_input_tokens_seen": 4473320,
      "step": 8065
    },
    {
      "epoch": 141.58407079646017,
      "grad_norm": 8.375,
      "learning_rate": 4.514549441586255e-05,
      "loss": 0.8104,
      "num_input_tokens_seen": 4476392,
      "step": 8070
    },
    {
      "epoch": 141.67256637168143,
      "grad_norm": 6.46875,
      "learning_rate": 4.513967933699051e-05,
      "loss": 0.9201,
      "num_input_tokens_seen": 4478968,
      "step": 8075
    },
    {
      "epoch": 141.76106194690266,
      "grad_norm": 8.0,
      "learning_rate": 4.513386115232684e-05,
      "loss": 1.0355,
      "num_input_tokens_seen": 4481528,
      "step": 8080
    },
    {
      "epoch": 141.8495575221239,
      "grad_norm": 8.9375,
      "learning_rate": 4.5128039862768745e-05,
      "loss": 1.0623,
      "num_input_tokens_seen": 4484744,
      "step": 8085
    },
    {
      "epoch": 141.93805309734512,
      "grad_norm": 11.0,
      "learning_rate": 4.512221546921397e-05,
      "loss": 1.3682,
      "num_input_tokens_seen": 4487480,
      "step": 8090
    },
    {
      "epoch": 142.01769911504425,
      "grad_norm": 7.0625,
      "learning_rate": 4.5116387972560694e-05,
      "loss": 0.9266,
      "num_input_tokens_seen": 4490048,
      "step": 8095
    },
    {
      "epoch": 142.10619469026548,
      "grad_norm": 14.8125,
      "learning_rate": 4.511055737370759e-05,
      "loss": 1.2963,
      "num_input_tokens_seen": 4493152,
      "step": 8100
    },
    {
      "epoch": 142.1946902654867,
      "grad_norm": 6.90625,
      "learning_rate": 4.510472367355383e-05,
      "loss": 1.012,
      "num_input_tokens_seen": 4495664,
      "step": 8105
    },
    {
      "epoch": 142.28318584070797,
      "grad_norm": 6.65625,
      "learning_rate": 4.509888687299901e-05,
      "loss": 0.9076,
      "num_input_tokens_seen": 4498112,
      "step": 8110
    },
    {
      "epoch": 142.3716814159292,
      "grad_norm": 7.84375,
      "learning_rate": 4.5093046972943266e-05,
      "loss": 0.9962,
      "num_input_tokens_seen": 4501152,
      "step": 8115
    },
    {
      "epoch": 142.46017699115043,
      "grad_norm": 8.25,
      "learning_rate": 4.508720397428717e-05,
      "loss": 1.095,
      "num_input_tokens_seen": 4503984,
      "step": 8120
    },
    {
      "epoch": 142.5486725663717,
      "grad_norm": 9.6875,
      "learning_rate": 4.508135787793178e-05,
      "loss": 1.0529,
      "num_input_tokens_seen": 4506576,
      "step": 8125
    },
    {
      "epoch": 142.63716814159292,
      "grad_norm": 7.8125,
      "learning_rate": 4.5075508684778664e-05,
      "loss": 1.0438,
      "num_input_tokens_seen": 4508928,
      "step": 8130
    },
    {
      "epoch": 142.72566371681415,
      "grad_norm": 7.75,
      "learning_rate": 4.506965639572982e-05,
      "loss": 1.0189,
      "num_input_tokens_seen": 4512000,
      "step": 8135
    },
    {
      "epoch": 142.81415929203538,
      "grad_norm": 9.5625,
      "learning_rate": 4.506380101168774e-05,
      "loss": 0.9936,
      "num_input_tokens_seen": 4515248,
      "step": 8140
    },
    {
      "epoch": 142.90265486725664,
      "grad_norm": 9.25,
      "learning_rate": 4.505794253355542e-05,
      "loss": 1.0804,
      "num_input_tokens_seen": 4518608,
      "step": 8145
    },
    {
      "epoch": 142.99115044247787,
      "grad_norm": 8.1875,
      "learning_rate": 4.5052080962236286e-05,
      "loss": 1.174,
      "num_input_tokens_seen": 4521104,
      "step": 8150
    },
    {
      "epoch": 143.07079646017698,
      "grad_norm": 7.125,
      "learning_rate": 4.504621629863428e-05,
      "loss": 1.0661,
      "num_input_tokens_seen": 4523544,
      "step": 8155
    },
    {
      "epoch": 143.15929203539824,
      "grad_norm": 8.3125,
      "learning_rate": 4.504034854365381e-05,
      "loss": 0.9285,
      "num_input_tokens_seen": 4526696,
      "step": 8160
    },
    {
      "epoch": 143.24778761061947,
      "grad_norm": 11.0,
      "learning_rate": 4.503447769819974e-05,
      "loss": 1.1801,
      "num_input_tokens_seen": 4529512,
      "step": 8165
    },
    {
      "epoch": 143.3362831858407,
      "grad_norm": 7.375,
      "learning_rate": 4.502860376317745e-05,
      "loss": 1.2118,
      "num_input_tokens_seen": 4532248,
      "step": 8170
    },
    {
      "epoch": 143.42477876106196,
      "grad_norm": 11.3125,
      "learning_rate": 4.502272673949276e-05,
      "loss": 0.9726,
      "num_input_tokens_seen": 4535224,
      "step": 8175
    },
    {
      "epoch": 143.5132743362832,
      "grad_norm": 9.25,
      "learning_rate": 4.501684662805199e-05,
      "loss": 1.0166,
      "num_input_tokens_seen": 4537944,
      "step": 8180
    },
    {
      "epoch": 143.60176991150442,
      "grad_norm": 11.9375,
      "learning_rate": 4.5010963429761924e-05,
      "loss": 1.0819,
      "num_input_tokens_seen": 4540552,
      "step": 8185
    },
    {
      "epoch": 143.69026548672565,
      "grad_norm": 8.6875,
      "learning_rate": 4.500507714552982e-05,
      "loss": 1.0867,
      "num_input_tokens_seen": 4543416,
      "step": 8190
    },
    {
      "epoch": 143.7787610619469,
      "grad_norm": 6.59375,
      "learning_rate": 4.499918777626342e-05,
      "loss": 0.9332,
      "num_input_tokens_seen": 4546472,
      "step": 8195
    },
    {
      "epoch": 143.86725663716814,
      "grad_norm": 9.25,
      "learning_rate": 4.499329532287093e-05,
      "loss": 1.1652,
      "num_input_tokens_seen": 4549512,
      "step": 8200
    },
    {
      "epoch": 143.86725663716814,
      "eval_loss": 1.0362967252731323,
      "eval_runtime": 0.9353,
      "eval_samples_per_second": 26.73,
      "eval_steps_per_second": 13.9,
      "num_input_tokens_seen": 4549512,
      "step": 8200
    },
    {
      "epoch": 143.95575221238937,
      "grad_norm": 6.9375,
      "learning_rate": 4.4987399786261064e-05,
      "loss": 1.0291,
      "num_input_tokens_seen": 4551880,
      "step": 8205
    },
    {
      "epoch": 144.0353982300885,
      "grad_norm": 9.4375,
      "learning_rate": 4.498150116734297e-05,
      "loss": 0.9747,
      "num_input_tokens_seen": 4553928,
      "step": 8210
    },
    {
      "epoch": 144.12389380530973,
      "grad_norm": 8.25,
      "learning_rate": 4.4975599467026294e-05,
      "loss": 1.1623,
      "num_input_tokens_seen": 4556184,
      "step": 8215
    },
    {
      "epoch": 144.21238938053096,
      "grad_norm": 12.9375,
      "learning_rate": 4.496969468622114e-05,
      "loss": 0.8754,
      "num_input_tokens_seen": 4559112,
      "step": 8220
    },
    {
      "epoch": 144.30088495575222,
      "grad_norm": 9.1875,
      "learning_rate": 4.496378682583813e-05,
      "loss": 0.9861,
      "num_input_tokens_seen": 4562280,
      "step": 8225
    },
    {
      "epoch": 144.38938053097345,
      "grad_norm": 7.75,
      "learning_rate": 4.495787588678829e-05,
      "loss": 1.0546,
      "num_input_tokens_seen": 4565256,
      "step": 8230
    },
    {
      "epoch": 144.47787610619469,
      "grad_norm": 8.75,
      "learning_rate": 4.4951961869983196e-05,
      "loss": 1.0254,
      "num_input_tokens_seen": 4568056,
      "step": 8235
    },
    {
      "epoch": 144.56637168141592,
      "grad_norm": 8.0625,
      "learning_rate": 4.494604477633485e-05,
      "loss": 1.0436,
      "num_input_tokens_seen": 4570872,
      "step": 8240
    },
    {
      "epoch": 144.65486725663717,
      "grad_norm": 9.4375,
      "learning_rate": 4.4940124606755734e-05,
      "loss": 1.1782,
      "num_input_tokens_seen": 4573336,
      "step": 8245
    },
    {
      "epoch": 144.7433628318584,
      "grad_norm": 9.4375,
      "learning_rate": 4.493420136215882e-05,
      "loss": 0.9933,
      "num_input_tokens_seen": 4576360,
      "step": 8250
    },
    {
      "epoch": 144.83185840707964,
      "grad_norm": 7.3125,
      "learning_rate": 4.492827504345756e-05,
      "loss": 0.9338,
      "num_input_tokens_seen": 4578920,
      "step": 8255
    },
    {
      "epoch": 144.9203539823009,
      "grad_norm": 8.4375,
      "learning_rate": 4.492234565156584e-05,
      "loss": 1.051,
      "num_input_tokens_seen": 4582232,
      "step": 8260
    },
    {
      "epoch": 145.0,
      "grad_norm": 9.875,
      "learning_rate": 4.491641318739807e-05,
      "loss": 1.3114,
      "num_input_tokens_seen": 4584632,
      "step": 8265
    },
    {
      "epoch": 145.08849557522123,
      "grad_norm": 8.3125,
      "learning_rate": 4.4910477651869096e-05,
      "loss": 1.1996,
      "num_input_tokens_seen": 4587688,
      "step": 8270
    },
    {
      "epoch": 145.1769911504425,
      "grad_norm": 8.3125,
      "learning_rate": 4.4904539045894254e-05,
      "loss": 1.0954,
      "num_input_tokens_seen": 4590088,
      "step": 8275
    },
    {
      "epoch": 145.26548672566372,
      "grad_norm": 8.5625,
      "learning_rate": 4.4898597370389364e-05,
      "loss": 0.9155,
      "num_input_tokens_seen": 4593064,
      "step": 8280
    },
    {
      "epoch": 145.35398230088495,
      "grad_norm": 9.6875,
      "learning_rate": 4.489265262627069e-05,
      "loss": 1.1193,
      "num_input_tokens_seen": 4595560,
      "step": 8285
    },
    {
      "epoch": 145.44247787610618,
      "grad_norm": 6.5,
      "learning_rate": 4.488670481445499e-05,
      "loss": 1.0446,
      "num_input_tokens_seen": 4598488,
      "step": 8290
    },
    {
      "epoch": 145.53097345132744,
      "grad_norm": 7.59375,
      "learning_rate": 4.488075393585951e-05,
      "loss": 1.0446,
      "num_input_tokens_seen": 4601352,
      "step": 8295
    },
    {
      "epoch": 145.61946902654867,
      "grad_norm": 9.625,
      "learning_rate": 4.487479999140193e-05,
      "loss": 1.0986,
      "num_input_tokens_seen": 4603896,
      "step": 8300
    },
    {
      "epoch": 145.7079646017699,
      "grad_norm": 7.21875,
      "learning_rate": 4.4868842982000425e-05,
      "loss": 1.0932,
      "num_input_tokens_seen": 4606520,
      "step": 8305
    },
    {
      "epoch": 145.79646017699116,
      "grad_norm": 9.875,
      "learning_rate": 4.486288290857365e-05,
      "loss": 1.1317,
      "num_input_tokens_seen": 4609688,
      "step": 8310
    },
    {
      "epoch": 145.8849557522124,
      "grad_norm": 10.6875,
      "learning_rate": 4.4856919772040715e-05,
      "loss": 1.0158,
      "num_input_tokens_seen": 4612456,
      "step": 8315
    },
    {
      "epoch": 145.97345132743362,
      "grad_norm": 6.71875,
      "learning_rate": 4.485095357332122e-05,
      "loss": 0.8609,
      "num_input_tokens_seen": 4615288,
      "step": 8320
    },
    {
      "epoch": 146.05309734513276,
      "grad_norm": 8.3125,
      "learning_rate": 4.484498431333521e-05,
      "loss": 1.0003,
      "num_input_tokens_seen": 4617448,
      "step": 8325
    },
    {
      "epoch": 146.141592920354,
      "grad_norm": 7.09375,
      "learning_rate": 4.4839011993003245e-05,
      "loss": 1.0466,
      "num_input_tokens_seen": 4620344,
      "step": 8330
    },
    {
      "epoch": 146.23008849557522,
      "grad_norm": 10.3125,
      "learning_rate": 4.4833036613246305e-05,
      "loss": 1.1028,
      "num_input_tokens_seen": 4623080,
      "step": 8335
    },
    {
      "epoch": 146.31858407079645,
      "grad_norm": 7.28125,
      "learning_rate": 4.482705817498589e-05,
      "loss": 0.9081,
      "num_input_tokens_seen": 4626072,
      "step": 8340
    },
    {
      "epoch": 146.4070796460177,
      "grad_norm": 9.625,
      "learning_rate": 4.4821076679143934e-05,
      "loss": 1.0189,
      "num_input_tokens_seen": 4628952,
      "step": 8345
    },
    {
      "epoch": 146.49557522123894,
      "grad_norm": 8.75,
      "learning_rate": 4.481509212664288e-05,
      "loss": 0.9489,
      "num_input_tokens_seen": 4631368,
      "step": 8350
    },
    {
      "epoch": 146.58407079646017,
      "grad_norm": 7.5,
      "learning_rate": 4.480910451840559e-05,
      "loss": 1.1694,
      "num_input_tokens_seen": 4633864,
      "step": 8355
    },
    {
      "epoch": 146.67256637168143,
      "grad_norm": 7.0,
      "learning_rate": 4.480311385535546e-05,
      "loss": 1.0304,
      "num_input_tokens_seen": 4636936,
      "step": 8360
    },
    {
      "epoch": 146.76106194690266,
      "grad_norm": 7.375,
      "learning_rate": 4.47971201384163e-05,
      "loss": 0.9697,
      "num_input_tokens_seen": 4639912,
      "step": 8365
    },
    {
      "epoch": 146.8495575221239,
      "grad_norm": 8.5625,
      "learning_rate": 4.4791123368512446e-05,
      "loss": 1.1011,
      "num_input_tokens_seen": 4642872,
      "step": 8370
    },
    {
      "epoch": 146.93805309734512,
      "grad_norm": 9.3125,
      "learning_rate": 4.478512354656864e-05,
      "loss": 1.0393,
      "num_input_tokens_seen": 4645688,
      "step": 8375
    },
    {
      "epoch": 147.01769911504425,
      "grad_norm": 6.90625,
      "learning_rate": 4.477912067351016e-05,
      "loss": 1.2466,
      "num_input_tokens_seen": 4648000,
      "step": 8380
    },
    {
      "epoch": 147.10619469026548,
      "grad_norm": 7.53125,
      "learning_rate": 4.477311475026271e-05,
      "loss": 1.0621,
      "num_input_tokens_seen": 4650624,
      "step": 8385
    },
    {
      "epoch": 147.1946902654867,
      "grad_norm": 8.5,
      "learning_rate": 4.476710577775248e-05,
      "loss": 1.0043,
      "num_input_tokens_seen": 4653536,
      "step": 8390
    },
    {
      "epoch": 147.28318584070797,
      "grad_norm": 12.5625,
      "learning_rate": 4.476109375690612e-05,
      "loss": 0.994,
      "num_input_tokens_seen": 4656336,
      "step": 8395
    },
    {
      "epoch": 147.3716814159292,
      "grad_norm": 7.0625,
      "learning_rate": 4.4755078688650784e-05,
      "loss": 1.1128,
      "num_input_tokens_seen": 4658800,
      "step": 8400
    },
    {
      "epoch": 147.3716814159292,
      "eval_loss": 1.0290031433105469,
      "eval_runtime": 0.9361,
      "eval_samples_per_second": 26.708,
      "eval_steps_per_second": 13.888,
      "num_input_tokens_seen": 4658800,
      "step": 8400
    },
    {
      "epoch": 147.46017699115043,
      "grad_norm": 11.4375,
      "learning_rate": 4.474906057391406e-05,
      "loss": 1.0928,
      "num_input_tokens_seen": 4661424,
      "step": 8405
    },
    {
      "epoch": 147.5486725663717,
      "grad_norm": 7.6875,
      "learning_rate": 4.4743039413624e-05,
      "loss": 1.0076,
      "num_input_tokens_seen": 4664128,
      "step": 8410
    },
    {
      "epoch": 147.63716814159292,
      "grad_norm": 9.25,
      "learning_rate": 4.473701520870916e-05,
      "loss": 1.1862,
      "num_input_tokens_seen": 4666816,
      "step": 8415
    },
    {
      "epoch": 147.72566371681415,
      "grad_norm": 8.625,
      "learning_rate": 4.4730987960098544e-05,
      "loss": 1.0441,
      "num_input_tokens_seen": 4669408,
      "step": 8420
    },
    {
      "epoch": 147.81415929203538,
      "grad_norm": 6.46875,
      "learning_rate": 4.4724957668721635e-05,
      "loss": 1.0263,
      "num_input_tokens_seen": 4672176,
      "step": 8425
    },
    {
      "epoch": 147.90265486725664,
      "grad_norm": 8.375,
      "learning_rate": 4.471892433550836e-05,
      "loss": 0.9469,
      "num_input_tokens_seen": 4674880,
      "step": 8430
    },
    {
      "epoch": 147.99115044247787,
      "grad_norm": 7.53125,
      "learning_rate": 4.471288796138916e-05,
      "loss": 1.1662,
      "num_input_tokens_seen": 4678080,
      "step": 8435
    },
    {
      "epoch": 148.07079646017698,
      "grad_norm": 8.75,
      "learning_rate": 4.470684854729491e-05,
      "loss": 0.9597,
      "num_input_tokens_seen": 4680400,
      "step": 8440
    },
    {
      "epoch": 148.15929203539824,
      "grad_norm": 9.75,
      "learning_rate": 4.4700806094156955e-05,
      "loss": 1.0583,
      "num_input_tokens_seen": 4683536,
      "step": 8445
    },
    {
      "epoch": 148.24778761061947,
      "grad_norm": 7.1875,
      "learning_rate": 4.469476060290713e-05,
      "loss": 1.0337,
      "num_input_tokens_seen": 4685872,
      "step": 8450
    },
    {
      "epoch": 148.3362831858407,
      "grad_norm": 9.9375,
      "learning_rate": 4.468871207447772e-05,
      "loss": 1.1014,
      "num_input_tokens_seen": 4688240,
      "step": 8455
    },
    {
      "epoch": 148.42477876106196,
      "grad_norm": 10.1875,
      "learning_rate": 4.4682660509801486e-05,
      "loss": 1.0409,
      "num_input_tokens_seen": 4690992,
      "step": 8460
    },
    {
      "epoch": 148.5132743362832,
      "grad_norm": 11.25,
      "learning_rate": 4.467660590981165e-05,
      "loss": 1.1835,
      "num_input_tokens_seen": 4693664,
      "step": 8465
    },
    {
      "epoch": 148.60176991150442,
      "grad_norm": 9.1875,
      "learning_rate": 4.467054827544191e-05,
      "loss": 1.1204,
      "num_input_tokens_seen": 4696400,
      "step": 8470
    },
    {
      "epoch": 148.69026548672565,
      "grad_norm": 8.0625,
      "learning_rate": 4.4664487607626434e-05,
      "loss": 1.1971,
      "num_input_tokens_seen": 4699568,
      "step": 8475
    },
    {
      "epoch": 148.7787610619469,
      "grad_norm": 8.0,
      "learning_rate": 4.4658423907299845e-05,
      "loss": 0.9805,
      "num_input_tokens_seen": 4702400,
      "step": 8480
    },
    {
      "epoch": 148.86725663716814,
      "grad_norm": 9.375,
      "learning_rate": 4.465235717539725e-05,
      "loss": 1.1048,
      "num_input_tokens_seen": 4705328,
      "step": 8485
    },
    {
      "epoch": 148.95575221238937,
      "grad_norm": 7.25,
      "learning_rate": 4.464628741285421e-05,
      "loss": 0.933,
      "num_input_tokens_seen": 4708160,
      "step": 8490
    },
    {
      "epoch": 149.0353982300885,
      "grad_norm": 7.40625,
      "learning_rate": 4.4640214620606754e-05,
      "loss": 1.0328,
      "num_input_tokens_seen": 4710920,
      "step": 8495
    },
    {
      "epoch": 149.12389380530973,
      "grad_norm": 9.6875,
      "learning_rate": 4.46341387995914e-05,
      "loss": 1.2164,
      "num_input_tokens_seen": 4713688,
      "step": 8500
    },
    {
      "epoch": 149.21238938053096,
      "grad_norm": 8.4375,
      "learning_rate": 4.4628059950745106e-05,
      "loss": 0.8994,
      "num_input_tokens_seen": 4716456,
      "step": 8505
    },
    {
      "epoch": 149.30088495575222,
      "grad_norm": 6.21875,
      "learning_rate": 4.4621978075005297e-05,
      "loss": 0.9203,
      "num_input_tokens_seen": 4719560,
      "step": 8510
    },
    {
      "epoch": 149.38938053097345,
      "grad_norm": 8.0625,
      "learning_rate": 4.461589317330989e-05,
      "loss": 1.0489,
      "num_input_tokens_seen": 4722488,
      "step": 8515
    },
    {
      "epoch": 149.47787610619469,
      "grad_norm": 7.875,
      "learning_rate": 4.460980524659724e-05,
      "loss": 1.1,
      "num_input_tokens_seen": 4725128,
      "step": 8520
    },
    {
      "epoch": 149.56637168141592,
      "grad_norm": 8.9375,
      "learning_rate": 4.46037142958062e-05,
      "loss": 1.0051,
      "num_input_tokens_seen": 4727480,
      "step": 8525
    },
    {
      "epoch": 149.65486725663717,
      "grad_norm": 8.4375,
      "learning_rate": 4.4597620321876046e-05,
      "loss": 1.1188,
      "num_input_tokens_seen": 4730008,
      "step": 8530
    },
    {
      "epoch": 149.7433628318584,
      "grad_norm": 8.0625,
      "learning_rate": 4.459152332574656e-05,
      "loss": 1.106,
      "num_input_tokens_seen": 4733064,
      "step": 8535
    },
    {
      "epoch": 149.83185840707964,
      "grad_norm": 8.5625,
      "learning_rate": 4.4585423308357985e-05,
      "loss": 1.0786,
      "num_input_tokens_seen": 4736040,
      "step": 8540
    },
    {
      "epoch": 149.9203539823009,
      "grad_norm": 7.46875,
      "learning_rate": 4.457932027065102e-05,
      "loss": 1.0369,
      "num_input_tokens_seen": 4738792,
      "step": 8545
    },
    {
      "epoch": 150.0,
      "grad_norm": 9.9375,
      "learning_rate": 4.45732142135668e-05,
      "loss": 0.8561,
      "num_input_tokens_seen": 4741464,
      "step": 8550
    },
    {
      "epoch": 150.08849557522123,
      "grad_norm": 9.75,
      "learning_rate": 4.4567105138046986e-05,
      "loss": 1.0224,
      "num_input_tokens_seen": 4743992,
      "step": 8555
    },
    {
      "epoch": 150.1769911504425,
      "grad_norm": 14.1875,
      "learning_rate": 4.456099304503365e-05,
      "loss": 1.0748,
      "num_input_tokens_seen": 4746920,
      "step": 8560
    },
    {
      "epoch": 150.26548672566372,
      "grad_norm": 8.5,
      "learning_rate": 4.455487793546939e-05,
      "loss": 1.133,
      "num_input_tokens_seen": 4749656,
      "step": 8565
    },
    {
      "epoch": 150.35398230088495,
      "grad_norm": 9.75,
      "learning_rate": 4.454875981029719e-05,
      "loss": 0.9986,
      "num_input_tokens_seen": 4752392,
      "step": 8570
    },
    {
      "epoch": 150.44247787610618,
      "grad_norm": 8.5,
      "learning_rate": 4.454263867046057e-05,
      "loss": 1.0534,
      "num_input_tokens_seen": 4755192,
      "step": 8575
    },
    {
      "epoch": 150.53097345132744,
      "grad_norm": 6.375,
      "learning_rate": 4.4536514516903484e-05,
      "loss": 1.003,
      "num_input_tokens_seen": 4757768,
      "step": 8580
    },
    {
      "epoch": 150.61946902654867,
      "grad_norm": 10.625,
      "learning_rate": 4.453038735057034e-05,
      "loss": 1.0625,
      "num_input_tokens_seen": 4760856,
      "step": 8585
    },
    {
      "epoch": 150.7079646017699,
      "grad_norm": 9.3125,
      "learning_rate": 4.4524257172406034e-05,
      "loss": 1.0736,
      "num_input_tokens_seen": 4763800,
      "step": 8590
    },
    {
      "epoch": 150.79646017699116,
      "grad_norm": 6.84375,
      "learning_rate": 4.451812398335592e-05,
      "loss": 1.0211,
      "num_input_tokens_seen": 4766648,
      "step": 8595
    },
    {
      "epoch": 150.8849557522124,
      "grad_norm": 9.375,
      "learning_rate": 4.4511987784365805e-05,
      "loss": 1.0173,
      "num_input_tokens_seen": 4769400,
      "step": 8600
    },
    {
      "epoch": 150.8849557522124,
      "eval_loss": 1.0291807651519775,
      "eval_runtime": 0.9316,
      "eval_samples_per_second": 26.835,
      "eval_steps_per_second": 13.954,
      "num_input_tokens_seen": 4769400,
      "step": 8600
    },
    {
      "epoch": 150.97345132743362,
      "grad_norm": 7.96875,
      "learning_rate": 4.450584857638197e-05,
      "loss": 1.1832,
      "num_input_tokens_seen": 4772472,
      "step": 8605
    },
    {
      "epoch": 151.05309734513276,
      "grad_norm": 7.90625,
      "learning_rate": 4.449970636035116e-05,
      "loss": 0.9238,
      "num_input_tokens_seen": 4774728,
      "step": 8610
    },
    {
      "epoch": 151.141592920354,
      "grad_norm": 7.78125,
      "learning_rate": 4.4493561137220574e-05,
      "loss": 1.1301,
      "num_input_tokens_seen": 4777176,
      "step": 8615
    },
    {
      "epoch": 151.23008849557522,
      "grad_norm": 8.4375,
      "learning_rate": 4.44874129079379e-05,
      "loss": 1.2035,
      "num_input_tokens_seen": 4779928,
      "step": 8620
    },
    {
      "epoch": 151.31858407079645,
      "grad_norm": 8.6875,
      "learning_rate": 4.4481261673451255e-05,
      "loss": 1.1201,
      "num_input_tokens_seen": 4782936,
      "step": 8625
    },
    {
      "epoch": 151.4070796460177,
      "grad_norm": 9.875,
      "learning_rate": 4.4475107434709245e-05,
      "loss": 1.0557,
      "num_input_tokens_seen": 4785688,
      "step": 8630
    },
    {
      "epoch": 151.49557522123894,
      "grad_norm": 7.90625,
      "learning_rate": 4.446895019266093e-05,
      "loss": 1.0522,
      "num_input_tokens_seen": 4788664,
      "step": 8635
    },
    {
      "epoch": 151.58407079646017,
      "grad_norm": 8.25,
      "learning_rate": 4.446278994825583e-05,
      "loss": 0.9492,
      "num_input_tokens_seen": 4791800,
      "step": 8640
    },
    {
      "epoch": 151.67256637168143,
      "grad_norm": 8.4375,
      "learning_rate": 4.445662670244394e-05,
      "loss": 0.9762,
      "num_input_tokens_seen": 4795096,
      "step": 8645
    },
    {
      "epoch": 151.76106194690266,
      "grad_norm": 7.1875,
      "learning_rate": 4.44504604561757e-05,
      "loss": 0.9977,
      "num_input_tokens_seen": 4797800,
      "step": 8650
    },
    {
      "epoch": 151.8495575221239,
      "grad_norm": 7.09375,
      "learning_rate": 4.4444291210402035e-05,
      "loss": 0.9574,
      "num_input_tokens_seen": 4800904,
      "step": 8655
    },
    {
      "epoch": 151.93805309734512,
      "grad_norm": 7.34375,
      "learning_rate": 4.443811896607431e-05,
      "loss": 1.0442,
      "num_input_tokens_seen": 4803640,
      "step": 8660
    },
    {
      "epoch": 152.01769911504425,
      "grad_norm": 8.25,
      "learning_rate": 4.443194372414436e-05,
      "loss": 1.0445,
      "num_input_tokens_seen": 4805952,
      "step": 8665
    },
    {
      "epoch": 152.10619469026548,
      "grad_norm": 10.375,
      "learning_rate": 4.442576548556449e-05,
      "loss": 1.0677,
      "num_input_tokens_seen": 4808464,
      "step": 8670
    },
    {
      "epoch": 152.1946902654867,
      "grad_norm": 8.5625,
      "learning_rate": 4.441958425128747e-05,
      "loss": 1.0209,
      "num_input_tokens_seen": 4811072,
      "step": 8675
    },
    {
      "epoch": 152.28318584070797,
      "grad_norm": 6.8125,
      "learning_rate": 4.4413400022266515e-05,
      "loss": 1.0889,
      "num_input_tokens_seen": 4814240,
      "step": 8680
    },
    {
      "epoch": 152.3716814159292,
      "grad_norm": 11.3125,
      "learning_rate": 4.4407212799455313e-05,
      "loss": 0.9084,
      "num_input_tokens_seen": 4816912,
      "step": 8685
    },
    {
      "epoch": 152.46017699115043,
      "grad_norm": 9.5,
      "learning_rate": 4.4401022583808003e-05,
      "loss": 1.158,
      "num_input_tokens_seen": 4819888,
      "step": 8690
    },
    {
      "epoch": 152.5486725663717,
      "grad_norm": 9.375,
      "learning_rate": 4.439482937627921e-05,
      "loss": 1.0872,
      "num_input_tokens_seen": 4823200,
      "step": 8695
    },
    {
      "epoch": 152.63716814159292,
      "grad_norm": 7.0625,
      "learning_rate": 4.4388633177824004e-05,
      "loss": 0.9753,
      "num_input_tokens_seen": 4825984,
      "step": 8700
    },
    {
      "epoch": 152.72566371681415,
      "grad_norm": 8.125,
      "learning_rate": 4.4382433989397895e-05,
      "loss": 1.0834,
      "num_input_tokens_seen": 4828544,
      "step": 8705
    },
    {
      "epoch": 152.81415929203538,
      "grad_norm": 7.34375,
      "learning_rate": 4.4376231811956895e-05,
      "loss": 0.9789,
      "num_input_tokens_seen": 4831920,
      "step": 8710
    },
    {
      "epoch": 152.90265486725664,
      "grad_norm": 5.9375,
      "learning_rate": 4.437002664645745e-05,
      "loss": 1.0489,
      "num_input_tokens_seen": 4834528,
      "step": 8715
    },
    {
      "epoch": 152.99115044247787,
      "grad_norm": 10.25,
      "learning_rate": 4.436381849385649e-05,
      "loss": 1.2594,
      "num_input_tokens_seen": 4837248,
      "step": 8720
    },
    {
      "epoch": 153.07079646017698,
      "grad_norm": 10.3125,
      "learning_rate": 4.435760735511136e-05,
      "loss": 0.8028,
      "num_input_tokens_seen": 4839336,
      "step": 8725
    },
    {
      "epoch": 153.15929203539824,
      "grad_norm": 10.0,
      "learning_rate": 4.435139323117992e-05,
      "loss": 1.1604,
      "num_input_tokens_seen": 4843048,
      "step": 8730
    },
    {
      "epoch": 153.24778761061947,
      "grad_norm": 7.9375,
      "learning_rate": 4.434517612302046e-05,
      "loss": 1.087,
      "num_input_tokens_seen": 4845800,
      "step": 8735
    },
    {
      "epoch": 153.3362831858407,
      "grad_norm": 11.625,
      "learning_rate": 4.433895603159174e-05,
      "loss": 1.1203,
      "num_input_tokens_seen": 4848280,
      "step": 8740
    },
    {
      "epoch": 153.42477876106196,
      "grad_norm": 8.375,
      "learning_rate": 4.433273295785296e-05,
      "loss": 0.9646,
      "num_input_tokens_seen": 4851608,
      "step": 8745
    },
    {
      "epoch": 153.5132743362832,
      "grad_norm": 8.5,
      "learning_rate": 4.432650690276382e-05,
      "loss": 0.955,
      "num_input_tokens_seen": 4854136,
      "step": 8750
    },
    {
      "epoch": 153.60176991150442,
      "grad_norm": 8.4375,
      "learning_rate": 4.4320277867284435e-05,
      "loss": 1.0698,
      "num_input_tokens_seen": 4857000,
      "step": 8755
    },
    {
      "epoch": 153.69026548672565,
      "grad_norm": 10.3125,
      "learning_rate": 4.431404585237541e-05,
      "loss": 1.084,
      "num_input_tokens_seen": 4859880,
      "step": 8760
    },
    {
      "epoch": 153.7787610619469,
      "grad_norm": 7.0,
      "learning_rate": 4.43078108589978e-05,
      "loss": 1.0114,
      "num_input_tokens_seen": 4862776,
      "step": 8765
    },
    {
      "epoch": 153.86725663716814,
      "grad_norm": 6.25,
      "learning_rate": 4.4301572888113116e-05,
      "loss": 1.0323,
      "num_input_tokens_seen": 4865416,
      "step": 8770
    },
    {
      "epoch": 153.95575221238937,
      "grad_norm": 7.71875,
      "learning_rate": 4.4295331940683337e-05,
      "loss": 1.1635,
      "num_input_tokens_seen": 4867896,
      "step": 8775
    },
    {
      "epoch": 154.0353982300885,
      "grad_norm": 7.625,
      "learning_rate": 4.428908801767089e-05,
      "loss": 0.9453,
      "num_input_tokens_seen": 4870152,
      "step": 8780
    },
    {
      "epoch": 154.12389380530973,
      "grad_norm": 7.25,
      "learning_rate": 4.428284112003868e-05,
      "loss": 1.0209,
      "num_input_tokens_seen": 4873064,
      "step": 8785
    },
    {
      "epoch": 154.21238938053096,
      "grad_norm": 7.15625,
      "learning_rate": 4.4276591248750033e-05,
      "loss": 1.1618,
      "num_input_tokens_seen": 4876152,
      "step": 8790
    },
    {
      "epoch": 154.30088495575222,
      "grad_norm": 6.875,
      "learning_rate": 4.4270338404768774e-05,
      "loss": 1.0899,
      "num_input_tokens_seen": 4878744,
      "step": 8795
    },
    {
      "epoch": 154.38938053097345,
      "grad_norm": 8.0625,
      "learning_rate": 4.426408258905917e-05,
      "loss": 0.9819,
      "num_input_tokens_seen": 4881880,
      "step": 8800
    },
    {
      "epoch": 154.38938053097345,
      "eval_loss": 1.0316694974899292,
      "eval_runtime": 0.9344,
      "eval_samples_per_second": 26.755,
      "eval_steps_per_second": 13.913,
      "num_input_tokens_seen": 4881880,
      "step": 8800
    },
    {
      "epoch": 154.47787610619469,
      "grad_norm": 7.375,
      "learning_rate": 4.425782380258594e-05,
      "loss": 1.0621,
      "num_input_tokens_seen": 4884520,
      "step": 8805
    },
    {
      "epoch": 154.56637168141592,
      "grad_norm": 8.125,
      "learning_rate": 4.425156204631427e-05,
      "loss": 1.4067,
      "num_input_tokens_seen": 4887560,
      "step": 8810
    },
    {
      "epoch": 154.65486725663717,
      "grad_norm": 7.1875,
      "learning_rate": 4.424529732120981e-05,
      "loss": 0.869,
      "num_input_tokens_seen": 4890264,
      "step": 8815
    },
    {
      "epoch": 154.7433628318584,
      "grad_norm": 8.0625,
      "learning_rate": 4.423902962823864e-05,
      "loss": 0.9935,
      "num_input_tokens_seen": 4893336,
      "step": 8820
    },
    {
      "epoch": 154.83185840707964,
      "grad_norm": 8.875,
      "learning_rate": 4.423275896836733e-05,
      "loss": 1.0856,
      "num_input_tokens_seen": 4896120,
      "step": 8825
    },
    {
      "epoch": 154.9203539823009,
      "grad_norm": 12.4375,
      "learning_rate": 4.42264853425629e-05,
      "loss": 0.9144,
      "num_input_tokens_seen": 4898616,
      "step": 8830
    },
    {
      "epoch": 155.0,
      "grad_norm": 16.25,
      "learning_rate": 4.4220208751792816e-05,
      "loss": 1.0063,
      "num_input_tokens_seen": 4900856,
      "step": 8835
    },
    {
      "epoch": 155.08849557522123,
      "grad_norm": 10.75,
      "learning_rate": 4.421392919702499e-05,
      "loss": 1.1787,
      "num_input_tokens_seen": 4903752,
      "step": 8840
    },
    {
      "epoch": 155.1769911504425,
      "grad_norm": 12.4375,
      "learning_rate": 4.4207646679227846e-05,
      "loss": 0.9929,
      "num_input_tokens_seen": 4906376,
      "step": 8845
    },
    {
      "epoch": 155.26548672566372,
      "grad_norm": 7.3125,
      "learning_rate": 4.42013611993702e-05,
      "loss": 0.9391,
      "num_input_tokens_seen": 4909480,
      "step": 8850
    },
    {
      "epoch": 155.35398230088495,
      "grad_norm": 8.0625,
      "learning_rate": 4.419507275842135e-05,
      "loss": 1.0547,
      "num_input_tokens_seen": 4912040,
      "step": 8855
    },
    {
      "epoch": 155.44247787610618,
      "grad_norm": 7.1875,
      "learning_rate": 4.418878135735106e-05,
      "loss": 1.0998,
      "num_input_tokens_seen": 4915112,
      "step": 8860
    },
    {
      "epoch": 155.53097345132744,
      "grad_norm": 6.46875,
      "learning_rate": 4.418248699712955e-05,
      "loss": 1.0068,
      "num_input_tokens_seen": 4917880,
      "step": 8865
    },
    {
      "epoch": 155.61946902654867,
      "grad_norm": 11.0625,
      "learning_rate": 4.417618967872748e-05,
      "loss": 1.2465,
      "num_input_tokens_seen": 4920712,
      "step": 8870
    },
    {
      "epoch": 155.7079646017699,
      "grad_norm": 8.5625,
      "learning_rate": 4.4169889403115985e-05,
      "loss": 0.967,
      "num_input_tokens_seen": 4923528,
      "step": 8875
    },
    {
      "epoch": 155.79646017699116,
      "grad_norm": 9.75,
      "learning_rate": 4.4163586171266627e-05,
      "loss": 1.1283,
      "num_input_tokens_seen": 4926104,
      "step": 8880
    },
    {
      "epoch": 155.8849557522124,
      "grad_norm": 10.3125,
      "learning_rate": 4.415727998415147e-05,
      "loss": 1.0987,
      "num_input_tokens_seen": 4928728,
      "step": 8885
    },
    {
      "epoch": 155.97345132743362,
      "grad_norm": 6.84375,
      "learning_rate": 4.4150970842742985e-05,
      "loss": 1.0485,
      "num_input_tokens_seen": 4931768,
      "step": 8890
    },
    {
      "epoch": 156.05309734513276,
      "grad_norm": 8.625,
      "learning_rate": 4.4144658748014134e-05,
      "loss": 0.7752,
      "num_input_tokens_seen": 4934128,
      "step": 8895
    },
    {
      "epoch": 156.141592920354,
      "grad_norm": 7.34375,
      "learning_rate": 4.413834370093831e-05,
      "loss": 1.0634,
      "num_input_tokens_seen": 4937024,
      "step": 8900
    },
    {
      "epoch": 156.23008849557522,
      "grad_norm": 6.65625,
      "learning_rate": 4.413202570248939e-05,
      "loss": 0.974,
      "num_input_tokens_seen": 4939936,
      "step": 8905
    },
    {
      "epoch": 156.31858407079645,
      "grad_norm": 9.6875,
      "learning_rate": 4.412570475364167e-05,
      "loss": 1.1801,
      "num_input_tokens_seen": 4942832,
      "step": 8910
    },
    {
      "epoch": 156.4070796460177,
      "grad_norm": 10.0625,
      "learning_rate": 4.411938085536994e-05,
      "loss": 0.9539,
      "num_input_tokens_seen": 4945168,
      "step": 8915
    },
    {
      "epoch": 156.49557522123894,
      "grad_norm": 8.3125,
      "learning_rate": 4.41130540086494e-05,
      "loss": 0.8528,
      "num_input_tokens_seen": 4948512,
      "step": 8920
    },
    {
      "epoch": 156.58407079646017,
      "grad_norm": 8.3125,
      "learning_rate": 4.4106724214455754e-05,
      "loss": 1.0571,
      "num_input_tokens_seen": 4951424,
      "step": 8925
    },
    {
      "epoch": 156.67256637168143,
      "grad_norm": 7.8125,
      "learning_rate": 4.4100391473765115e-05,
      "loss": 1.2063,
      "num_input_tokens_seen": 4953840,
      "step": 8930
    },
    {
      "epoch": 156.76106194690266,
      "grad_norm": 9.6875,
      "learning_rate": 4.409405578755408e-05,
      "loss": 1.1911,
      "num_input_tokens_seen": 4956128,
      "step": 8935
    },
    {
      "epoch": 156.8495575221239,
      "grad_norm": 9.0625,
      "learning_rate": 4.4087717156799705e-05,
      "loss": 0.9775,
      "num_input_tokens_seen": 4959056,
      "step": 8940
    },
    {
      "epoch": 156.93805309734512,
      "grad_norm": 9.625,
      "learning_rate": 4.408137558247946e-05,
      "loss": 1.1119,
      "num_input_tokens_seen": 4962256,
      "step": 8945
    },
    {
      "epoch": 157.01769911504425,
      "grad_norm": 7.09375,
      "learning_rate": 4.4075031065571306e-05,
      "loss": 0.8782,
      "num_input_tokens_seen": 4964344,
      "step": 8950
    },
    {
      "epoch": 157.10619469026548,
      "grad_norm": 12.0625,
      "learning_rate": 4.406868360705366e-05,
      "loss": 0.9615,
      "num_input_tokens_seen": 4967480,
      "step": 8955
    },
    {
      "epoch": 157.1946902654867,
      "grad_norm": 7.875,
      "learning_rate": 4.406233320790536e-05,
      "loss": 1.181,
      "num_input_tokens_seen": 4970280,
      "step": 8960
    },
    {
      "epoch": 157.28318584070797,
      "grad_norm": 6.65625,
      "learning_rate": 4.4055979869105734e-05,
      "loss": 1.0426,
      "num_input_tokens_seen": 4972952,
      "step": 8965
    },
    {
      "epoch": 157.3716814159292,
      "grad_norm": 11.5,
      "learning_rate": 4.404962359163454e-05,
      "loss": 1.0474,
      "num_input_tokens_seen": 4975928,
      "step": 8970
    },
    {
      "epoch": 157.46017699115043,
      "grad_norm": 7.15625,
      "learning_rate": 4.404326437647199e-05,
      "loss": 1.1402,
      "num_input_tokens_seen": 4978696,
      "step": 8975
    },
    {
      "epoch": 157.5486725663717,
      "grad_norm": 8.5625,
      "learning_rate": 4.403690222459877e-05,
      "loss": 0.9069,
      "num_input_tokens_seen": 4981448,
      "step": 8980
    },
    {
      "epoch": 157.63716814159292,
      "grad_norm": 8.6875,
      "learning_rate": 4.4030537136995984e-05,
      "loss": 1.0852,
      "num_input_tokens_seen": 4984120,
      "step": 8985
    },
    {
      "epoch": 157.72566371681415,
      "grad_norm": 12.4375,
      "learning_rate": 4.402416911464523e-05,
      "loss": 1.0399,
      "num_input_tokens_seen": 4986888,
      "step": 8990
    },
    {
      "epoch": 157.81415929203538,
      "grad_norm": 8.125,
      "learning_rate": 4.4017798158528516e-05,
      "loss": 0.9552,
      "num_input_tokens_seen": 4989544,
      "step": 8995
    },
    {
      "epoch": 157.90265486725664,
      "grad_norm": 7.0625,
      "learning_rate": 4.401142426962834e-05,
      "loss": 1.1285,
      "num_input_tokens_seen": 4992488,
      "step": 9000
    },
    {
      "epoch": 157.90265486725664,
      "eval_loss": 1.0267274379730225,
      "eval_runtime": 0.9342,
      "eval_samples_per_second": 26.762,
      "eval_steps_per_second": 13.916,
      "num_input_tokens_seen": 4992488,
      "step": 9000
    },
    {
      "epoch": 157.99115044247787,
      "grad_norm": 7.15625,
      "learning_rate": 4.400504744892763e-05,
      "loss": 1.1715,
      "num_input_tokens_seen": 4995288,
      "step": 9005
    },
    {
      "epoch": 158.07079646017698,
      "grad_norm": 9.125,
      "learning_rate": 4.399866769740975e-05,
      "loss": 1.1295,
      "num_input_tokens_seen": 4997600,
      "step": 9010
    },
    {
      "epoch": 158.15929203539824,
      "grad_norm": 6.625,
      "learning_rate": 4.399228501605859e-05,
      "loss": 0.9989,
      "num_input_tokens_seen": 5000208,
      "step": 9015
    },
    {
      "epoch": 158.24778761061947,
      "grad_norm": 8.3125,
      "learning_rate": 4.398589940585839e-05,
      "loss": 1.0671,
      "num_input_tokens_seen": 5002736,
      "step": 9020
    },
    {
      "epoch": 158.3362831858407,
      "grad_norm": 10.6875,
      "learning_rate": 4.3979510867793917e-05,
      "loss": 1.0416,
      "num_input_tokens_seen": 5005920,
      "step": 9025
    },
    {
      "epoch": 158.42477876106196,
      "grad_norm": 8.125,
      "learning_rate": 4.3973119402850346e-05,
      "loss": 0.9425,
      "num_input_tokens_seen": 5008736,
      "step": 9030
    },
    {
      "epoch": 158.5132743362832,
      "grad_norm": 7.6875,
      "learning_rate": 4.396672501201334e-05,
      "loss": 1.0236,
      "num_input_tokens_seen": 5011376,
      "step": 9035
    },
    {
      "epoch": 158.60176991150442,
      "grad_norm": 8.375,
      "learning_rate": 4.396032769626899e-05,
      "loss": 0.9882,
      "num_input_tokens_seen": 5014560,
      "step": 9040
    },
    {
      "epoch": 158.69026548672565,
      "grad_norm": 8.8125,
      "learning_rate": 4.395392745660384e-05,
      "loss": 1.1401,
      "num_input_tokens_seen": 5017200,
      "step": 9045
    },
    {
      "epoch": 158.7787610619469,
      "grad_norm": 9.0,
      "learning_rate": 4.394752429400488e-05,
      "loss": 1.0917,
      "num_input_tokens_seen": 5020192,
      "step": 9050
    },
    {
      "epoch": 158.86725663716814,
      "grad_norm": 10.9375,
      "learning_rate": 4.394111820945957e-05,
      "loss": 1.269,
      "num_input_tokens_seen": 5022848,
      "step": 9055
    },
    {
      "epoch": 158.95575221238937,
      "grad_norm": 10.875,
      "learning_rate": 4.393470920395579e-05,
      "loss": 0.9155,
      "num_input_tokens_seen": 5026048,
      "step": 9060
    },
    {
      "epoch": 159.0353982300885,
      "grad_norm": 7.78125,
      "learning_rate": 4.392829727848192e-05,
      "loss": 0.9958,
      "num_input_tokens_seen": 5028680,
      "step": 9065
    },
    {
      "epoch": 159.12389380530973,
      "grad_norm": 7.96875,
      "learning_rate": 4.392188243402673e-05,
      "loss": 1.1236,
      "num_input_tokens_seen": 5031592,
      "step": 9070
    },
    {
      "epoch": 159.21238938053096,
      "grad_norm": 9.5,
      "learning_rate": 4.391546467157949e-05,
      "loss": 1.1179,
      "num_input_tokens_seen": 5034424,
      "step": 9075
    },
    {
      "epoch": 159.30088495575222,
      "grad_norm": 7.75,
      "learning_rate": 4.390904399212988e-05,
      "loss": 0.937,
      "num_input_tokens_seen": 5037256,
      "step": 9080
    },
    {
      "epoch": 159.38938053097345,
      "grad_norm": 9.8125,
      "learning_rate": 4.390262039666807e-05,
      "loss": 1.0154,
      "num_input_tokens_seen": 5039976,
      "step": 9085
    },
    {
      "epoch": 159.47787610619469,
      "grad_norm": 8.6875,
      "learning_rate": 4.389619388618464e-05,
      "loss": 1.1352,
      "num_input_tokens_seen": 5042840,
      "step": 9090
    },
    {
      "epoch": 159.56637168141592,
      "grad_norm": 9.375,
      "learning_rate": 4.3889764461670655e-05,
      "loss": 1.1578,
      "num_input_tokens_seen": 5045576,
      "step": 9095
    },
    {
      "epoch": 159.65486725663717,
      "grad_norm": 8.875,
      "learning_rate": 4.38833321241176e-05,
      "loss": 1.0954,
      "num_input_tokens_seen": 5048744,
      "step": 9100
    },
    {
      "epoch": 159.7433628318584,
      "grad_norm": 11.5,
      "learning_rate": 4.3876896874517434e-05,
      "loss": 1.2066,
      "num_input_tokens_seen": 5051720,
      "step": 9105
    },
    {
      "epoch": 159.83185840707964,
      "grad_norm": 8.3125,
      "learning_rate": 4.3870458713862554e-05,
      "loss": 1.06,
      "num_input_tokens_seen": 5054200,
      "step": 9110
    },
    {
      "epoch": 159.9203539823009,
      "grad_norm": 8.125,
      "learning_rate": 4.386401764314579e-05,
      "loss": 0.8925,
      "num_input_tokens_seen": 5056584,
      "step": 9115
    },
    {
      "epoch": 160.0,
      "grad_norm": 34.75,
      "learning_rate": 4.385757366336045e-05,
      "loss": 0.9885,
      "num_input_tokens_seen": 5058800,
      "step": 9120
    },
    {
      "epoch": 160.08849557522123,
      "grad_norm": 8.8125,
      "learning_rate": 4.385112677550027e-05,
      "loss": 1.1555,
      "num_input_tokens_seen": 5061792,
      "step": 9125
    },
    {
      "epoch": 160.1769911504425,
      "grad_norm": 8.1875,
      "learning_rate": 4.384467698055945e-05,
      "loss": 1.2952,
      "num_input_tokens_seen": 5064496,
      "step": 9130
    },
    {
      "epoch": 160.26548672566372,
      "grad_norm": 9.25,
      "learning_rate": 4.383822427953261e-05,
      "loss": 0.9281,
      "num_input_tokens_seen": 5067392,
      "step": 9135
    },
    {
      "epoch": 160.35398230088495,
      "grad_norm": 10.0625,
      "learning_rate": 4.3831768673414864e-05,
      "loss": 0.8522,
      "num_input_tokens_seen": 5070112,
      "step": 9140
    },
    {
      "epoch": 160.44247787610618,
      "grad_norm": 8.4375,
      "learning_rate": 4.382531016320173e-05,
      "loss": 1.1743,
      "num_input_tokens_seen": 5072720,
      "step": 9145
    },
    {
      "epoch": 160.53097345132744,
      "grad_norm": 8.3125,
      "learning_rate": 4.3818848749889184e-05,
      "loss": 0.9319,
      "num_input_tokens_seen": 5075312,
      "step": 9150
    },
    {
      "epoch": 160.61946902654867,
      "grad_norm": 10.0625,
      "learning_rate": 4.381238443447368e-05,
      "loss": 0.9692,
      "num_input_tokens_seen": 5078256,
      "step": 9155
    },
    {
      "epoch": 160.7079646017699,
      "grad_norm": 8.3125,
      "learning_rate": 4.380591721795208e-05,
      "loss": 0.953,
      "num_input_tokens_seen": 5081232,
      "step": 9160
    },
    {
      "epoch": 160.79646017699116,
      "grad_norm": 7.1875,
      "learning_rate": 4.3799447101321723e-05,
      "loss": 1.1806,
      "num_input_tokens_seen": 5084304,
      "step": 9165
    },
    {
      "epoch": 160.8849557522124,
      "grad_norm": 8.875,
      "learning_rate": 4.379297408558036e-05,
      "loss": 1.1443,
      "num_input_tokens_seen": 5087056,
      "step": 9170
    },
    {
      "epoch": 160.97345132743362,
      "grad_norm": 8.25,
      "learning_rate": 4.378649817172624e-05,
      "loss": 1.0352,
      "num_input_tokens_seen": 5089808,
      "step": 9175
    },
    {
      "epoch": 161.05309734513276,
      "grad_norm": 11.0,
      "learning_rate": 4.378001936075801e-05,
      "loss": 1.155,
      "num_input_tokens_seen": 5092088,
      "step": 9180
    },
    {
      "epoch": 161.141592920354,
      "grad_norm": 7.875,
      "learning_rate": 4.377353765367479e-05,
      "loss": 1.0471,
      "num_input_tokens_seen": 5094664,
      "step": 9185
    },
    {
      "epoch": 161.23008849557522,
      "grad_norm": 16.625,
      "learning_rate": 4.376705305147614e-05,
      "loss": 0.9355,
      "num_input_tokens_seen": 5097080,
      "step": 9190
    },
    {
      "epoch": 161.31858407079645,
      "grad_norm": 7.03125,
      "learning_rate": 4.376056555516206e-05,
      "loss": 1.103,
      "num_input_tokens_seen": 5099688,
      "step": 9195
    },
    {
      "epoch": 161.4070796460177,
      "grad_norm": 10.3125,
      "learning_rate": 4.375407516573302e-05,
      "loss": 1.2546,
      "num_input_tokens_seen": 5103032,
      "step": 9200
    },
    {
      "epoch": 161.4070796460177,
      "eval_loss": 1.0334337949752808,
      "eval_runtime": 0.9325,
      "eval_samples_per_second": 26.809,
      "eval_steps_per_second": 13.941,
      "num_input_tokens_seen": 5103032,
      "step": 9200
    },
    {
      "epoch": 161.49557522123894,
      "grad_norm": 9.3125,
      "learning_rate": 4.3747581884189913e-05,
      "loss": 1.1482,
      "num_input_tokens_seen": 5105944,
      "step": 9205
    },
    {
      "epoch": 161.58407079646017,
      "grad_norm": 8.125,
      "learning_rate": 4.374108571153408e-05,
      "loss": 0.9295,
      "num_input_tokens_seen": 5108776,
      "step": 9210
    },
    {
      "epoch": 161.67256637168143,
      "grad_norm": 6.34375,
      "learning_rate": 4.3734586648767316e-05,
      "loss": 1.024,
      "num_input_tokens_seen": 5111912,
      "step": 9215
    },
    {
      "epoch": 161.76106194690266,
      "grad_norm": 12.0625,
      "learning_rate": 4.372808469689186e-05,
      "loss": 1.1197,
      "num_input_tokens_seen": 5114712,
      "step": 9220
    },
    {
      "epoch": 161.8495575221239,
      "grad_norm": 8.0625,
      "learning_rate": 4.372157985691039e-05,
      "loss": 1.0219,
      "num_input_tokens_seen": 5117080,
      "step": 9225
    },
    {
      "epoch": 161.93805309734512,
      "grad_norm": 8.25,
      "learning_rate": 4.371507212982603e-05,
      "loss": 1.0081,
      "num_input_tokens_seen": 5119544,
      "step": 9230
    },
    {
      "epoch": 162.01769911504425,
      "grad_norm": 12.75,
      "learning_rate": 4.370856151664236e-05,
      "loss": 1.0907,
      "num_input_tokens_seen": 5121864,
      "step": 9235
    },
    {
      "epoch": 162.10619469026548,
      "grad_norm": 6.4375,
      "learning_rate": 4.3702048018363404e-05,
      "loss": 1.1994,
      "num_input_tokens_seen": 5124632,
      "step": 9240
    },
    {
      "epoch": 162.1946902654867,
      "grad_norm": 8.6875,
      "learning_rate": 4.369553163599362e-05,
      "loss": 1.223,
      "num_input_tokens_seen": 5127176,
      "step": 9245
    },
    {
      "epoch": 162.28318584070797,
      "grad_norm": 10.75,
      "learning_rate": 4.3689012370537904e-05,
      "loss": 1.1045,
      "num_input_tokens_seen": 5130024,
      "step": 9250
    },
    {
      "epoch": 162.3716814159292,
      "grad_norm": 11.0,
      "learning_rate": 4.368249022300164e-05,
      "loss": 0.8799,
      "num_input_tokens_seen": 5132712,
      "step": 9255
    },
    {
      "epoch": 162.46017699115043,
      "grad_norm": 8.8125,
      "learning_rate": 4.367596519439059e-05,
      "loss": 0.8766,
      "num_input_tokens_seen": 5135976,
      "step": 9260
    },
    {
      "epoch": 162.5486725663717,
      "grad_norm": 8.3125,
      "learning_rate": 4.366943728571101e-05,
      "loss": 0.8834,
      "num_input_tokens_seen": 5138744,
      "step": 9265
    },
    {
      "epoch": 162.63716814159292,
      "grad_norm": 8.8125,
      "learning_rate": 4.366290649796959e-05,
      "loss": 1.0721,
      "num_input_tokens_seen": 5141320,
      "step": 9270
    },
    {
      "epoch": 162.72566371681415,
      "grad_norm": 7.71875,
      "learning_rate": 4.3656372832173456e-05,
      "loss": 1.0784,
      "num_input_tokens_seen": 5144104,
      "step": 9275
    },
    {
      "epoch": 162.81415929203538,
      "grad_norm": 8.0625,
      "learning_rate": 4.364983628933017e-05,
      "loss": 1.0355,
      "num_input_tokens_seen": 5147096,
      "step": 9280
    },
    {
      "epoch": 162.90265486725664,
      "grad_norm": 7.65625,
      "learning_rate": 4.364329687044777e-05,
      "loss": 1.125,
      "num_input_tokens_seen": 5150024,
      "step": 9285
    },
    {
      "epoch": 162.99115044247787,
      "grad_norm": 8.75,
      "learning_rate": 4.36367545765347e-05,
      "loss": 1.0457,
      "num_input_tokens_seen": 5153016,
      "step": 9290
    },
    {
      "epoch": 163.07079646017698,
      "grad_norm": 7.125,
      "learning_rate": 4.363020940859988e-05,
      "loss": 0.7244,
      "num_input_tokens_seen": 5155640,
      "step": 9295
    },
    {
      "epoch": 163.15929203539824,
      "grad_norm": 9.0,
      "learning_rate": 4.362366136765263e-05,
      "loss": 1.0068,
      "num_input_tokens_seen": 5158264,
      "step": 9300
    },
    {
      "epoch": 163.24778761061947,
      "grad_norm": 7.9375,
      "learning_rate": 4.361711045470278e-05,
      "loss": 1.1069,
      "num_input_tokens_seen": 5161128,
      "step": 9305
    },
    {
      "epoch": 163.3362831858407,
      "grad_norm": 7.03125,
      "learning_rate": 4.3610556670760524e-05,
      "loss": 1.2988,
      "num_input_tokens_seen": 5163960,
      "step": 9310
    },
    {
      "epoch": 163.42477876106196,
      "grad_norm": 10.75,
      "learning_rate": 4.360400001683657e-05,
      "loss": 1.096,
      "num_input_tokens_seen": 5166696,
      "step": 9315
    },
    {
      "epoch": 163.5132743362832,
      "grad_norm": 7.5625,
      "learning_rate": 4.3597440493942e-05,
      "loss": 1.0566,
      "num_input_tokens_seen": 5169112,
      "step": 9320
    },
    {
      "epoch": 163.60176991150442,
      "grad_norm": 11.0625,
      "learning_rate": 4.3590878103088405e-05,
      "loss": 1.1821,
      "num_input_tokens_seen": 5171864,
      "step": 9325
    },
    {
      "epoch": 163.69026548672565,
      "grad_norm": 5.5625,
      "learning_rate": 4.358431284528779e-05,
      "loss": 0.8373,
      "num_input_tokens_seen": 5174728,
      "step": 9330
    },
    {
      "epoch": 163.7787610619469,
      "grad_norm": 8.4375,
      "learning_rate": 4.357774472155257e-05,
      "loss": 1.1339,
      "num_input_tokens_seen": 5177800,
      "step": 9335
    },
    {
      "epoch": 163.86725663716814,
      "grad_norm": 10.875,
      "learning_rate": 4.3571173732895664e-05,
      "loss": 0.9166,
      "num_input_tokens_seen": 5181016,
      "step": 9340
    },
    {
      "epoch": 163.95575221238937,
      "grad_norm": 5.96875,
      "learning_rate": 4.356459988033039e-05,
      "loss": 1.0397,
      "num_input_tokens_seen": 5183592,
      "step": 9345
    },
    {
      "epoch": 164.0353982300885,
      "grad_norm": 6.75,
      "learning_rate": 4.355802316487051e-05,
      "loss": 1.1265,
      "num_input_tokens_seen": 5185656,
      "step": 9350
    },
    {
      "epoch": 164.12389380530973,
      "grad_norm": 8.5625,
      "learning_rate": 4.355144358753025e-05,
      "loss": 1.0238,
      "num_input_tokens_seen": 5188520,
      "step": 9355
    },
    {
      "epoch": 164.21238938053096,
      "grad_norm": 8.5,
      "learning_rate": 4.354486114932425e-05,
      "loss": 1.0366,
      "num_input_tokens_seen": 5191464,
      "step": 9360
    },
    {
      "epoch": 164.30088495575222,
      "grad_norm": 8.1875,
      "learning_rate": 4.353827585126762e-05,
      "loss": 1.0623,
      "num_input_tokens_seen": 5194312,
      "step": 9365
    },
    {
      "epoch": 164.38938053097345,
      "grad_norm": 10.3125,
      "learning_rate": 4.353168769437588e-05,
      "loss": 0.9243,
      "num_input_tokens_seen": 5197048,
      "step": 9370
    },
    {
      "epoch": 164.47787610619469,
      "grad_norm": 7.0,
      "learning_rate": 4.3525096679665014e-05,
      "loss": 1.029,
      "num_input_tokens_seen": 5199960,
      "step": 9375
    },
    {
      "epoch": 164.56637168141592,
      "grad_norm": 13.5625,
      "learning_rate": 4.351850280815144e-05,
      "loss": 1.0129,
      "num_input_tokens_seen": 5202536,
      "step": 9380
    },
    {
      "epoch": 164.65486725663717,
      "grad_norm": 10.125,
      "learning_rate": 4.3511906080852014e-05,
      "loss": 1.3752,
      "num_input_tokens_seen": 5205528,
      "step": 9385
    },
    {
      "epoch": 164.7433628318584,
      "grad_norm": 9.5,
      "learning_rate": 4.350530649878404e-05,
      "loss": 1.1537,
      "num_input_tokens_seen": 5208664,
      "step": 9390
    },
    {
      "epoch": 164.83185840707964,
      "grad_norm": 7.15625,
      "learning_rate": 4.3498704062965246e-05,
      "loss": 0.8587,
      "num_input_tokens_seen": 5211432,
      "step": 9395
    },
    {
      "epoch": 164.9203539823009,
      "grad_norm": 8.6875,
      "learning_rate": 4.3492098774413815e-05,
      "loss": 0.9652,
      "num_input_tokens_seen": 5214280,
      "step": 9400
    },
    {
      "epoch": 164.9203539823009,
      "eval_loss": 1.0267728567123413,
      "eval_runtime": 0.9326,
      "eval_samples_per_second": 26.808,
      "eval_steps_per_second": 13.94,
      "num_input_tokens_seen": 5214280,
      "step": 9400
    },
    {
      "epoch": 165.0,
      "grad_norm": 8.75,
      "learning_rate": 4.3485490634148375e-05,
      "loss": 1.0908,
      "num_input_tokens_seen": 5216520,
      "step": 9405
    },
    {
      "epoch": 165.08849557522123,
      "grad_norm": 8.25,
      "learning_rate": 4.347887964318797e-05,
      "loss": 1.0109,
      "num_input_tokens_seen": 5219656,
      "step": 9410
    },
    {
      "epoch": 165.1769911504425,
      "grad_norm": 8.3125,
      "learning_rate": 4.34722658025521e-05,
      "loss": 1.1058,
      "num_input_tokens_seen": 5222200,
      "step": 9415
    },
    {
      "epoch": 165.26548672566372,
      "grad_norm": 6.09375,
      "learning_rate": 4.346564911326071e-05,
      "loss": 1.1078,
      "num_input_tokens_seen": 5225256,
      "step": 9420
    },
    {
      "epoch": 165.35398230088495,
      "grad_norm": 9.0625,
      "learning_rate": 4.345902957633418e-05,
      "loss": 1.1433,
      "num_input_tokens_seen": 5228280,
      "step": 9425
    },
    {
      "epoch": 165.44247787610618,
      "grad_norm": 9.375,
      "learning_rate": 4.345240719279331e-05,
      "loss": 1.0511,
      "num_input_tokens_seen": 5230872,
      "step": 9430
    },
    {
      "epoch": 165.53097345132744,
      "grad_norm": 10.25,
      "learning_rate": 4.3445781963659374e-05,
      "loss": 1.0278,
      "num_input_tokens_seen": 5233544,
      "step": 9435
    },
    {
      "epoch": 165.61946902654867,
      "grad_norm": 7.875,
      "learning_rate": 4.3439153889954045e-05,
      "loss": 1.1497,
      "num_input_tokens_seen": 5236280,
      "step": 9440
    },
    {
      "epoch": 165.7079646017699,
      "grad_norm": 7.59375,
      "learning_rate": 4.343252297269946e-05,
      "loss": 0.9051,
      "num_input_tokens_seen": 5238792,
      "step": 9445
    },
    {
      "epoch": 165.79646017699116,
      "grad_norm": 7.8125,
      "learning_rate": 4.342588921291821e-05,
      "loss": 1.1857,
      "num_input_tokens_seen": 5241320,
      "step": 9450
    },
    {
      "epoch": 165.8849557522124,
      "grad_norm": 12.5625,
      "learning_rate": 4.341925261163328e-05,
      "loss": 1.0214,
      "num_input_tokens_seen": 5244376,
      "step": 9455
    },
    {
      "epoch": 165.97345132743362,
      "grad_norm": 8.1875,
      "learning_rate": 4.341261316986813e-05,
      "loss": 0.9722,
      "num_input_tokens_seen": 5247512,
      "step": 9460
    },
    {
      "epoch": 166.05309734513276,
      "grad_norm": 4.90625,
      "learning_rate": 4.340597088864664e-05,
      "loss": 0.8829,
      "num_input_tokens_seen": 5249560,
      "step": 9465
    },
    {
      "epoch": 166.141592920354,
      "grad_norm": 7.46875,
      "learning_rate": 4.339932576899313e-05,
      "loss": 0.9702,
      "num_input_tokens_seen": 5252360,
      "step": 9470
    },
    {
      "epoch": 166.23008849557522,
      "grad_norm": 7.53125,
      "learning_rate": 4.3392677811932375e-05,
      "loss": 0.9959,
      "num_input_tokens_seen": 5256040,
      "step": 9475
    },
    {
      "epoch": 166.31858407079645,
      "grad_norm": 10.75,
      "learning_rate": 4.338602701848956e-05,
      "loss": 1.1368,
      "num_input_tokens_seen": 5258872,
      "step": 9480
    },
    {
      "epoch": 166.4070796460177,
      "grad_norm": 12.0625,
      "learning_rate": 4.337937338969033e-05,
      "loss": 1.3325,
      "num_input_tokens_seen": 5261416,
      "step": 9485
    },
    {
      "epoch": 166.49557522123894,
      "grad_norm": 4.8125,
      "learning_rate": 4.337271692656075e-05,
      "loss": 1.0073,
      "num_input_tokens_seen": 5263720,
      "step": 9490
    },
    {
      "epoch": 166.58407079646017,
      "grad_norm": 7.90625,
      "learning_rate": 4.336605763012733e-05,
      "loss": 1.1463,
      "num_input_tokens_seen": 5266392,
      "step": 9495
    },
    {
      "epoch": 166.67256637168143,
      "grad_norm": 8.6875,
      "learning_rate": 4.3359395501417026e-05,
      "loss": 0.9848,
      "num_input_tokens_seen": 5269448,
      "step": 9500
    },
    {
      "epoch": 166.76106194690266,
      "grad_norm": 11.5625,
      "learning_rate": 4.335273054145722e-05,
      "loss": 0.9735,
      "num_input_tokens_seen": 5272440,
      "step": 9505
    },
    {
      "epoch": 166.8495575221239,
      "grad_norm": 7.6875,
      "learning_rate": 4.334606275127572e-05,
      "loss": 0.991,
      "num_input_tokens_seen": 5275352,
      "step": 9510
    },
    {
      "epoch": 166.93805309734512,
      "grad_norm": 8.1875,
      "learning_rate": 4.33393921319008e-05,
      "loss": 1.0936,
      "num_input_tokens_seen": 5278008,
      "step": 9515
    },
    {
      "epoch": 167.01769911504425,
      "grad_norm": 8.625,
      "learning_rate": 4.3332718684361146e-05,
      "loss": 1.1868,
      "num_input_tokens_seen": 5280112,
      "step": 9520
    },
    {
      "epoch": 167.10619469026548,
      "grad_norm": 9.4375,
      "learning_rate": 4.332604240968588e-05,
      "loss": 1.1234,
      "num_input_tokens_seen": 5282688,
      "step": 9525
    },
    {
      "epoch": 167.1946902654867,
      "grad_norm": 8.375,
      "learning_rate": 4.331936330890459e-05,
      "loss": 0.9357,
      "num_input_tokens_seen": 5285376,
      "step": 9530
    },
    {
      "epoch": 167.28318584070797,
      "grad_norm": 9.5625,
      "learning_rate": 4.331268138304725e-05,
      "loss": 1.0903,
      "num_input_tokens_seen": 5288016,
      "step": 9535
    },
    {
      "epoch": 167.3716814159292,
      "grad_norm": 9.5,
      "learning_rate": 4.330599663314431e-05,
      "loss": 1.1809,
      "num_input_tokens_seen": 5290480,
      "step": 9540
    },
    {
      "epoch": 167.46017699115043,
      "grad_norm": 8.5625,
      "learning_rate": 4.329930906022665e-05,
      "loss": 1.2085,
      "num_input_tokens_seen": 5293472,
      "step": 9545
    },
    {
      "epoch": 167.5486725663717,
      "grad_norm": 6.53125,
      "learning_rate": 4.3292618665325564e-05,
      "loss": 1.0502,
      "num_input_tokens_seen": 5296144,
      "step": 9550
    },
    {
      "epoch": 167.63716814159292,
      "grad_norm": 8.875,
      "learning_rate": 4.3285925449472796e-05,
      "loss": 0.936,
      "num_input_tokens_seen": 5299136,
      "step": 9555
    },
    {
      "epoch": 167.72566371681415,
      "grad_norm": 9.6875,
      "learning_rate": 4.327922941370054e-05,
      "loss": 1.1833,
      "num_input_tokens_seen": 5302048,
      "step": 9560
    },
    {
      "epoch": 167.81415929203538,
      "grad_norm": 9.0,
      "learning_rate": 4.3272530559041384e-05,
      "loss": 0.9526,
      "num_input_tokens_seen": 5305184,
      "step": 9565
    },
    {
      "epoch": 167.90265486725664,
      "grad_norm": 9.375,
      "learning_rate": 4.32658288865284e-05,
      "loss": 1.0161,
      "num_input_tokens_seen": 5308096,
      "step": 9570
    },
    {
      "epoch": 167.99115044247787,
      "grad_norm": 8.375,
      "learning_rate": 4.325912439719505e-05,
      "loss": 0.8916,
      "num_input_tokens_seen": 5311072,
      "step": 9575
    },
    {
      "epoch": 168.07079646017698,
      "grad_norm": 7.46875,
      "learning_rate": 4.3252417092075266e-05,
      "loss": 0.9611,
      "num_input_tokens_seen": 5312992,
      "step": 9580
    },
    {
      "epoch": 168.15929203539824,
      "grad_norm": 9.6875,
      "learning_rate": 4.3245706972203385e-05,
      "loss": 1.0964,
      "num_input_tokens_seen": 5315632,
      "step": 9585
    },
    {
      "epoch": 168.24778761061947,
      "grad_norm": 8.8125,
      "learning_rate": 4.323899403861421e-05,
      "loss": 0.889,
      "num_input_tokens_seen": 5318208,
      "step": 9590
    },
    {
      "epoch": 168.3362831858407,
      "grad_norm": 8.75,
      "learning_rate": 4.3232278292342935e-05,
      "loss": 0.9795,
      "num_input_tokens_seen": 5320624,
      "step": 9595
    },
    {
      "epoch": 168.42477876106196,
      "grad_norm": 9.0625,
      "learning_rate": 4.322555973442524e-05,
      "loss": 1.0387,
      "num_input_tokens_seen": 5323664,
      "step": 9600
    },
    {
      "epoch": 168.42477876106196,
      "eval_loss": 1.0263911485671997,
      "eval_runtime": 0.933,
      "eval_samples_per_second": 26.796,
      "eval_steps_per_second": 13.934,
      "num_input_tokens_seen": 5323664,
      "step": 9600
    },
    {
      "epoch": 168.5132743362832,
      "grad_norm": 11.0,
      "learning_rate": 4.3218838365897184e-05,
      "loss": 0.978,
      "num_input_tokens_seen": 5326048,
      "step": 9605
    },
    {
      "epoch": 168.60176991150442,
      "grad_norm": 7.03125,
      "learning_rate": 4.3212114187795306e-05,
      "loss": 1.0482,
      "num_input_tokens_seen": 5328992,
      "step": 9610
    },
    {
      "epoch": 168.69026548672565,
      "grad_norm": 7.84375,
      "learning_rate": 4.320538720115656e-05,
      "loss": 1.3118,
      "num_input_tokens_seen": 5332400,
      "step": 9615
    },
    {
      "epoch": 168.7787610619469,
      "grad_norm": 7.3125,
      "learning_rate": 4.319865740701831e-05,
      "loss": 0.8341,
      "num_input_tokens_seen": 5335136,
      "step": 9620
    },
    {
      "epoch": 168.86725663716814,
      "grad_norm": 8.125,
      "learning_rate": 4.3191924806418396e-05,
      "loss": 1.3801,
      "num_input_tokens_seen": 5337840,
      "step": 9625
    },
    {
      "epoch": 168.95575221238937,
      "grad_norm": 8.375,
      "learning_rate": 4.318518940039507e-05,
      "loss": 1.0911,
      "num_input_tokens_seen": 5341296,
      "step": 9630
    },
    {
      "epoch": 169.0353982300885,
      "grad_norm": 10.75,
      "learning_rate": 4.3178451189987e-05,
      "loss": 1.1707,
      "num_input_tokens_seen": 5343632,
      "step": 9635
    },
    {
      "epoch": 169.12389380530973,
      "grad_norm": 7.5625,
      "learning_rate": 4.3171710176233315e-05,
      "loss": 1.1236,
      "num_input_tokens_seen": 5346976,
      "step": 9640
    },
    {
      "epoch": 169.21238938053096,
      "grad_norm": 6.75,
      "learning_rate": 4.316496636017355e-05,
      "loss": 1.0213,
      "num_input_tokens_seen": 5349712,
      "step": 9645
    },
    {
      "epoch": 169.30088495575222,
      "grad_norm": 8.25,
      "learning_rate": 4.315821974284771e-05,
      "loss": 1.1181,
      "num_input_tokens_seen": 5352464,
      "step": 9650
    },
    {
      "epoch": 169.38938053097345,
      "grad_norm": 9.4375,
      "learning_rate": 4.315147032529619e-05,
      "loss": 1.228,
      "num_input_tokens_seen": 5355696,
      "step": 9655
    },
    {
      "epoch": 169.47787610619469,
      "grad_norm": 7.8125,
      "learning_rate": 4.3144718108559845e-05,
      "loss": 1.017,
      "num_input_tokens_seen": 5358528,
      "step": 9660
    },
    {
      "epoch": 169.56637168141592,
      "grad_norm": 8.5625,
      "learning_rate": 4.3137963093679945e-05,
      "loss": 1.1087,
      "num_input_tokens_seen": 5361136,
      "step": 9665
    },
    {
      "epoch": 169.65486725663717,
      "grad_norm": 11.0625,
      "learning_rate": 4.31312052816982e-05,
      "loss": 1.0193,
      "num_input_tokens_seen": 5363584,
      "step": 9670
    },
    {
      "epoch": 169.7433628318584,
      "grad_norm": 6.78125,
      "learning_rate": 4.312444467365675e-05,
      "loss": 0.9607,
      "num_input_tokens_seen": 5366576,
      "step": 9675
    },
    {
      "epoch": 169.83185840707964,
      "grad_norm": 8.25,
      "learning_rate": 4.311768127059816e-05,
      "loss": 0.9949,
      "num_input_tokens_seen": 5369200,
      "step": 9680
    },
    {
      "epoch": 169.9203539823009,
      "grad_norm": 9.625,
      "learning_rate": 4.3110915073565444e-05,
      "loss": 1.1239,
      "num_input_tokens_seen": 5371408,
      "step": 9685
    },
    {
      "epoch": 170.0,
      "grad_norm": 9.6875,
      "learning_rate": 4.310414608360203e-05,
      "loss": 0.9057,
      "num_input_tokens_seen": 5374264,
      "step": 9690
    },
    {
      "epoch": 170.08849557522123,
      "grad_norm": 9.5,
      "learning_rate": 4.309737430175177e-05,
      "loss": 1.142,
      "num_input_tokens_seen": 5377064,
      "step": 9695
    },
    {
      "epoch": 170.1769911504425,
      "grad_norm": 8.625,
      "learning_rate": 4.309059972905897e-05,
      "loss": 1.0202,
      "num_input_tokens_seen": 5379928,
      "step": 9700
    },
    {
      "epoch": 170.26548672566372,
      "grad_norm": 7.65625,
      "learning_rate": 4.308382236656836e-05,
      "loss": 1.0401,
      "num_input_tokens_seen": 5382680,
      "step": 9705
    },
    {
      "epoch": 170.35398230088495,
      "grad_norm": 8.6875,
      "learning_rate": 4.307704221532507e-05,
      "loss": 1.1064,
      "num_input_tokens_seen": 5385992,
      "step": 9710
    },
    {
      "epoch": 170.44247787610618,
      "grad_norm": 6.78125,
      "learning_rate": 4.307025927637471e-05,
      "loss": 1.0012,
      "num_input_tokens_seen": 5389096,
      "step": 9715
    },
    {
      "epoch": 170.53097345132744,
      "grad_norm": 13.1875,
      "learning_rate": 4.306347355076328e-05,
      "loss": 0.9717,
      "num_input_tokens_seen": 5392120,
      "step": 9720
    },
    {
      "epoch": 170.61946902654867,
      "grad_norm": 7.25,
      "learning_rate": 4.305668503953724e-05,
      "loss": 0.9631,
      "num_input_tokens_seen": 5395112,
      "step": 9725
    },
    {
      "epoch": 170.7079646017699,
      "grad_norm": 6.6875,
      "learning_rate": 4.3049893743743436e-05,
      "loss": 0.9658,
      "num_input_tokens_seen": 5397688,
      "step": 9730
    },
    {
      "epoch": 170.79646017699116,
      "grad_norm": 9.375,
      "learning_rate": 4.304309966442919e-05,
      "loss": 1.2349,
      "num_input_tokens_seen": 5400168,
      "step": 9735
    },
    {
      "epoch": 170.8849557522124,
      "grad_norm": 10.0,
      "learning_rate": 4.303630280264224e-05,
      "loss": 1.0648,
      "num_input_tokens_seen": 5403032,
      "step": 9740
    },
    {
      "epoch": 170.97345132743362,
      "grad_norm": 7.03125,
      "learning_rate": 4.302950315943074e-05,
      "loss": 1.1,
      "num_input_tokens_seen": 5405336,
      "step": 9745
    },
    {
      "epoch": 171.05309734513276,
      "grad_norm": 7.6875,
      "learning_rate": 4.3022700735843275e-05,
      "loss": 1.0262,
      "num_input_tokens_seen": 5407744,
      "step": 9750
    },
    {
      "epoch": 171.141592920354,
      "grad_norm": 11.4375,
      "learning_rate": 4.301589553292887e-05,
      "loss": 0.9412,
      "num_input_tokens_seen": 5410512,
      "step": 9755
    },
    {
      "epoch": 171.23008849557522,
      "grad_norm": 8.125,
      "learning_rate": 4.300908755173697e-05,
      "loss": 0.9439,
      "num_input_tokens_seen": 5413216,
      "step": 9760
    },
    {
      "epoch": 171.31858407079645,
      "grad_norm": 10.3125,
      "learning_rate": 4.300227679331745e-05,
      "loss": 1.0947,
      "num_input_tokens_seen": 5416544,
      "step": 9765
    },
    {
      "epoch": 171.4070796460177,
      "grad_norm": 7.625,
      "learning_rate": 4.299546325872063e-05,
      "loss": 1.1379,
      "num_input_tokens_seen": 5419600,
      "step": 9770
    },
    {
      "epoch": 171.49557522123894,
      "grad_norm": 9.25,
      "learning_rate": 4.2988646948997225e-05,
      "loss": 1.2189,
      "num_input_tokens_seen": 5423072,
      "step": 9775
    },
    {
      "epoch": 171.58407079646017,
      "grad_norm": 8.9375,
      "learning_rate": 4.29818278651984e-05,
      "loss": 0.9954,
      "num_input_tokens_seen": 5425712,
      "step": 9780
    },
    {
      "epoch": 171.67256637168143,
      "grad_norm": 8.6875,
      "learning_rate": 4.297500600837574e-05,
      "loss": 1.0801,
      "num_input_tokens_seen": 5428704,
      "step": 9785
    },
    {
      "epoch": 171.76106194690266,
      "grad_norm": 6.78125,
      "learning_rate": 4.2968181379581276e-05,
      "loss": 1.034,
      "num_input_tokens_seen": 5431424,
      "step": 9790
    },
    {
      "epoch": 171.8495575221239,
      "grad_norm": 10.125,
      "learning_rate": 4.296135397986743e-05,
      "loss": 1.1928,
      "num_input_tokens_seen": 5434080,
      "step": 9795
    },
    {
      "epoch": 171.93805309734512,
      "grad_norm": 8.3125,
      "learning_rate": 4.295452381028709e-05,
      "loss": 1.0908,
      "num_input_tokens_seen": 5436384,
      "step": 9800
    },
    {
      "epoch": 171.93805309734512,
      "eval_loss": 1.0306485891342163,
      "eval_runtime": 0.933,
      "eval_samples_per_second": 26.795,
      "eval_steps_per_second": 13.934,
      "num_input_tokens_seen": 5436384,
      "step": 9800
    },
    {
      "epoch": 172.01769911504425,
      "grad_norm": 7.65625,
      "learning_rate": 4.294769087189354e-05,
      "loss": 0.9305,
      "num_input_tokens_seen": 5438704,
      "step": 9805
    },
    {
      "epoch": 172.10619469026548,
      "grad_norm": 6.4375,
      "learning_rate": 4.294085516574052e-05,
      "loss": 0.8123,
      "num_input_tokens_seen": 5442256,
      "step": 9810
    },
    {
      "epoch": 172.1946902654867,
      "grad_norm": 11.375,
      "learning_rate": 4.2934016692882176e-05,
      "loss": 1.2573,
      "num_input_tokens_seen": 5444752,
      "step": 9815
    },
    {
      "epoch": 172.28318584070797,
      "grad_norm": 7.03125,
      "learning_rate": 4.292717545437308e-05,
      "loss": 1.1188,
      "num_input_tokens_seen": 5447184,
      "step": 9820
    },
    {
      "epoch": 172.3716814159292,
      "grad_norm": 9.25,
      "learning_rate": 4.292033145126825e-05,
      "loss": 1.0862,
      "num_input_tokens_seen": 5449824,
      "step": 9825
    },
    {
      "epoch": 172.46017699115043,
      "grad_norm": 8.0,
      "learning_rate": 4.29134846846231e-05,
      "loss": 0.979,
      "num_input_tokens_seen": 5452736,
      "step": 9830
    },
    {
      "epoch": 172.5486725663717,
      "grad_norm": 8.0,
      "learning_rate": 4.29066351554935e-05,
      "loss": 1.1414,
      "num_input_tokens_seen": 5455520,
      "step": 9835
    },
    {
      "epoch": 172.63716814159292,
      "grad_norm": 10.375,
      "learning_rate": 4.289978286493574e-05,
      "loss": 1.0866,
      "num_input_tokens_seen": 5458192,
      "step": 9840
    },
    {
      "epoch": 172.72566371681415,
      "grad_norm": 8.0,
      "learning_rate": 4.28929278140065e-05,
      "loss": 1.2768,
      "num_input_tokens_seen": 5460944,
      "step": 9845
    },
    {
      "epoch": 172.81415929203538,
      "grad_norm": 7.09375,
      "learning_rate": 4.288607000376295e-05,
      "loss": 0.8734,
      "num_input_tokens_seen": 5463776,
      "step": 9850
    },
    {
      "epoch": 172.90265486725664,
      "grad_norm": 9.125,
      "learning_rate": 4.2879209435262624e-05,
      "loss": 1.0084,
      "num_input_tokens_seen": 5466640,
      "step": 9855
    },
    {
      "epoch": 172.99115044247787,
      "grad_norm": 9.125,
      "learning_rate": 4.287234610956353e-05,
      "loss": 0.8869,
      "num_input_tokens_seen": 5469680,
      "step": 9860
    },
    {
      "epoch": 173.07079646017698,
      "grad_norm": 6.34375,
      "learning_rate": 4.2865480027724056e-05,
      "loss": 1.1415,
      "num_input_tokens_seen": 5472224,
      "step": 9865
    },
    {
      "epoch": 173.15929203539824,
      "grad_norm": 7.875,
      "learning_rate": 4.285861119080306e-05,
      "loss": 1.0571,
      "num_input_tokens_seen": 5475088,
      "step": 9870
    },
    {
      "epoch": 173.24778761061947,
      "grad_norm": 9.5625,
      "learning_rate": 4.2851739599859784e-05,
      "loss": 1.0505,
      "num_input_tokens_seen": 5477632,
      "step": 9875
    },
    {
      "epoch": 173.3362831858407,
      "grad_norm": 8.125,
      "learning_rate": 4.2844865255953934e-05,
      "loss": 0.9945,
      "num_input_tokens_seen": 5480896,
      "step": 9880
    },
    {
      "epoch": 173.42477876106196,
      "grad_norm": 8.3125,
      "learning_rate": 4.2837988160145605e-05,
      "loss": 1.1737,
      "num_input_tokens_seen": 5483760,
      "step": 9885
    },
    {
      "epoch": 173.5132743362832,
      "grad_norm": 8.1875,
      "learning_rate": 4.2831108313495336e-05,
      "loss": 0.911,
      "num_input_tokens_seen": 5486400,
      "step": 9890
    },
    {
      "epoch": 173.60176991150442,
      "grad_norm": 5.96875,
      "learning_rate": 4.282422571706408e-05,
      "loss": 0.9703,
      "num_input_tokens_seen": 5488928,
      "step": 9895
    },
    {
      "epoch": 173.69026548672565,
      "grad_norm": 8.125,
      "learning_rate": 4.281734037191323e-05,
      "loss": 1.1215,
      "num_input_tokens_seen": 5491696,
      "step": 9900
    },
    {
      "epoch": 173.7787610619469,
      "grad_norm": 8.75,
      "learning_rate": 4.281045227910459e-05,
      "loss": 1.0413,
      "num_input_tokens_seen": 5494848,
      "step": 9905
    },
    {
      "epoch": 173.86725663716814,
      "grad_norm": 9.25,
      "learning_rate": 4.280356143970038e-05,
      "loss": 1.232,
      "num_input_tokens_seen": 5497616,
      "step": 9910
    },
    {
      "epoch": 173.95575221238937,
      "grad_norm": 7.6875,
      "learning_rate": 4.279666785476327e-05,
      "loss": 1.138,
      "num_input_tokens_seen": 5500432,
      "step": 9915
    },
    {
      "epoch": 174.0353982300885,
      "grad_norm": 7.84375,
      "learning_rate": 4.2789771525356325e-05,
      "loss": 0.8999,
      "num_input_tokens_seen": 5502496,
      "step": 9920
    },
    {
      "epoch": 174.12389380530973,
      "grad_norm": 8.75,
      "learning_rate": 4.2782872452543056e-05,
      "loss": 1.1012,
      "num_input_tokens_seen": 5505328,
      "step": 9925
    },
    {
      "epoch": 174.21238938053096,
      "grad_norm": 8.6875,
      "learning_rate": 4.2775970637387376e-05,
      "loss": 1.0638,
      "num_input_tokens_seen": 5508064,
      "step": 9930
    },
    {
      "epoch": 174.30088495575222,
      "grad_norm": 11.0625,
      "learning_rate": 4.276906608095363e-05,
      "loss": 0.9994,
      "num_input_tokens_seen": 5511072,
      "step": 9935
    },
    {
      "epoch": 174.38938053097345,
      "grad_norm": 7.625,
      "learning_rate": 4.276215878430661e-05,
      "loss": 1.0285,
      "num_input_tokens_seen": 5514128,
      "step": 9940
    },
    {
      "epoch": 174.47787610619469,
      "grad_norm": 9.0625,
      "learning_rate": 4.275524874851149e-05,
      "loss": 0.9537,
      "num_input_tokens_seen": 5516976,
      "step": 9945
    },
    {
      "epoch": 174.56637168141592,
      "grad_norm": 8.4375,
      "learning_rate": 4.274833597463388e-05,
      "loss": 0.9346,
      "num_input_tokens_seen": 5519632,
      "step": 9950
    },
    {
      "epoch": 174.65486725663717,
      "grad_norm": 8.1875,
      "learning_rate": 4.2741420463739824e-05,
      "loss": 1.1739,
      "num_input_tokens_seen": 5522752,
      "step": 9955
    },
    {
      "epoch": 174.7433628318584,
      "grad_norm": 6.84375,
      "learning_rate": 4.273450221689578e-05,
      "loss": 1.0092,
      "num_input_tokens_seen": 5525616,
      "step": 9960
    },
    {
      "epoch": 174.83185840707964,
      "grad_norm": 8.0,
      "learning_rate": 4.272758123516863e-05,
      "loss": 1.2335,
      "num_input_tokens_seen": 5528224,
      "step": 9965
    },
    {
      "epoch": 174.9203539823009,
      "grad_norm": 8.5,
      "learning_rate": 4.272065751962567e-05,
      "loss": 1.0643,
      "num_input_tokens_seen": 5531152,
      "step": 9970
    },
    {
      "epoch": 175.0,
      "grad_norm": 9.875,
      "learning_rate": 4.271373107133464e-05,
      "loss": 1.0847,
      "num_input_tokens_seen": 5533376,
      "step": 9975
    },
    {
      "epoch": 175.08849557522123,
      "grad_norm": 7.5,
      "learning_rate": 4.270680189136366e-05,
      "loss": 0.9312,
      "num_input_tokens_seen": 5536160,
      "step": 9980
    },
    {
      "epoch": 175.1769911504425,
      "grad_norm": 7.875,
      "learning_rate": 4.269986998078132e-05,
      "loss": 1.0935,
      "num_input_tokens_seen": 5538976,
      "step": 9985
    },
    {
      "epoch": 175.26548672566372,
      "grad_norm": 7.84375,
      "learning_rate": 4.2692935340656595e-05,
      "loss": 1.1428,
      "num_input_tokens_seen": 5541520,
      "step": 9990
    },
    {
      "epoch": 175.35398230088495,
      "grad_norm": 8.875,
      "learning_rate": 4.26859979720589e-05,
      "loss": 1.0828,
      "num_input_tokens_seen": 5544656,
      "step": 9995
    },
    {
      "epoch": 175.44247787610618,
      "grad_norm": 7.0,
      "learning_rate": 4.267905787605806e-05,
      "loss": 1.0437,
      "num_input_tokens_seen": 5547152,
      "step": 10000
    },
    {
      "epoch": 175.44247787610618,
      "eval_loss": 1.0295060873031616,
      "eval_runtime": 0.9368,
      "eval_samples_per_second": 26.688,
      "eval_steps_per_second": 13.878,
      "num_input_tokens_seen": 5547152,
      "step": 10000
    },
    {
      "epoch": 175.53097345132744,
      "grad_norm": 8.875,
      "learning_rate": 4.267211505372433e-05,
      "loss": 1.1484,
      "num_input_tokens_seen": 5549760,
      "step": 10005
    },
    {
      "epoch": 175.61946902654867,
      "grad_norm": 6.21875,
      "learning_rate": 4.266516950612837e-05,
      "loss": 1.0793,
      "num_input_tokens_seen": 5552832,
      "step": 10010
    },
    {
      "epoch": 175.7079646017699,
      "grad_norm": 7.21875,
      "learning_rate": 4.265822123434128e-05,
      "loss": 1.0192,
      "num_input_tokens_seen": 5556000,
      "step": 10015
    },
    {
      "epoch": 175.79646017699116,
      "grad_norm": 9.0,
      "learning_rate": 4.265127023943457e-05,
      "loss": 1.2056,
      "num_input_tokens_seen": 5558304,
      "step": 10020
    },
    {
      "epoch": 175.8849557522124,
      "grad_norm": 10.0625,
      "learning_rate": 4.2644316522480176e-05,
      "loss": 0.8886,
      "num_input_tokens_seen": 5561088,
      "step": 10025
    },
    {
      "epoch": 175.97345132743362,
      "grad_norm": 12.6875,
      "learning_rate": 4.263736008455044e-05,
      "loss": 1.0293,
      "num_input_tokens_seen": 5564384,
      "step": 10030
    },
    {
      "epoch": 176.05309734513276,
      "grad_norm": 6.65625,
      "learning_rate": 4.2630400926718125e-05,
      "loss": 0.8704,
      "num_input_tokens_seen": 5566944,
      "step": 10035
    },
    {
      "epoch": 176.141592920354,
      "grad_norm": 8.25,
      "learning_rate": 4.262343905005644e-05,
      "loss": 1.069,
      "num_input_tokens_seen": 5569664,
      "step": 10040
    },
    {
      "epoch": 176.23008849557522,
      "grad_norm": 7.46875,
      "learning_rate": 4.261647445563897e-05,
      "loss": 0.8828,
      "num_input_tokens_seen": 5572720,
      "step": 10045
    },
    {
      "epoch": 176.31858407079645,
      "grad_norm": 6.71875,
      "learning_rate": 4.260950714453976e-05,
      "loss": 0.9564,
      "num_input_tokens_seen": 5575744,
      "step": 10050
    },
    {
      "epoch": 176.4070796460177,
      "grad_norm": 9.0625,
      "learning_rate": 4.2602537117833266e-05,
      "loss": 1.1911,
      "num_input_tokens_seen": 5578368,
      "step": 10055
    },
    {
      "epoch": 176.49557522123894,
      "grad_norm": 6.5,
      "learning_rate": 4.259556437659433e-05,
      "loss": 1.1246,
      "num_input_tokens_seen": 5581120,
      "step": 10060
    },
    {
      "epoch": 176.58407079646017,
      "grad_norm": 6.875,
      "learning_rate": 4.258858892189825e-05,
      "loss": 0.9227,
      "num_input_tokens_seen": 5584560,
      "step": 10065
    },
    {
      "epoch": 176.67256637168143,
      "grad_norm": 7.75,
      "learning_rate": 4.2581610754820725e-05,
      "loss": 0.9659,
      "num_input_tokens_seen": 5587760,
      "step": 10070
    },
    {
      "epoch": 176.76106194690266,
      "grad_norm": 8.375,
      "learning_rate": 4.2574629876437876e-05,
      "loss": 1.2664,
      "num_input_tokens_seen": 5590144,
      "step": 10075
    },
    {
      "epoch": 176.8495575221239,
      "grad_norm": 9.6875,
      "learning_rate": 4.256764628782625e-05,
      "loss": 0.9948,
      "num_input_tokens_seen": 5593072,
      "step": 10080
    },
    {
      "epoch": 176.93805309734512,
      "grad_norm": 8.8125,
      "learning_rate": 4.256065999006279e-05,
      "loss": 1.0837,
      "num_input_tokens_seen": 5595728,
      "step": 10085
    },
    {
      "epoch": 177.01769911504425,
      "grad_norm": 11.3125,
      "learning_rate": 4.2553670984224885e-05,
      "loss": 1.304,
      "num_input_tokens_seen": 5597704,
      "step": 10090
    },
    {
      "epoch": 177.10619469026548,
      "grad_norm": 6.03125,
      "learning_rate": 4.254667927139032e-05,
      "loss": 1.0065,
      "num_input_tokens_seen": 5600360,
      "step": 10095
    },
    {
      "epoch": 177.1946902654867,
      "grad_norm": 10.9375,
      "learning_rate": 4.2539684852637295e-05,
      "loss": 1.2481,
      "num_input_tokens_seen": 5602680,
      "step": 10100
    },
    {
      "epoch": 177.28318584070797,
      "grad_norm": 7.71875,
      "learning_rate": 4.253268772904446e-05,
      "loss": 1.243,
      "num_input_tokens_seen": 5605576,
      "step": 10105
    },
    {
      "epoch": 177.3716814159292,
      "grad_norm": 8.4375,
      "learning_rate": 4.252568790169085e-05,
      "loss": 0.9013,
      "num_input_tokens_seen": 5608440,
      "step": 10110
    },
    {
      "epoch": 177.46017699115043,
      "grad_norm": 6.625,
      "learning_rate": 4.251868537165592e-05,
      "loss": 1.0371,
      "num_input_tokens_seen": 5611336,
      "step": 10115
    },
    {
      "epoch": 177.5486725663717,
      "grad_norm": 7.0,
      "learning_rate": 4.251168014001955e-05,
      "loss": 0.8711,
      "num_input_tokens_seen": 5614264,
      "step": 10120
    },
    {
      "epoch": 177.63716814159292,
      "grad_norm": 7.4375,
      "learning_rate": 4.250467220786204e-05,
      "loss": 1.091,
      "num_input_tokens_seen": 5617192,
      "step": 10125
    },
    {
      "epoch": 177.72566371681415,
      "grad_norm": 7.84375,
      "learning_rate": 4.249766157626409e-05,
      "loss": 0.9223,
      "num_input_tokens_seen": 5620280,
      "step": 10130
    },
    {
      "epoch": 177.81415929203538,
      "grad_norm": 8.5,
      "learning_rate": 4.249064824630684e-05,
      "loss": 0.9338,
      "num_input_tokens_seen": 5623016,
      "step": 10135
    },
    {
      "epoch": 177.90265486725664,
      "grad_norm": 8.1875,
      "learning_rate": 4.248363221907183e-05,
      "loss": 1.1489,
      "num_input_tokens_seen": 5625528,
      "step": 10140
    },
    {
      "epoch": 177.99115044247787,
      "grad_norm": 9.5,
      "learning_rate": 4.2476613495641026e-05,
      "loss": 1.1603,
      "num_input_tokens_seen": 5628392,
      "step": 10145
    },
    {
      "epoch": 178.07079646017698,
      "grad_norm": 8.125,
      "learning_rate": 4.246959207709679e-05,
      "loss": 1.011,
      "num_input_tokens_seen": 5630848,
      "step": 10150
    },
    {
      "epoch": 178.15929203539824,
      "grad_norm": 8.4375,
      "learning_rate": 4.246256796452192e-05,
      "loss": 0.894,
      "num_input_tokens_seen": 5633440,
      "step": 10155
    },
    {
      "epoch": 178.24778761061947,
      "grad_norm": 7.8125,
      "learning_rate": 4.245554115899962e-05,
      "loss": 0.9872,
      "num_input_tokens_seen": 5636400,
      "step": 10160
    },
    {
      "epoch": 178.3362831858407,
      "grad_norm": 7.03125,
      "learning_rate": 4.2448511661613514e-05,
      "loss": 1.1289,
      "num_input_tokens_seen": 5639088,
      "step": 10165
    },
    {
      "epoch": 178.42477876106196,
      "grad_norm": 11.0,
      "learning_rate": 4.2441479473447635e-05,
      "loss": 1.0659,
      "num_input_tokens_seen": 5642160,
      "step": 10170
    },
    {
      "epoch": 178.5132743362832,
      "grad_norm": 12.0,
      "learning_rate": 4.243444459558644e-05,
      "loss": 1.1893,
      "num_input_tokens_seen": 5644592,
      "step": 10175
    },
    {
      "epoch": 178.60176991150442,
      "grad_norm": 8.4375,
      "learning_rate": 4.24274070291148e-05,
      "loss": 1.133,
      "num_input_tokens_seen": 5647328,
      "step": 10180
    },
    {
      "epoch": 178.69026548672565,
      "grad_norm": 9.4375,
      "learning_rate": 4.242036677511798e-05,
      "loss": 1.1592,
      "num_input_tokens_seen": 5650176,
      "step": 10185
    },
    {
      "epoch": 178.7787610619469,
      "grad_norm": 7.78125,
      "learning_rate": 4.241332383468169e-05,
      "loss": 1.0359,
      "num_input_tokens_seen": 5653216,
      "step": 10190
    },
    {
      "epoch": 178.86725663716814,
      "grad_norm": 8.5625,
      "learning_rate": 4.2406278208892034e-05,
      "loss": 1.1047,
      "num_input_tokens_seen": 5655840,
      "step": 10195
    },
    {
      "epoch": 178.95575221238937,
      "grad_norm": 7.59375,
      "learning_rate": 4.2399229898835536e-05,
      "loss": 0.9347,
      "num_input_tokens_seen": 5658656,
      "step": 10200
    },
    {
      "epoch": 178.95575221238937,
      "eval_loss": 1.025412678718567,
      "eval_runtime": 0.9365,
      "eval_samples_per_second": 26.695,
      "eval_steps_per_second": 13.881,
      "num_input_tokens_seen": 5658656,
      "step": 10200
    },
    {
      "epoch": 179.0353982300885,
      "grad_norm": 9.5,
      "learning_rate": 4.239217890559914e-05,
      "loss": 1.1956,
      "num_input_tokens_seen": 5661056,
      "step": 10205
    },
    {
      "epoch": 179.12389380530973,
      "grad_norm": 8.375,
      "learning_rate": 4.238512523027019e-05,
      "loss": 1.128,
      "num_input_tokens_seen": 5664192,
      "step": 10210
    },
    {
      "epoch": 179.21238938053096,
      "grad_norm": 8.25,
      "learning_rate": 4.237806887393645e-05,
      "loss": 0.9834,
      "num_input_tokens_seen": 5667168,
      "step": 10215
    },
    {
      "epoch": 179.30088495575222,
      "grad_norm": 7.75,
      "learning_rate": 4.237100983768611e-05,
      "loss": 1.0366,
      "num_input_tokens_seen": 5669680,
      "step": 10220
    },
    {
      "epoch": 179.38938053097345,
      "grad_norm": 7.8125,
      "learning_rate": 4.2363948122607756e-05,
      "loss": 0.9815,
      "num_input_tokens_seen": 5672256,
      "step": 10225
    },
    {
      "epoch": 179.47787610619469,
      "grad_norm": 9.1875,
      "learning_rate": 4.235688372979039e-05,
      "loss": 1.2119,
      "num_input_tokens_seen": 5674928,
      "step": 10230
    },
    {
      "epoch": 179.56637168141592,
      "grad_norm": 8.25,
      "learning_rate": 4.234981666032343e-05,
      "loss": 0.9823,
      "num_input_tokens_seen": 5678016,
      "step": 10235
    },
    {
      "epoch": 179.65486725663717,
      "grad_norm": 9.1875,
      "learning_rate": 4.2342746915296704e-05,
      "loss": 1.1545,
      "num_input_tokens_seen": 5680768,
      "step": 10240
    },
    {
      "epoch": 179.7433628318584,
      "grad_norm": 9.5,
      "learning_rate": 4.233567449580047e-05,
      "loss": 0.985,
      "num_input_tokens_seen": 5683920,
      "step": 10245
    },
    {
      "epoch": 179.83185840707964,
      "grad_norm": 8.625,
      "learning_rate": 4.232859940292537e-05,
      "loss": 1.0628,
      "num_input_tokens_seen": 5686080,
      "step": 10250
    },
    {
      "epoch": 179.9203539823009,
      "grad_norm": 10.25,
      "learning_rate": 4.232152163776248e-05,
      "loss": 0.8973,
      "num_input_tokens_seen": 5688656,
      "step": 10255
    },
    {
      "epoch": 180.0,
      "grad_norm": 26.5,
      "learning_rate": 4.231444120140328e-05,
      "loss": 1.4854,
      "num_input_tokens_seen": 5691408,
      "step": 10260
    },
    {
      "epoch": 180.08849557522123,
      "grad_norm": 7.625,
      "learning_rate": 4.230735809493967e-05,
      "loss": 1.2142,
      "num_input_tokens_seen": 5694048,
      "step": 10265
    },
    {
      "epoch": 180.1769911504425,
      "grad_norm": 9.0,
      "learning_rate": 4.2300272319463926e-05,
      "loss": 1.0779,
      "num_input_tokens_seen": 5696624,
      "step": 10270
    },
    {
      "epoch": 180.26548672566372,
      "grad_norm": 9.0,
      "learning_rate": 4.2293183876068786e-05,
      "loss": 1.1626,
      "num_input_tokens_seen": 5699968,
      "step": 10275
    },
    {
      "epoch": 180.35398230088495,
      "grad_norm": 6.75,
      "learning_rate": 4.228609276584737e-05,
      "loss": 0.7538,
      "num_input_tokens_seen": 5703232,
      "step": 10280
    },
    {
      "epoch": 180.44247787610618,
      "grad_norm": 7.875,
      "learning_rate": 4.227899898989323e-05,
      "loss": 1.0789,
      "num_input_tokens_seen": 5706064,
      "step": 10285
    },
    {
      "epoch": 180.53097345132744,
      "grad_norm": 8.75,
      "learning_rate": 4.2271902549300293e-05,
      "loss": 1.0201,
      "num_input_tokens_seen": 5708976,
      "step": 10290
    },
    {
      "epoch": 180.61946902654867,
      "grad_norm": 8.3125,
      "learning_rate": 4.226480344516294e-05,
      "loss": 1.0588,
      "num_input_tokens_seen": 5711936,
      "step": 10295
    },
    {
      "epoch": 180.7079646017699,
      "grad_norm": 6.03125,
      "learning_rate": 4.2257701678575925e-05,
      "loss": 0.9659,
      "num_input_tokens_seen": 5714496,
      "step": 10300
    },
    {
      "epoch": 180.79646017699116,
      "grad_norm": 9.6875,
      "learning_rate": 4.225059725063444e-05,
      "loss": 1.024,
      "num_input_tokens_seen": 5717584,
      "step": 10305
    },
    {
      "epoch": 180.8849557522124,
      "grad_norm": 10.875,
      "learning_rate": 4.2243490162434074e-05,
      "loss": 1.0832,
      "num_input_tokens_seen": 5720016,
      "step": 10310
    },
    {
      "epoch": 180.97345132743362,
      "grad_norm": 8.8125,
      "learning_rate": 4.223638041507083e-05,
      "loss": 1.0852,
      "num_input_tokens_seen": 5722752,
      "step": 10315
    },
    {
      "epoch": 181.05309734513276,
      "grad_norm": 13.25,
      "learning_rate": 4.2229268009641124e-05,
      "loss": 1.1887,
      "num_input_tokens_seen": 5725096,
      "step": 10320
    },
    {
      "epoch": 181.141592920354,
      "grad_norm": 10.125,
      "learning_rate": 4.222215294724177e-05,
      "loss": 0.9199,
      "num_input_tokens_seen": 5727896,
      "step": 10325
    },
    {
      "epoch": 181.23008849557522,
      "grad_norm": 7.625,
      "learning_rate": 4.2215035228970005e-05,
      "loss": 1.0297,
      "num_input_tokens_seen": 5730888,
      "step": 10330
    },
    {
      "epoch": 181.31858407079645,
      "grad_norm": 7.625,
      "learning_rate": 4.2207914855923464e-05,
      "loss": 1.0997,
      "num_input_tokens_seen": 5733352,
      "step": 10335
    },
    {
      "epoch": 181.4070796460177,
      "grad_norm": 10.0,
      "learning_rate": 4.220079182920021e-05,
      "loss": 1.1739,
      "num_input_tokens_seen": 5735928,
      "step": 10340
    },
    {
      "epoch": 181.49557522123894,
      "grad_norm": 7.78125,
      "learning_rate": 4.2193666149898705e-05,
      "loss": 0.9944,
      "num_input_tokens_seen": 5738584,
      "step": 10345
    },
    {
      "epoch": 181.58407079646017,
      "grad_norm": 11.3125,
      "learning_rate": 4.21865378191178e-05,
      "loss": 1.0846,
      "num_input_tokens_seen": 5741672,
      "step": 10350
    },
    {
      "epoch": 181.67256637168143,
      "grad_norm": 9.6875,
      "learning_rate": 4.217940683795678e-05,
      "loss": 1.1114,
      "num_input_tokens_seen": 5744136,
      "step": 10355
    },
    {
      "epoch": 181.76106194690266,
      "grad_norm": 8.75,
      "learning_rate": 4.217227320751534e-05,
      "loss": 1.0643,
      "num_input_tokens_seen": 5746872,
      "step": 10360
    },
    {
      "epoch": 181.8495575221239,
      "grad_norm": 7.25,
      "learning_rate": 4.216513692889358e-05,
      "loss": 1.0428,
      "num_input_tokens_seen": 5749624,
      "step": 10365
    },
    {
      "epoch": 181.93805309734512,
      "grad_norm": 7.53125,
      "learning_rate": 4.215799800319199e-05,
      "loss": 1.0521,
      "num_input_tokens_seen": 5752728,
      "step": 10370
    },
    {
      "epoch": 182.01769911504425,
      "grad_norm": 9.0625,
      "learning_rate": 4.2150856431511485e-05,
      "loss": 0.9548,
      "num_input_tokens_seen": 5755208,
      "step": 10375
    },
    {
      "epoch": 182.10619469026548,
      "grad_norm": 12.9375,
      "learning_rate": 4.214371221495339e-05,
      "loss": 1.142,
      "num_input_tokens_seen": 5757832,
      "step": 10380
    },
    {
      "epoch": 182.1946902654867,
      "grad_norm": 8.5625,
      "learning_rate": 4.213656535461942e-05,
      "loss": 0.9088,
      "num_input_tokens_seen": 5760408,
      "step": 10385
    },
    {
      "epoch": 182.28318584070797,
      "grad_norm": 8.375,
      "learning_rate": 4.2129415851611734e-05,
      "loss": 0.9843,
      "num_input_tokens_seen": 5762952,
      "step": 10390
    },
    {
      "epoch": 182.3716814159292,
      "grad_norm": 11.5,
      "learning_rate": 4.2122263707032855e-05,
      "loss": 0.9455,
      "num_input_tokens_seen": 5765816,
      "step": 10395
    },
    {
      "epoch": 182.46017699115043,
      "grad_norm": 8.375,
      "learning_rate": 4.211510892198574e-05,
      "loss": 1.1187,
      "num_input_tokens_seen": 5768616,
      "step": 10400
    },
    {
      "epoch": 182.46017699115043,
      "eval_loss": 1.0259926319122314,
      "eval_runtime": 0.9421,
      "eval_samples_per_second": 26.536,
      "eval_steps_per_second": 13.798,
      "num_input_tokens_seen": 5768616,
      "step": 10400
    },
    {
      "epoch": 182.5486725663717,
      "grad_norm": 7.9375,
      "learning_rate": 4.210795149757375e-05,
      "loss": 0.9563,
      "num_input_tokens_seen": 5771656,
      "step": 10405
    },
    {
      "epoch": 182.63716814159292,
      "grad_norm": 10.25,
      "learning_rate": 4.210079143490065e-05,
      "loss": 1.0728,
      "num_input_tokens_seen": 5774632,
      "step": 10410
    },
    {
      "epoch": 182.72566371681415,
      "grad_norm": 14.0625,
      "learning_rate": 4.2093628735070604e-05,
      "loss": 1.1559,
      "num_input_tokens_seen": 5777160,
      "step": 10415
    },
    {
      "epoch": 182.81415929203538,
      "grad_norm": 7.875,
      "learning_rate": 4.208646339918819e-05,
      "loss": 1.0827,
      "num_input_tokens_seen": 5780072,
      "step": 10420
    },
    {
      "epoch": 182.90265486725664,
      "grad_norm": 8.4375,
      "learning_rate": 4.2079295428358414e-05,
      "loss": 1.0881,
      "num_input_tokens_seen": 5783032,
      "step": 10425
    },
    {
      "epoch": 182.99115044247787,
      "grad_norm": 6.3125,
      "learning_rate": 4.207212482368664e-05,
      "loss": 1.1406,
      "num_input_tokens_seen": 5785736,
      "step": 10430
    },
    {
      "epoch": 183.07079646017698,
      "grad_norm": 7.6875,
      "learning_rate": 4.206495158627867e-05,
      "loss": 1.1064,
      "num_input_tokens_seen": 5788192,
      "step": 10435
    },
    {
      "epoch": 183.15929203539824,
      "grad_norm": 8.125,
      "learning_rate": 4.205777571724073e-05,
      "loss": 1.0003,
      "num_input_tokens_seen": 5791328,
      "step": 10440
    },
    {
      "epoch": 183.24778761061947,
      "grad_norm": 7.9375,
      "learning_rate": 4.20505972176794e-05,
      "loss": 1.0453,
      "num_input_tokens_seen": 5794048,
      "step": 10445
    },
    {
      "epoch": 183.3362831858407,
      "grad_norm": 9.875,
      "learning_rate": 4.204341608870171e-05,
      "loss": 1.2215,
      "num_input_tokens_seen": 5797264,
      "step": 10450
    },
    {
      "epoch": 183.42477876106196,
      "grad_norm": 6.96875,
      "learning_rate": 4.203623233141508e-05,
      "loss": 0.9218,
      "num_input_tokens_seen": 5800224,
      "step": 10455
    },
    {
      "epoch": 183.5132743362832,
      "grad_norm": 8.625,
      "learning_rate": 4.2029045946927334e-05,
      "loss": 1.0457,
      "num_input_tokens_seen": 5802576,
      "step": 10460
    },
    {
      "epoch": 183.60176991150442,
      "grad_norm": 7.78125,
      "learning_rate": 4.20218569363467e-05,
      "loss": 1.2569,
      "num_input_tokens_seen": 5805152,
      "step": 10465
    },
    {
      "epoch": 183.69026548672565,
      "grad_norm": 11.1875,
      "learning_rate": 4.2014665300781834e-05,
      "loss": 0.962,
      "num_input_tokens_seen": 5808112,
      "step": 10470
    },
    {
      "epoch": 183.7787610619469,
      "grad_norm": 9.0,
      "learning_rate": 4.200747104134174e-05,
      "loss": 1.0077,
      "num_input_tokens_seen": 5810992,
      "step": 10475
    },
    {
      "epoch": 183.86725663716814,
      "grad_norm": 8.375,
      "learning_rate": 4.200027415913588e-05,
      "loss": 1.127,
      "num_input_tokens_seen": 5813536,
      "step": 10480
    },
    {
      "epoch": 183.95575221238937,
      "grad_norm": 7.125,
      "learning_rate": 4.1993074655274126e-05,
      "loss": 0.9541,
      "num_input_tokens_seen": 5816064,
      "step": 10485
    },
    {
      "epoch": 184.0353982300885,
      "grad_norm": 6.125,
      "learning_rate": 4.198587253086669e-05,
      "loss": 1.0367,
      "num_input_tokens_seen": 5818432,
      "step": 10490
    },
    {
      "epoch": 184.12389380530973,
      "grad_norm": 9.5625,
      "learning_rate": 4.197866778702426e-05,
      "loss": 1.0733,
      "num_input_tokens_seen": 5821136,
      "step": 10495
    },
    {
      "epoch": 184.21238938053096,
      "grad_norm": 9.0625,
      "learning_rate": 4.197146042485789e-05,
      "loss": 1.2912,
      "num_input_tokens_seen": 5823744,
      "step": 10500
    },
    {
      "epoch": 184.30088495575222,
      "grad_norm": 8.375,
      "learning_rate": 4.1964250445479046e-05,
      "loss": 0.8222,
      "num_input_tokens_seen": 5827136,
      "step": 10505
    },
    {
      "epoch": 184.38938053097345,
      "grad_norm": 10.75,
      "learning_rate": 4.19570378499996e-05,
      "loss": 1.0683,
      "num_input_tokens_seen": 5830288,
      "step": 10510
    },
    {
      "epoch": 184.47787610619469,
      "grad_norm": 8.375,
      "learning_rate": 4.194982263953182e-05,
      "loss": 1.1019,
      "num_input_tokens_seen": 5832912,
      "step": 10515
    },
    {
      "epoch": 184.56637168141592,
      "grad_norm": 7.75,
      "learning_rate": 4.194260481518838e-05,
      "loss": 0.9834,
      "num_input_tokens_seen": 5835552,
      "step": 10520
    },
    {
      "epoch": 184.65486725663717,
      "grad_norm": 5.625,
      "learning_rate": 4.1935384378082366e-05,
      "loss": 1.001,
      "num_input_tokens_seen": 5838272,
      "step": 10525
    },
    {
      "epoch": 184.7433628318584,
      "grad_norm": 7.03125,
      "learning_rate": 4.1928161329327267e-05,
      "loss": 1.1385,
      "num_input_tokens_seen": 5840848,
      "step": 10530
    },
    {
      "epoch": 184.83185840707964,
      "grad_norm": 7.5,
      "learning_rate": 4.1920935670036945e-05,
      "loss": 1.1199,
      "num_input_tokens_seen": 5843392,
      "step": 10535
    },
    {
      "epoch": 184.9203539823009,
      "grad_norm": 6.6875,
      "learning_rate": 4.1913707401325705e-05,
      "loss": 0.9873,
      "num_input_tokens_seen": 5845872,
      "step": 10540
    },
    {
      "epoch": 185.0,
      "grad_norm": 12.375,
      "learning_rate": 4.1906476524308235e-05,
      "loss": 1.059,
      "num_input_tokens_seen": 5848600,
      "step": 10545
    },
    {
      "epoch": 185.08849557522123,
      "grad_norm": 10.9375,
      "learning_rate": 4.189924304009962e-05,
      "loss": 0.9655,
      "num_input_tokens_seen": 5851496,
      "step": 10550
    },
    {
      "epoch": 185.1769911504425,
      "grad_norm": 8.125,
      "learning_rate": 4.189200694981537e-05,
      "loss": 0.9395,
      "num_input_tokens_seen": 5854616,
      "step": 10555
    },
    {
      "epoch": 185.26548672566372,
      "grad_norm": 9.125,
      "learning_rate": 4.188476825457136e-05,
      "loss": 1.0417,
      "num_input_tokens_seen": 5857512,
      "step": 10560
    },
    {
      "epoch": 185.35398230088495,
      "grad_norm": 8.875,
      "learning_rate": 4.18775269554839e-05,
      "loss": 1.0099,
      "num_input_tokens_seen": 5860072,
      "step": 10565
    },
    {
      "epoch": 185.44247787610618,
      "grad_norm": 8.5,
      "learning_rate": 4.187028305366969e-05,
      "loss": 1.0242,
      "num_input_tokens_seen": 5862904,
      "step": 10570
    },
    {
      "epoch": 185.53097345132744,
      "grad_norm": 8.625,
      "learning_rate": 4.1863036550245824e-05,
      "loss": 1.0688,
      "num_input_tokens_seen": 5865736,
      "step": 10575
    },
    {
      "epoch": 185.61946902654867,
      "grad_norm": 7.375,
      "learning_rate": 4.1855787446329806e-05,
      "loss": 1.0653,
      "num_input_tokens_seen": 5868600,
      "step": 10580
    },
    {
      "epoch": 185.7079646017699,
      "grad_norm": 7.75,
      "learning_rate": 4.184853574303955e-05,
      "loss": 1.135,
      "num_input_tokens_seen": 5871224,
      "step": 10585
    },
    {
      "epoch": 185.79646017699116,
      "grad_norm": 9.0,
      "learning_rate": 4.184128144149334e-05,
      "loss": 1.0621,
      "num_input_tokens_seen": 5873720,
      "step": 10590
    },
    {
      "epoch": 185.8849557522124,
      "grad_norm": 6.03125,
      "learning_rate": 4.1834024542809896e-05,
      "loss": 1.0072,
      "num_input_tokens_seen": 5876552,
      "step": 10595
    },
    {
      "epoch": 185.97345132743362,
      "grad_norm": 8.125,
      "learning_rate": 4.1826765048108315e-05,
      "loss": 1.1472,
      "num_input_tokens_seen": 5879304,
      "step": 10600
    },
    {
      "epoch": 185.97345132743362,
      "eval_loss": 1.0324807167053223,
      "eval_runtime": 0.9349,
      "eval_samples_per_second": 26.741,
      "eval_steps_per_second": 13.905,
      "num_input_tokens_seen": 5879304,
      "step": 10600
    },
    {
      "epoch": 186.05309734513276,
      "grad_norm": 9.0,
      "learning_rate": 4.181950295850811e-05,
      "loss": 1.3013,
      "num_input_tokens_seen": 5881280,
      "step": 10605
    },
    {
      "epoch": 186.141592920354,
      "grad_norm": 8.75,
      "learning_rate": 4.181223827512918e-05,
      "loss": 1.0544,
      "num_input_tokens_seen": 5884096,
      "step": 10610
    },
    {
      "epoch": 186.23008849557522,
      "grad_norm": 8.125,
      "learning_rate": 4.180497099909183e-05,
      "loss": 1.2089,
      "num_input_tokens_seen": 5886976,
      "step": 10615
    },
    {
      "epoch": 186.31858407079645,
      "grad_norm": 5.53125,
      "learning_rate": 4.179770113151677e-05,
      "loss": 0.9388,
      "num_input_tokens_seen": 5889728,
      "step": 10620
    },
    {
      "epoch": 186.4070796460177,
      "grad_norm": 10.5,
      "learning_rate": 4.179042867352511e-05,
      "loss": 1.0652,
      "num_input_tokens_seen": 5892320,
      "step": 10625
    },
    {
      "epoch": 186.49557522123894,
      "grad_norm": 8.0625,
      "learning_rate": 4.1783153626238334e-05,
      "loss": 0.9501,
      "num_input_tokens_seen": 5895248,
      "step": 10630
    },
    {
      "epoch": 186.58407079646017,
      "grad_norm": 8.625,
      "learning_rate": 4.177587599077836e-05,
      "loss": 0.9814,
      "num_input_tokens_seen": 5898048,
      "step": 10635
    },
    {
      "epoch": 186.67256637168143,
      "grad_norm": 8.625,
      "learning_rate": 4.1768595768267494e-05,
      "loss": 1.0883,
      "num_input_tokens_seen": 5900848,
      "step": 10640
    },
    {
      "epoch": 186.76106194690266,
      "grad_norm": 8.25,
      "learning_rate": 4.176131295982843e-05,
      "loss": 0.9677,
      "num_input_tokens_seen": 5903600,
      "step": 10645
    },
    {
      "epoch": 186.8495575221239,
      "grad_norm": 7.59375,
      "learning_rate": 4.1754027566584276e-05,
      "loss": 0.9936,
      "num_input_tokens_seen": 5907024,
      "step": 10650
    },
    {
      "epoch": 186.93805309734512,
      "grad_norm": 11.5,
      "learning_rate": 4.174673958965852e-05,
      "loss": 1.2503,
      "num_input_tokens_seen": 5909808,
      "step": 10655
    },
    {
      "epoch": 187.01769911504425,
      "grad_norm": 8.6875,
      "learning_rate": 4.173944903017507e-05,
      "loss": 1.0272,
      "num_input_tokens_seen": 5912320,
      "step": 10660
    },
    {
      "epoch": 187.10619469026548,
      "grad_norm": 8.0625,
      "learning_rate": 4.173215588925822e-05,
      "loss": 0.9339,
      "num_input_tokens_seen": 5914672,
      "step": 10665
    },
    {
      "epoch": 187.1946902654867,
      "grad_norm": 6.9375,
      "learning_rate": 4.172486016803266e-05,
      "loss": 0.8601,
      "num_input_tokens_seen": 5917968,
      "step": 10670
    },
    {
      "epoch": 187.28318584070797,
      "grad_norm": 7.84375,
      "learning_rate": 4.171756186762349e-05,
      "loss": 1.1168,
      "num_input_tokens_seen": 5920864,
      "step": 10675
    },
    {
      "epoch": 187.3716814159292,
      "grad_norm": 7.21875,
      "learning_rate": 4.171026098915619e-05,
      "loss": 1.0374,
      "num_input_tokens_seen": 5923728,
      "step": 10680
    },
    {
      "epoch": 187.46017699115043,
      "grad_norm": 11.0625,
      "learning_rate": 4.170295753375665e-05,
      "loss": 1.1692,
      "num_input_tokens_seen": 5926608,
      "step": 10685
    },
    {
      "epoch": 187.5486725663717,
      "grad_norm": 8.3125,
      "learning_rate": 4.169565150255117e-05,
      "loss": 1.1672,
      "num_input_tokens_seen": 5929024,
      "step": 10690
    },
    {
      "epoch": 187.63716814159292,
      "grad_norm": 7.6875,
      "learning_rate": 4.16883428966664e-05,
      "loss": 1.1004,
      "num_input_tokens_seen": 5931904,
      "step": 10695
    },
    {
      "epoch": 187.72566371681415,
      "grad_norm": 6.625,
      "learning_rate": 4.168103171722944e-05,
      "loss": 1.11,
      "num_input_tokens_seen": 5934560,
      "step": 10700
    },
    {
      "epoch": 187.81415929203538,
      "grad_norm": 11.8125,
      "learning_rate": 4.167371796536777e-05,
      "loss": 0.8643,
      "num_input_tokens_seen": 5937456,
      "step": 10705
    },
    {
      "epoch": 187.90265486725664,
      "grad_norm": 7.75,
      "learning_rate": 4.166640164220924e-05,
      "loss": 1.1401,
      "num_input_tokens_seen": 5940528,
      "step": 10710
    },
    {
      "epoch": 187.99115044247787,
      "grad_norm": 8.375,
      "learning_rate": 4.1659082748882144e-05,
      "loss": 1.1355,
      "num_input_tokens_seen": 5943136,
      "step": 10715
    },
    {
      "epoch": 188.07079646017698,
      "grad_norm": 6.21875,
      "learning_rate": 4.1651761286515135e-05,
      "loss": 0.713,
      "num_input_tokens_seen": 5945576,
      "step": 10720
    },
    {
      "epoch": 188.15929203539824,
      "grad_norm": 7.71875,
      "learning_rate": 4.164443725623728e-05,
      "loss": 0.9429,
      "num_input_tokens_seen": 5948968,
      "step": 10725
    },
    {
      "epoch": 188.24778761061947,
      "grad_norm": 8.125,
      "learning_rate": 4.163711065917802e-05,
      "loss": 1.0443,
      "num_input_tokens_seen": 5951512,
      "step": 10730
    },
    {
      "epoch": 188.3362831858407,
      "grad_norm": 10.9375,
      "learning_rate": 4.1629781496467234e-05,
      "loss": 1.226,
      "num_input_tokens_seen": 5954184,
      "step": 10735
    },
    {
      "epoch": 188.42477876106196,
      "grad_norm": 8.5,
      "learning_rate": 4.1622449769235164e-05,
      "loss": 0.8028,
      "num_input_tokens_seen": 5957672,
      "step": 10740
    },
    {
      "epoch": 188.5132743362832,
      "grad_norm": 7.59375,
      "learning_rate": 4.161511547861243e-05,
      "loss": 1.0155,
      "num_input_tokens_seen": 5960632,
      "step": 10745
    },
    {
      "epoch": 188.60176991150442,
      "grad_norm": 8.25,
      "learning_rate": 4.1607778625730104e-05,
      "loss": 1.1924,
      "num_input_tokens_seen": 5963432,
      "step": 10750
    },
    {
      "epoch": 188.69026548672565,
      "grad_norm": 8.1875,
      "learning_rate": 4.160043921171961e-05,
      "loss": 1.112,
      "num_input_tokens_seen": 5966360,
      "step": 10755
    },
    {
      "epoch": 188.7787610619469,
      "grad_norm": 8.375,
      "learning_rate": 4.159309723771276e-05,
      "loss": 0.9653,
      "num_input_tokens_seen": 5968568,
      "step": 10760
    },
    {
      "epoch": 188.86725663716814,
      "grad_norm": 7.125,
      "learning_rate": 4.158575270484181e-05,
      "loss": 1.215,
      "num_input_tokens_seen": 5970984,
      "step": 10765
    },
    {
      "epoch": 188.95575221238937,
      "grad_norm": 14.625,
      "learning_rate": 4.157840561423936e-05,
      "loss": 1.2726,
      "num_input_tokens_seen": 5973960,
      "step": 10770
    },
    {
      "epoch": 189.0353982300885,
      "grad_norm": 8.5,
      "learning_rate": 4.1571055967038416e-05,
      "loss": 0.8876,
      "num_input_tokens_seen": 5976440,
      "step": 10775
    },
    {
      "epoch": 189.12389380530973,
      "grad_norm": 8.1875,
      "learning_rate": 4.156370376437241e-05,
      "loss": 1.0378,
      "num_input_tokens_seen": 5979400,
      "step": 10780
    },
    {
      "epoch": 189.21238938053096,
      "grad_norm": 7.09375,
      "learning_rate": 4.155634900737513e-05,
      "loss": 1.0055,
      "num_input_tokens_seen": 5982008,
      "step": 10785
    },
    {
      "epoch": 189.30088495575222,
      "grad_norm": 9.0,
      "learning_rate": 4.1548991697180764e-05,
      "loss": 1.0532,
      "num_input_tokens_seen": 5984648,
      "step": 10790
    },
    {
      "epoch": 189.38938053097345,
      "grad_norm": 7.875,
      "learning_rate": 4.1541631834923914e-05,
      "loss": 1.0039,
      "num_input_tokens_seen": 5987496,
      "step": 10795
    },
    {
      "epoch": 189.47787610619469,
      "grad_norm": 8.5,
      "learning_rate": 4.153426942173956e-05,
      "loss": 1.0755,
      "num_input_tokens_seen": 5990296,
      "step": 10800
    },
    {
      "epoch": 189.47787610619469,
      "eval_loss": 1.0415838956832886,
      "eval_runtime": 0.933,
      "eval_samples_per_second": 26.796,
      "eval_steps_per_second": 13.934,
      "num_input_tokens_seen": 5990296,
      "step": 10800
    },
    {
      "epoch": 189.56637168141592,
      "grad_norm": 7.0,
      "learning_rate": 4.152690445876308e-05,
      "loss": 0.9795,
      "num_input_tokens_seen": 5992984,
      "step": 10805
    },
    {
      "epoch": 189.65486725663717,
      "grad_norm": 6.4375,
      "learning_rate": 4.1519536947130245e-05,
      "loss": 1.0257,
      "num_input_tokens_seen": 5995720,
      "step": 10810
    },
    {
      "epoch": 189.7433628318584,
      "grad_norm": 8.875,
      "learning_rate": 4.151216688797722e-05,
      "loss": 1.2791,
      "num_input_tokens_seen": 5998296,
      "step": 10815
    },
    {
      "epoch": 189.83185840707964,
      "grad_norm": 9.625,
      "learning_rate": 4.150479428244054e-05,
      "loss": 0.9938,
      "num_input_tokens_seen": 6001560,
      "step": 10820
    },
    {
      "epoch": 189.9203539823009,
      "grad_norm": 11.8125,
      "learning_rate": 4.1497419131657176e-05,
      "loss": 1.2484,
      "num_input_tokens_seen": 6004920,
      "step": 10825
    },
    {
      "epoch": 190.0,
      "grad_norm": 5.28125,
      "learning_rate": 4.149004143676447e-05,
      "loss": 0.8493,
      "num_input_tokens_seen": 6007160,
      "step": 10830
    },
    {
      "epoch": 190.08849557522123,
      "grad_norm": 12.6875,
      "learning_rate": 4.148266119890015e-05,
      "loss": 1.2451,
      "num_input_tokens_seen": 6010424,
      "step": 10835
    },
    {
      "epoch": 190.1769911504425,
      "grad_norm": 7.375,
      "learning_rate": 4.1475278419202324e-05,
      "loss": 0.9479,
      "num_input_tokens_seen": 6013384,
      "step": 10840
    },
    {
      "epoch": 190.26548672566372,
      "grad_norm": 6.90625,
      "learning_rate": 4.146789309880953e-05,
      "loss": 1.0363,
      "num_input_tokens_seen": 6016056,
      "step": 10845
    },
    {
      "epoch": 190.35398230088495,
      "grad_norm": 9.8125,
      "learning_rate": 4.146050523886068e-05,
      "loss": 1.2199,
      "num_input_tokens_seen": 6018760,
      "step": 10850
    },
    {
      "epoch": 190.44247787610618,
      "grad_norm": 8.875,
      "learning_rate": 4.1453114840495055e-05,
      "loss": 0.9382,
      "num_input_tokens_seen": 6021384,
      "step": 10855
    },
    {
      "epoch": 190.53097345132744,
      "grad_norm": 8.0,
      "learning_rate": 4.1445721904852364e-05,
      "loss": 1.1284,
      "num_input_tokens_seen": 6023944,
      "step": 10860
    },
    {
      "epoch": 190.61946902654867,
      "grad_norm": 8.4375,
      "learning_rate": 4.143832643307269e-05,
      "loss": 1.0208,
      "num_input_tokens_seen": 6026568,
      "step": 10865
    },
    {
      "epoch": 190.7079646017699,
      "grad_norm": 8.8125,
      "learning_rate": 4.1430928426296503e-05,
      "loss": 0.7861,
      "num_input_tokens_seen": 6029432,
      "step": 10870
    },
    {
      "epoch": 190.79646017699116,
      "grad_norm": 9.9375,
      "learning_rate": 4.142352788566466e-05,
      "loss": 1.0674,
      "num_input_tokens_seen": 6032456,
      "step": 10875
    },
    {
      "epoch": 190.8849557522124,
      "grad_norm": 8.0,
      "learning_rate": 4.1416124812318424e-05,
      "loss": 1.0601,
      "num_input_tokens_seen": 6035032,
      "step": 10880
    },
    {
      "epoch": 190.97345132743362,
      "grad_norm": 8.9375,
      "learning_rate": 4.1408719207399453e-05,
      "loss": 1.1148,
      "num_input_tokens_seen": 6037768,
      "step": 10885
    },
    {
      "epoch": 191.05309734513276,
      "grad_norm": 8.3125,
      "learning_rate": 4.140131107204978e-05,
      "loss": 1.2175,
      "num_input_tokens_seen": 6039864,
      "step": 10890
    },
    {
      "epoch": 191.141592920354,
      "grad_norm": 8.8125,
      "learning_rate": 4.139390040741182e-05,
      "loss": 1.253,
      "num_input_tokens_seen": 6042808,
      "step": 10895
    },
    {
      "epoch": 191.23008849557522,
      "grad_norm": 8.0,
      "learning_rate": 4.1386487214628396e-05,
      "loss": 1.0118,
      "num_input_tokens_seen": 6045336,
      "step": 10900
    },
    {
      "epoch": 191.31858407079645,
      "grad_norm": 9.5625,
      "learning_rate": 4.137907149484272e-05,
      "loss": 1.0592,
      "num_input_tokens_seen": 6048504,
      "step": 10905
    },
    {
      "epoch": 191.4070796460177,
      "grad_norm": 9.9375,
      "learning_rate": 4.137165324919839e-05,
      "loss": 1.2159,
      "num_input_tokens_seen": 6051048,
      "step": 10910
    },
    {
      "epoch": 191.49557522123894,
      "grad_norm": 8.4375,
      "learning_rate": 4.136423247883939e-05,
      "loss": 0.9772,
      "num_input_tokens_seen": 6054216,
      "step": 10915
    },
    {
      "epoch": 191.58407079646017,
      "grad_norm": 7.34375,
      "learning_rate": 4.135680918491009e-05,
      "loss": 1.1091,
      "num_input_tokens_seen": 6056568,
      "step": 10920
    },
    {
      "epoch": 191.67256637168143,
      "grad_norm": 8.6875,
      "learning_rate": 4.1349383368555265e-05,
      "loss": 0.9261,
      "num_input_tokens_seen": 6059368,
      "step": 10925
    },
    {
      "epoch": 191.76106194690266,
      "grad_norm": 8.9375,
      "learning_rate": 4.1341955030920065e-05,
      "loss": 1.0001,
      "num_input_tokens_seen": 6061960,
      "step": 10930
    },
    {
      "epoch": 191.8495575221239,
      "grad_norm": 6.96875,
      "learning_rate": 4.1334524173150036e-05,
      "loss": 0.9035,
      "num_input_tokens_seen": 6064888,
      "step": 10935
    },
    {
      "epoch": 191.93805309734512,
      "grad_norm": 7.9375,
      "learning_rate": 4.13270907963911e-05,
      "loss": 0.992,
      "num_input_tokens_seen": 6067912,
      "step": 10940
    },
    {
      "epoch": 192.01769911504425,
      "grad_norm": 9.1875,
      "learning_rate": 4.131965490178959e-05,
      "loss": 1.1962,
      "num_input_tokens_seen": 6070248,
      "step": 10945
    },
    {
      "epoch": 192.10619469026548,
      "grad_norm": 9.1875,
      "learning_rate": 4.131221649049222e-05,
      "loss": 1.0781,
      "num_input_tokens_seen": 6072920,
      "step": 10950
    },
    {
      "epoch": 192.1946902654867,
      "grad_norm": 7.9375,
      "learning_rate": 4.130477556364606e-05,
      "loss": 1.0372,
      "num_input_tokens_seen": 6076184,
      "step": 10955
    },
    {
      "epoch": 192.28318584070797,
      "grad_norm": 8.875,
      "learning_rate": 4.129733212239861e-05,
      "loss": 1.1383,
      "num_input_tokens_seen": 6078776,
      "step": 10960
    },
    {
      "epoch": 192.3716814159292,
      "grad_norm": 9.3125,
      "learning_rate": 4.128988616789774e-05,
      "loss": 0.9416,
      "num_input_tokens_seen": 6081560,
      "step": 10965
    },
    {
      "epoch": 192.46017699115043,
      "grad_norm": 8.4375,
      "learning_rate": 4.1282437701291724e-05,
      "loss": 1.0278,
      "num_input_tokens_seen": 6084216,
      "step": 10970
    },
    {
      "epoch": 192.5486725663717,
      "grad_norm": 8.375,
      "learning_rate": 4.1274986723729184e-05,
      "loss": 1.1344,
      "num_input_tokens_seen": 6087096,
      "step": 10975
    },
    {
      "epoch": 192.63716814159292,
      "grad_norm": 7.375,
      "learning_rate": 4.126753323635917e-05,
      "loss": 0.9401,
      "num_input_tokens_seen": 6089672,
      "step": 10980
    },
    {
      "epoch": 192.72566371681415,
      "grad_norm": 8.6875,
      "learning_rate": 4.12600772403311e-05,
      "loss": 1.0394,
      "num_input_tokens_seen": 6092584,
      "step": 10985
    },
    {
      "epoch": 192.81415929203538,
      "grad_norm": 8.875,
      "learning_rate": 4.125261873679479e-05,
      "loss": 1.0665,
      "num_input_tokens_seen": 6095640,
      "step": 10990
    },
    {
      "epoch": 192.90265486725664,
      "grad_norm": 6.90625,
      "learning_rate": 4.124515772690042e-05,
      "loss": 0.9792,
      "num_input_tokens_seen": 6098376,
      "step": 10995
    },
    {
      "epoch": 192.99115044247787,
      "grad_norm": 8.3125,
      "learning_rate": 4.123769421179858e-05,
      "loss": 1.1533,
      "num_input_tokens_seen": 6101128,
      "step": 11000
    },
    {
      "epoch": 192.99115044247787,
      "eval_loss": 1.0258084535598755,
      "eval_runtime": 0.9404,
      "eval_samples_per_second": 26.585,
      "eval_steps_per_second": 13.824,
      "num_input_tokens_seen": 6101128,
      "step": 11000
    },
    {
      "epoch": 193.07079646017698,
      "grad_norm": 7.21875,
      "learning_rate": 4.1230228192640236e-05,
      "loss": 0.9656,
      "num_input_tokens_seen": 6103168,
      "step": 11005
    },
    {
      "epoch": 193.15929203539824,
      "grad_norm": 10.5625,
      "learning_rate": 4.122275967057675e-05,
      "loss": 1.1277,
      "num_input_tokens_seen": 6105920,
      "step": 11010
    },
    {
      "epoch": 193.24778761061947,
      "grad_norm": 6.53125,
      "learning_rate": 4.1215288646759846e-05,
      "loss": 1.0575,
      "num_input_tokens_seen": 6108480,
      "step": 11015
    },
    {
      "epoch": 193.3362831858407,
      "grad_norm": 6.96875,
      "learning_rate": 4.120781512234166e-05,
      "loss": 0.8903,
      "num_input_tokens_seen": 6111424,
      "step": 11020
    },
    {
      "epoch": 193.42477876106196,
      "grad_norm": 6.90625,
      "learning_rate": 4.120033909847471e-05,
      "loss": 0.8834,
      "num_input_tokens_seen": 6114912,
      "step": 11025
    },
    {
      "epoch": 193.5132743362832,
      "grad_norm": 11.6875,
      "learning_rate": 4.119286057631187e-05,
      "loss": 1.1047,
      "num_input_tokens_seen": 6117728,
      "step": 11030
    },
    {
      "epoch": 193.60176991150442,
      "grad_norm": 9.375,
      "learning_rate": 4.118537955700646e-05,
      "loss": 1.2807,
      "num_input_tokens_seen": 6120352,
      "step": 11035
    },
    {
      "epoch": 193.69026548672565,
      "grad_norm": 9.0625,
      "learning_rate": 4.11778960417121e-05,
      "loss": 1.0691,
      "num_input_tokens_seen": 6123600,
      "step": 11040
    },
    {
      "epoch": 193.7787610619469,
      "grad_norm": 7.59375,
      "learning_rate": 4.117041003158288e-05,
      "loss": 0.8072,
      "num_input_tokens_seen": 6126960,
      "step": 11045
    },
    {
      "epoch": 193.86725663716814,
      "grad_norm": 9.875,
      "learning_rate": 4.1162921527773215e-05,
      "loss": 1.271,
      "num_input_tokens_seen": 6129376,
      "step": 11050
    },
    {
      "epoch": 193.95575221238937,
      "grad_norm": 7.5625,
      "learning_rate": 4.115543053143794e-05,
      "loss": 1.0728,
      "num_input_tokens_seen": 6132192,
      "step": 11055
    },
    {
      "epoch": 194.0353982300885,
      "grad_norm": 8.5,
      "learning_rate": 4.114793704373226e-05,
      "loss": 0.9246,
      "num_input_tokens_seen": 6134472,
      "step": 11060
    },
    {
      "epoch": 194.12389380530973,
      "grad_norm": 11.1875,
      "learning_rate": 4.114044106581175e-05,
      "loss": 1.1459,
      "num_input_tokens_seen": 6137384,
      "step": 11065
    },
    {
      "epoch": 194.21238938053096,
      "grad_norm": 8.125,
      "learning_rate": 4.11329425988324e-05,
      "loss": 1.0213,
      "num_input_tokens_seen": 6140200,
      "step": 11070
    },
    {
      "epoch": 194.30088495575222,
      "grad_norm": 7.34375,
      "learning_rate": 4.112544164395056e-05,
      "loss": 1.0367,
      "num_input_tokens_seen": 6142904,
      "step": 11075
    },
    {
      "epoch": 194.38938053097345,
      "grad_norm": 6.75,
      "learning_rate": 4.111793820232297e-05,
      "loss": 1.066,
      "num_input_tokens_seen": 6145368,
      "step": 11080
    },
    {
      "epoch": 194.47787610619469,
      "grad_norm": 10.0,
      "learning_rate": 4.1110432275106767e-05,
      "loss": 0.9627,
      "num_input_tokens_seen": 6147992,
      "step": 11085
    },
    {
      "epoch": 194.56637168141592,
      "grad_norm": 8.875,
      "learning_rate": 4.110292386345944e-05,
      "loss": 1.1544,
      "num_input_tokens_seen": 6150680,
      "step": 11090
    },
    {
      "epoch": 194.65486725663717,
      "grad_norm": 8.4375,
      "learning_rate": 4.109541296853891e-05,
      "loss": 1.0403,
      "num_input_tokens_seen": 6153304,
      "step": 11095
    },
    {
      "epoch": 194.7433628318584,
      "grad_norm": 8.6875,
      "learning_rate": 4.108789959150341e-05,
      "loss": 1.1649,
      "num_input_tokens_seen": 6156360,
      "step": 11100
    },
    {
      "epoch": 194.83185840707964,
      "grad_norm": 7.90625,
      "learning_rate": 4.108038373351163e-05,
      "loss": 1.1144,
      "num_input_tokens_seen": 6159624,
      "step": 11105
    },
    {
      "epoch": 194.9203539823009,
      "grad_norm": 9.0,
      "learning_rate": 4.10728653957226e-05,
      "loss": 1.0442,
      "num_input_tokens_seen": 6162392,
      "step": 11110
    },
    {
      "epoch": 195.0,
      "grad_norm": 4.0625,
      "learning_rate": 4.106534457929575e-05,
      "loss": 0.8223,
      "num_input_tokens_seen": 6165064,
      "step": 11115
    },
    {
      "epoch": 195.08849557522123,
      "grad_norm": 7.09375,
      "learning_rate": 4.105782128539086e-05,
      "loss": 0.7634,
      "num_input_tokens_seen": 6168008,
      "step": 11120
    },
    {
      "epoch": 195.1769911504425,
      "grad_norm": 8.0,
      "learning_rate": 4.1050295515168144e-05,
      "loss": 1.1357,
      "num_input_tokens_seen": 6170776,
      "step": 11125
    },
    {
      "epoch": 195.26548672566372,
      "grad_norm": 7.4375,
      "learning_rate": 4.1042767269788155e-05,
      "loss": 1.1259,
      "num_input_tokens_seen": 6173400,
      "step": 11130
    },
    {
      "epoch": 195.35398230088495,
      "grad_norm": 7.65625,
      "learning_rate": 4.103523655041185e-05,
      "loss": 1.0117,
      "num_input_tokens_seen": 6176088,
      "step": 11135
    },
    {
      "epoch": 195.44247787610618,
      "grad_norm": 9.375,
      "learning_rate": 4.102770335820055e-05,
      "loss": 1.1872,
      "num_input_tokens_seen": 6179176,
      "step": 11140
    },
    {
      "epoch": 195.53097345132744,
      "grad_norm": 8.9375,
      "learning_rate": 4.1020167694315984e-05,
      "loss": 0.9554,
      "num_input_tokens_seen": 6181880,
      "step": 11145
    },
    {
      "epoch": 195.61946902654867,
      "grad_norm": 8.375,
      "learning_rate": 4.101262955992023e-05,
      "loss": 1.0522,
      "num_input_tokens_seen": 6184360,
      "step": 11150
    },
    {
      "epoch": 195.7079646017699,
      "grad_norm": 8.5,
      "learning_rate": 4.100508895617578e-05,
      "loss": 1.2197,
      "num_input_tokens_seen": 6187240,
      "step": 11155
    },
    {
      "epoch": 195.79646017699116,
      "grad_norm": 9.8125,
      "learning_rate": 4.099754588424547e-05,
      "loss": 1.0895,
      "num_input_tokens_seen": 6189944,
      "step": 11160
    },
    {
      "epoch": 195.8849557522124,
      "grad_norm": 7.3125,
      "learning_rate": 4.0990000345292546e-05,
      "loss": 0.9983,
      "num_input_tokens_seen": 6192904,
      "step": 11165
    },
    {
      "epoch": 195.97345132743362,
      "grad_norm": 7.625,
      "learning_rate": 4.098245234048064e-05,
      "loss": 1.0517,
      "num_input_tokens_seen": 6195816,
      "step": 11170
    },
    {
      "epoch": 196.05309734513276,
      "grad_norm": 7.84375,
      "learning_rate": 4.0974901870973726e-05,
      "loss": 1.1185,
      "num_input_tokens_seen": 6198232,
      "step": 11175
    },
    {
      "epoch": 196.141592920354,
      "grad_norm": 7.8125,
      "learning_rate": 4.096734893793619e-05,
      "loss": 1.0712,
      "num_input_tokens_seen": 6201432,
      "step": 11180
    },
    {
      "epoch": 196.23008849557522,
      "grad_norm": 8.5,
      "learning_rate": 4.095979354253279e-05,
      "loss": 1.0053,
      "num_input_tokens_seen": 6204296,
      "step": 11185
    },
    {
      "epoch": 196.31858407079645,
      "grad_norm": 6.03125,
      "learning_rate": 4.0952235685928656e-05,
      "loss": 0.8962,
      "num_input_tokens_seen": 6206504,
      "step": 11190
    },
    {
      "epoch": 196.4070796460177,
      "grad_norm": 9.4375,
      "learning_rate": 4.094467536928932e-05,
      "loss": 0.9007,
      "num_input_tokens_seen": 6209320,
      "step": 11195
    },
    {
      "epoch": 196.49557522123894,
      "grad_norm": 11.6875,
      "learning_rate": 4.093711259378067e-05,
      "loss": 1.1898,
      "num_input_tokens_seen": 6212008,
      "step": 11200
    },
    {
      "epoch": 196.49557522123894,
      "eval_loss": 1.0284020900726318,
      "eval_runtime": 0.9356,
      "eval_samples_per_second": 26.72,
      "eval_steps_per_second": 13.894,
      "num_input_tokens_seen": 6212008,
      "step": 11200
    },
    {
      "epoch": 196.58407079646017,
      "grad_norm": 6.125,
      "learning_rate": 4.092954736056897e-05,
      "loss": 0.9682,
      "num_input_tokens_seen": 6214888,
      "step": 11205
    },
    {
      "epoch": 196.67256637168143,
      "grad_norm": 8.75,
      "learning_rate": 4.09219796708209e-05,
      "loss": 1.0563,
      "num_input_tokens_seen": 6217560,
      "step": 11210
    },
    {
      "epoch": 196.76106194690266,
      "grad_norm": 8.0625,
      "learning_rate": 4.0914409525703464e-05,
      "loss": 1.178,
      "num_input_tokens_seen": 6220360,
      "step": 11215
    },
    {
      "epoch": 196.8495575221239,
      "grad_norm": 9.3125,
      "learning_rate": 4.090683692638408e-05,
      "loss": 1.2419,
      "num_input_tokens_seen": 6222888,
      "step": 11220
    },
    {
      "epoch": 196.93805309734512,
      "grad_norm": 10.1875,
      "learning_rate": 4.089926187403056e-05,
      "loss": 0.9299,
      "num_input_tokens_seen": 6226040,
      "step": 11225
    },
    {
      "epoch": 197.01769911504425,
      "grad_norm": 8.25,
      "learning_rate": 4.0891684369811044e-05,
      "loss": 1.1237,
      "num_input_tokens_seen": 6228192,
      "step": 11230
    },
    {
      "epoch": 197.10619469026548,
      "grad_norm": 12.1875,
      "learning_rate": 4.0884104414894107e-05,
      "loss": 1.0736,
      "num_input_tokens_seen": 6231360,
      "step": 11235
    },
    {
      "epoch": 197.1946902654867,
      "grad_norm": 7.96875,
      "learning_rate": 4.087652201044864e-05,
      "loss": 0.8698,
      "num_input_tokens_seen": 6234304,
      "step": 11240
    },
    {
      "epoch": 197.28318584070797,
      "grad_norm": 7.78125,
      "learning_rate": 4.086893715764397e-05,
      "loss": 0.9732,
      "num_input_tokens_seen": 6236800,
      "step": 11245
    },
    {
      "epoch": 197.3716814159292,
      "grad_norm": 9.875,
      "learning_rate": 4.086134985764977e-05,
      "loss": 1.179,
      "num_input_tokens_seen": 6239504,
      "step": 11250
    },
    {
      "epoch": 197.46017699115043,
      "grad_norm": 8.125,
      "learning_rate": 4.0853760111636085e-05,
      "loss": 1.1205,
      "num_input_tokens_seen": 6242496,
      "step": 11255
    },
    {
      "epoch": 197.5486725663717,
      "grad_norm": 8.125,
      "learning_rate": 4.084616792077337e-05,
      "loss": 1.064,
      "num_input_tokens_seen": 6245072,
      "step": 11260
    },
    {
      "epoch": 197.63716814159292,
      "grad_norm": 10.0625,
      "learning_rate": 4.083857328623243e-05,
      "loss": 0.9887,
      "num_input_tokens_seen": 6247616,
      "step": 11265
    },
    {
      "epoch": 197.72566371681415,
      "grad_norm": 9.4375,
      "learning_rate": 4.083097620918444e-05,
      "loss": 1.0479,
      "num_input_tokens_seen": 6250672,
      "step": 11270
    },
    {
      "epoch": 197.81415929203538,
      "grad_norm": 10.0625,
      "learning_rate": 4.082337669080097e-05,
      "loss": 1.1392,
      "num_input_tokens_seen": 6253200,
      "step": 11275
    },
    {
      "epoch": 197.90265486725664,
      "grad_norm": 8.4375,
      "learning_rate": 4.081577473225398e-05,
      "loss": 1.1001,
      "num_input_tokens_seen": 6255968,
      "step": 11280
    },
    {
      "epoch": 197.99115044247787,
      "grad_norm": 8.0,
      "learning_rate": 4.080817033471577e-05,
      "loss": 0.9846,
      "num_input_tokens_seen": 6258784,
      "step": 11285
    },
    {
      "epoch": 198.07079646017698,
      "grad_norm": 7.0,
      "learning_rate": 4.080056349935903e-05,
      "loss": 1.1018,
      "num_input_tokens_seen": 6261016,
      "step": 11290
    },
    {
      "epoch": 198.15929203539824,
      "grad_norm": 8.875,
      "learning_rate": 4.079295422735684e-05,
      "loss": 0.8106,
      "num_input_tokens_seen": 6263400,
      "step": 11295
    },
    {
      "epoch": 198.24778761061947,
      "grad_norm": 10.25,
      "learning_rate": 4.078534251988264e-05,
      "loss": 1.2239,
      "num_input_tokens_seen": 6266344,
      "step": 11300
    },
    {
      "epoch": 198.3362831858407,
      "grad_norm": 7.96875,
      "learning_rate": 4.077772837811025e-05,
      "loss": 1.0717,
      "num_input_tokens_seen": 6269256,
      "step": 11305
    },
    {
      "epoch": 198.42477876106196,
      "grad_norm": 5.96875,
      "learning_rate": 4.0770111803213874e-05,
      "loss": 0.9906,
      "num_input_tokens_seen": 6271912,
      "step": 11310
    },
    {
      "epoch": 198.5132743362832,
      "grad_norm": 6.9375,
      "learning_rate": 4.076249279636807e-05,
      "loss": 0.9635,
      "num_input_tokens_seen": 6275240,
      "step": 11315
    },
    {
      "epoch": 198.60176991150442,
      "grad_norm": 8.625,
      "learning_rate": 4.075487135874781e-05,
      "loss": 1.0939,
      "num_input_tokens_seen": 6277720,
      "step": 11320
    },
    {
      "epoch": 198.69026548672565,
      "grad_norm": 9.5,
      "learning_rate": 4.074724749152837e-05,
      "loss": 1.0415,
      "num_input_tokens_seen": 6280792,
      "step": 11325
    },
    {
      "epoch": 198.7787610619469,
      "grad_norm": 8.25,
      "learning_rate": 4.07396211958855e-05,
      "loss": 1.1244,
      "num_input_tokens_seen": 6283560,
      "step": 11330
    },
    {
      "epoch": 198.86725663716814,
      "grad_norm": 14.0625,
      "learning_rate": 4.073199247299523e-05,
      "loss": 1.1867,
      "num_input_tokens_seen": 6286456,
      "step": 11335
    },
    {
      "epoch": 198.95575221238937,
      "grad_norm": 8.0625,
      "learning_rate": 4.072436132403403e-05,
      "loss": 1.0569,
      "num_input_tokens_seen": 6288968,
      "step": 11340
    },
    {
      "epoch": 199.0353982300885,
      "grad_norm": 7.1875,
      "learning_rate": 4.0716727750178704e-05,
      "loss": 0.9656,
      "num_input_tokens_seen": 6291200,
      "step": 11345
    },
    {
      "epoch": 199.12389380530973,
      "grad_norm": 6.28125,
      "learning_rate": 4.0709091752606455e-05,
      "loss": 1.1146,
      "num_input_tokens_seen": 6293712,
      "step": 11350
    },
    {
      "epoch": 199.21238938053096,
      "grad_norm": 6.96875,
      "learning_rate": 4.070145333249484e-05,
      "loss": 1.187,
      "num_input_tokens_seen": 6296976,
      "step": 11355
    },
    {
      "epoch": 199.30088495575222,
      "grad_norm": 7.09375,
      "learning_rate": 4.069381249102181e-05,
      "loss": 0.9839,
      "num_input_tokens_seen": 6300048,
      "step": 11360
    },
    {
      "epoch": 199.38938053097345,
      "grad_norm": 12.3125,
      "learning_rate": 4.0686169229365665e-05,
      "loss": 1.0348,
      "num_input_tokens_seen": 6302768,
      "step": 11365
    },
    {
      "epoch": 199.47787610619469,
      "grad_norm": 8.9375,
      "learning_rate": 4.067852354870511e-05,
      "loss": 1.2244,
      "num_input_tokens_seen": 6305360,
      "step": 11370
    },
    {
      "epoch": 199.56637168141592,
      "grad_norm": 8.4375,
      "learning_rate": 4.067087545021919e-05,
      "loss": 1.0371,
      "num_input_tokens_seen": 6308208,
      "step": 11375
    },
    {
      "epoch": 199.65486725663717,
      "grad_norm": 8.375,
      "learning_rate": 4.066322493508734e-05,
      "loss": 1.0496,
      "num_input_tokens_seen": 6310992,
      "step": 11380
    },
    {
      "epoch": 199.7433628318584,
      "grad_norm": 9.3125,
      "learning_rate": 4.065557200448937e-05,
      "loss": 1.0923,
      "num_input_tokens_seen": 6313472,
      "step": 11385
    },
    {
      "epoch": 199.83185840707964,
      "grad_norm": 8.5,
      "learning_rate": 4.064791665960546e-05,
      "loss": 0.9819,
      "num_input_tokens_seen": 6316224,
      "step": 11390
    },
    {
      "epoch": 199.9203539823009,
      "grad_norm": 7.78125,
      "learning_rate": 4.064025890161615e-05,
      "loss": 0.8499,
      "num_input_tokens_seen": 6319344,
      "step": 11395
    },
    {
      "epoch": 200.0,
      "grad_norm": 9.0625,
      "learning_rate": 4.0632598731702373e-05,
      "loss": 1.0763,
      "num_input_tokens_seen": 6321568,
      "step": 11400
    },
    {
      "epoch": 200.0,
      "eval_loss": 1.0304739475250244,
      "eval_runtime": 0.9335,
      "eval_samples_per_second": 26.781,
      "eval_steps_per_second": 13.926,
      "num_input_tokens_seen": 6321568,
      "step": 11400
    },
    {
      "epoch": 200.08849557522123,
      "grad_norm": 8.0625,
      "learning_rate": 4.0624936151045426e-05,
      "loss": 1.3375,
      "num_input_tokens_seen": 6324320,
      "step": 11405
    },
    {
      "epoch": 200.1769911504425,
      "grad_norm": 11.6875,
      "learning_rate": 4.061727116082696e-05,
      "loss": 0.8608,
      "num_input_tokens_seen": 6327216,
      "step": 11410
    },
    {
      "epoch": 200.26548672566372,
      "grad_norm": 8.5,
      "learning_rate": 4.060960376222903e-05,
      "loss": 1.1217,
      "num_input_tokens_seen": 6330208,
      "step": 11415
    },
    {
      "epoch": 200.35398230088495,
      "grad_norm": 6.03125,
      "learning_rate": 4.0601933956434034e-05,
      "loss": 0.8895,
      "num_input_tokens_seen": 6332720,
      "step": 11420
    },
    {
      "epoch": 200.44247787610618,
      "grad_norm": 7.125,
      "learning_rate": 4.059426174462476e-05,
      "loss": 0.8561,
      "num_input_tokens_seen": 6335504,
      "step": 11425
    },
    {
      "epoch": 200.53097345132744,
      "grad_norm": 9.8125,
      "learning_rate": 4.058658712798435e-05,
      "loss": 1.0412,
      "num_input_tokens_seen": 6338528,
      "step": 11430
    },
    {
      "epoch": 200.61946902654867,
      "grad_norm": 8.5,
      "learning_rate": 4.0578910107696336e-05,
      "loss": 1.1327,
      "num_input_tokens_seen": 6341008,
      "step": 11435
    },
    {
      "epoch": 200.7079646017699,
      "grad_norm": 7.5,
      "learning_rate": 4.05712306849446e-05,
      "loss": 0.9974,
      "num_input_tokens_seen": 6343808,
      "step": 11440
    },
    {
      "epoch": 200.79646017699116,
      "grad_norm": 7.625,
      "learning_rate": 4.0563548860913415e-05,
      "loss": 1.0679,
      "num_input_tokens_seen": 6346416,
      "step": 11445
    },
    {
      "epoch": 200.8849557522124,
      "grad_norm": 9.1875,
      "learning_rate": 4.0555864636787414e-05,
      "loss": 1.0413,
      "num_input_tokens_seen": 6349664,
      "step": 11450
    },
    {
      "epoch": 200.97345132743362,
      "grad_norm": 10.875,
      "learning_rate": 4.054817801375159e-05,
      "loss": 1.1673,
      "num_input_tokens_seen": 6352640,
      "step": 11455
    },
    {
      "epoch": 201.05309734513276,
      "grad_norm": 8.4375,
      "learning_rate": 4.054048899299134e-05,
      "loss": 1.1518,
      "num_input_tokens_seen": 6354944,
      "step": 11460
    },
    {
      "epoch": 201.141592920354,
      "grad_norm": 7.46875,
      "learning_rate": 4.0532797575692385e-05,
      "loss": 1.096,
      "num_input_tokens_seen": 6357376,
      "step": 11465
    },
    {
      "epoch": 201.23008849557522,
      "grad_norm": 9.3125,
      "learning_rate": 4.052510376304085e-05,
      "loss": 1.133,
      "num_input_tokens_seen": 6360608,
      "step": 11470
    },
    {
      "epoch": 201.31858407079645,
      "grad_norm": 7.0625,
      "learning_rate": 4.051740755622321e-05,
      "loss": 1.1852,
      "num_input_tokens_seen": 6362928,
      "step": 11475
    },
    {
      "epoch": 201.4070796460177,
      "grad_norm": 7.34375,
      "learning_rate": 4.050970895642632e-05,
      "loss": 0.9552,
      "num_input_tokens_seen": 6365504,
      "step": 11480
    },
    {
      "epoch": 201.49557522123894,
      "grad_norm": 8.0625,
      "learning_rate": 4.050200796483741e-05,
      "loss": 1.0891,
      "num_input_tokens_seen": 6368480,
      "step": 11485
    },
    {
      "epoch": 201.58407079646017,
      "grad_norm": 8.0625,
      "learning_rate": 4.049430458264405e-05,
      "loss": 1.0613,
      "num_input_tokens_seen": 6371648,
      "step": 11490
    },
    {
      "epoch": 201.67256637168143,
      "grad_norm": 6.59375,
      "learning_rate": 4.048659881103422e-05,
      "loss": 0.8822,
      "num_input_tokens_seen": 6374176,
      "step": 11495
    },
    {
      "epoch": 201.76106194690266,
      "grad_norm": 6.125,
      "learning_rate": 4.0478890651196235e-05,
      "loss": 0.9433,
      "num_input_tokens_seen": 6377264,
      "step": 11500
    },
    {
      "epoch": 201.8495575221239,
      "grad_norm": 7.875,
      "learning_rate": 4.047118010431879e-05,
      "loss": 0.9188,
      "num_input_tokens_seen": 6380128,
      "step": 11505
    },
    {
      "epoch": 201.93805309734512,
      "grad_norm": 7.59375,
      "learning_rate": 4.046346717159094e-05,
      "loss": 0.9808,
      "num_input_tokens_seen": 6382672,
      "step": 11510
    },
    {
      "epoch": 202.01769911504425,
      "grad_norm": 8.125,
      "learning_rate": 4.045575185420214e-05,
      "loss": 1.4308,
      "num_input_tokens_seen": 6385216,
      "step": 11515
    },
    {
      "epoch": 202.10619469026548,
      "grad_norm": 8.75,
      "learning_rate": 4.0448034153342165e-05,
      "loss": 1.1015,
      "num_input_tokens_seen": 6387904,
      "step": 11520
    },
    {
      "epoch": 202.1946902654867,
      "grad_norm": 9.875,
      "learning_rate": 4.0440314070201194e-05,
      "loss": 1.0439,
      "num_input_tokens_seen": 6390528,
      "step": 11525
    },
    {
      "epoch": 202.28318584070797,
      "grad_norm": 7.53125,
      "learning_rate": 4.043259160596976e-05,
      "loss": 1.0737,
      "num_input_tokens_seen": 6393152,
      "step": 11530
    },
    {
      "epoch": 202.3716814159292,
      "grad_norm": 9.5625,
      "learning_rate": 4.0424866761838767e-05,
      "loss": 0.9773,
      "num_input_tokens_seen": 6396064,
      "step": 11535
    },
    {
      "epoch": 202.46017699115043,
      "grad_norm": 11.125,
      "learning_rate": 4.041713953899948e-05,
      "loss": 1.049,
      "num_input_tokens_seen": 6399152,
      "step": 11540
    },
    {
      "epoch": 202.5486725663717,
      "grad_norm": 7.8125,
      "learning_rate": 4.0409409938643515e-05,
      "loss": 0.9919,
      "num_input_tokens_seen": 6402032,
      "step": 11545
    },
    {
      "epoch": 202.63716814159292,
      "grad_norm": 10.3125,
      "learning_rate": 4.0401677961962904e-05,
      "loss": 1.2039,
      "num_input_tokens_seen": 6404784,
      "step": 11550
    },
    {
      "epoch": 202.72566371681415,
      "grad_norm": 9.375,
      "learning_rate": 4.039394361015001e-05,
      "loss": 1.1644,
      "num_input_tokens_seen": 6407232,
      "step": 11555
    },
    {
      "epoch": 202.81415929203538,
      "grad_norm": 8.625,
      "learning_rate": 4.038620688439755e-05,
      "loss": 0.9366,
      "num_input_tokens_seen": 6410016,
      "step": 11560
    },
    {
      "epoch": 202.90265486725664,
      "grad_norm": 7.71875,
      "learning_rate": 4.037846778589862e-05,
      "loss": 0.8388,
      "num_input_tokens_seen": 6412608,
      "step": 11565
    },
    {
      "epoch": 202.99115044247787,
      "grad_norm": 8.3125,
      "learning_rate": 4.0370726315846715e-05,
      "loss": 1.1739,
      "num_input_tokens_seen": 6415376,
      "step": 11570
    },
    {
      "epoch": 203.07079646017698,
      "grad_norm": 8.5,
      "learning_rate": 4.036298247543565e-05,
      "loss": 1.1792,
      "num_input_tokens_seen": 6417600,
      "step": 11575
    },
    {
      "epoch": 203.15929203539824,
      "grad_norm": 6.53125,
      "learning_rate": 4.035523626585962e-05,
      "loss": 0.9125,
      "num_input_tokens_seen": 6420944,
      "step": 11580
    },
    {
      "epoch": 203.24778761061947,
      "grad_norm": 7.6875,
      "learning_rate": 4.0347487688313194e-05,
      "loss": 0.9942,
      "num_input_tokens_seen": 6423872,
      "step": 11585
    },
    {
      "epoch": 203.3362831858407,
      "grad_norm": 11.5625,
      "learning_rate": 4.0339736743991296e-05,
      "loss": 1.0361,
      "num_input_tokens_seen": 6426880,
      "step": 11590
    },
    {
      "epoch": 203.42477876106196,
      "grad_norm": 8.0625,
      "learning_rate": 4.0331983434089227e-05,
      "loss": 0.8969,
      "num_input_tokens_seen": 6429632,
      "step": 11595
    },
    {
      "epoch": 203.5132743362832,
      "grad_norm": 9.1875,
      "learning_rate": 4.032422775980264e-05,
      "loss": 1.0221,
      "num_input_tokens_seen": 6432384,
      "step": 11600
    },
    {
      "epoch": 203.5132743362832,
      "eval_loss": 1.0272767543792725,
      "eval_runtime": 0.9403,
      "eval_samples_per_second": 26.587,
      "eval_steps_per_second": 13.825,
      "num_input_tokens_seen": 6432384,
      "step": 11600
    },
    {
      "epoch": 203.60176991150442,
      "grad_norm": 11.625,
      "learning_rate": 4.031646972232754e-05,
      "loss": 1.1056,
      "num_input_tokens_seen": 6434864,
      "step": 11605
    },
    {
      "epoch": 203.69026548672565,
      "grad_norm": 7.96875,
      "learning_rate": 4.0308709322860344e-05,
      "loss": 0.9706,
      "num_input_tokens_seen": 6438080,
      "step": 11610
    },
    {
      "epoch": 203.7787610619469,
      "grad_norm": 7.53125,
      "learning_rate": 4.0300946562597784e-05,
      "loss": 1.2112,
      "num_input_tokens_seen": 6440720,
      "step": 11615
    },
    {
      "epoch": 203.86725663716814,
      "grad_norm": 8.625,
      "learning_rate": 4.029318144273698e-05,
      "loss": 1.1802,
      "num_input_tokens_seen": 6443984,
      "step": 11620
    },
    {
      "epoch": 203.95575221238937,
      "grad_norm": 9.8125,
      "learning_rate": 4.0285413964475415e-05,
      "loss": 1.2716,
      "num_input_tokens_seen": 6446192,
      "step": 11625
    },
    {
      "epoch": 204.0353982300885,
      "grad_norm": 7.0,
      "learning_rate": 4.0277644129010927e-05,
      "loss": 0.8918,
      "num_input_tokens_seen": 6448224,
      "step": 11630
    },
    {
      "epoch": 204.12389380530973,
      "grad_norm": 11.125,
      "learning_rate": 4.0269871937541724e-05,
      "loss": 1.2366,
      "num_input_tokens_seen": 6450800,
      "step": 11635
    },
    {
      "epoch": 204.21238938053096,
      "grad_norm": 8.625,
      "learning_rate": 4.026209739126637e-05,
      "loss": 1.0921,
      "num_input_tokens_seen": 6453456,
      "step": 11640
    },
    {
      "epoch": 204.30088495575222,
      "grad_norm": 7.65625,
      "learning_rate": 4.025432049138381e-05,
      "loss": 1.0474,
      "num_input_tokens_seen": 6456704,
      "step": 11645
    },
    {
      "epoch": 204.38938053097345,
      "grad_norm": 8.1875,
      "learning_rate": 4.0246541239093325e-05,
      "loss": 0.9766,
      "num_input_tokens_seen": 6459296,
      "step": 11650
    },
    {
      "epoch": 204.47787610619469,
      "grad_norm": 5.71875,
      "learning_rate": 4.023875963559459e-05,
      "loss": 0.9434,
      "num_input_tokens_seen": 6462048,
      "step": 11655
    },
    {
      "epoch": 204.56637168141592,
      "grad_norm": 10.875,
      "learning_rate": 4.023097568208761e-05,
      "loss": 0.9842,
      "num_input_tokens_seen": 6464992,
      "step": 11660
    },
    {
      "epoch": 204.65486725663717,
      "grad_norm": 7.84375,
      "learning_rate": 4.022318937977277e-05,
      "loss": 1.0573,
      "num_input_tokens_seen": 6467568,
      "step": 11665
    },
    {
      "epoch": 204.7433628318584,
      "grad_norm": 8.75,
      "learning_rate": 4.021540072985084e-05,
      "loss": 1.0442,
      "num_input_tokens_seen": 6470384,
      "step": 11670
    },
    {
      "epoch": 204.83185840707964,
      "grad_norm": 7.3125,
      "learning_rate": 4.020760973352289e-05,
      "loss": 0.9771,
      "num_input_tokens_seen": 6473664,
      "step": 11675
    },
    {
      "epoch": 204.9203539823009,
      "grad_norm": 8.6875,
      "learning_rate": 4.019981639199042e-05,
      "loss": 1.0945,
      "num_input_tokens_seen": 6476560,
      "step": 11680
    },
    {
      "epoch": 205.0,
      "grad_norm": 9.375,
      "learning_rate": 4.0192020706455245e-05,
      "loss": 1.1852,
      "num_input_tokens_seen": 6478616,
      "step": 11685
    },
    {
      "epoch": 205.08849557522123,
      "grad_norm": 13.5,
      "learning_rate": 4.018422267811956e-05,
      "loss": 1.2822,
      "num_input_tokens_seen": 6481224,
      "step": 11690
    },
    {
      "epoch": 205.1769911504425,
      "grad_norm": 8.1875,
      "learning_rate": 4.017642230818592e-05,
      "loss": 0.9354,
      "num_input_tokens_seen": 6484632,
      "step": 11695
    },
    {
      "epoch": 205.26548672566372,
      "grad_norm": 8.5,
      "learning_rate": 4.0168619597857246e-05,
      "loss": 1.2651,
      "num_input_tokens_seen": 6486840,
      "step": 11700
    },
    {
      "epoch": 205.35398230088495,
      "grad_norm": 7.8125,
      "learning_rate": 4.016081454833681e-05,
      "loss": 1.094,
      "num_input_tokens_seen": 6489640,
      "step": 11705
    },
    {
      "epoch": 205.44247787610618,
      "grad_norm": 8.75,
      "learning_rate": 4.0153007160828245e-05,
      "loss": 1.0385,
      "num_input_tokens_seen": 6492536,
      "step": 11710
    },
    {
      "epoch": 205.53097345132744,
      "grad_norm": 8.625,
      "learning_rate": 4.0145197436535555e-05,
      "loss": 0.9926,
      "num_input_tokens_seen": 6495256,
      "step": 11715
    },
    {
      "epoch": 205.61946902654867,
      "grad_norm": 12.3125,
      "learning_rate": 4.0137385376663095e-05,
      "loss": 1.1624,
      "num_input_tokens_seen": 6498056,
      "step": 11720
    },
    {
      "epoch": 205.7079646017699,
      "grad_norm": 8.125,
      "learning_rate": 4.012957098241558e-05,
      "loss": 0.8719,
      "num_input_tokens_seen": 6501448,
      "step": 11725
    },
    {
      "epoch": 205.79646017699116,
      "grad_norm": 5.78125,
      "learning_rate": 4.0121754254998076e-05,
      "loss": 0.8989,
      "num_input_tokens_seen": 6504184,
      "step": 11730
    },
    {
      "epoch": 205.8849557522124,
      "grad_norm": 9.9375,
      "learning_rate": 4.011393519561606e-05,
      "loss": 1.175,
      "num_input_tokens_seen": 6506728,
      "step": 11735
    },
    {
      "epoch": 205.97345132743362,
      "grad_norm": 7.4375,
      "learning_rate": 4.010611380547529e-05,
      "loss": 0.9367,
      "num_input_tokens_seen": 6509496,
      "step": 11740
    },
    {
      "epoch": 206.05309734513276,
      "grad_norm": 6.78125,
      "learning_rate": 4.009829008578192e-05,
      "loss": 1.0743,
      "num_input_tokens_seen": 6511680,
      "step": 11745
    },
    {
      "epoch": 206.141592920354,
      "grad_norm": 7.75,
      "learning_rate": 4.00904640377425e-05,
      "loss": 0.7812,
      "num_input_tokens_seen": 6514336,
      "step": 11750
    },
    {
      "epoch": 206.23008849557522,
      "grad_norm": 9.6875,
      "learning_rate": 4.0082635662563886e-05,
      "loss": 1.3669,
      "num_input_tokens_seen": 6517264,
      "step": 11755
    },
    {
      "epoch": 206.31858407079645,
      "grad_norm": 9.9375,
      "learning_rate": 4.007480496145331e-05,
      "loss": 1.0151,
      "num_input_tokens_seen": 6520176,
      "step": 11760
    },
    {
      "epoch": 206.4070796460177,
      "grad_norm": 10.0625,
      "learning_rate": 4.006697193561837e-05,
      "loss": 1.1821,
      "num_input_tokens_seen": 6522784,
      "step": 11765
    },
    {
      "epoch": 206.49557522123894,
      "grad_norm": 7.8125,
      "learning_rate": 4.005913658626701e-05,
      "loss": 1.1393,
      "num_input_tokens_seen": 6526080,
      "step": 11770
    },
    {
      "epoch": 206.58407079646017,
      "grad_norm": 12.25,
      "learning_rate": 4.005129891460754e-05,
      "loss": 1.2181,
      "num_input_tokens_seen": 6528816,
      "step": 11775
    },
    {
      "epoch": 206.67256637168143,
      "grad_norm": 9.0625,
      "learning_rate": 4.004345892184864e-05,
      "loss": 1.1709,
      "num_input_tokens_seen": 6531728,
      "step": 11780
    },
    {
      "epoch": 206.76106194690266,
      "grad_norm": 7.28125,
      "learning_rate": 4.003561660919932e-05,
      "loss": 0.7266,
      "num_input_tokens_seen": 6534384,
      "step": 11785
    },
    {
      "epoch": 206.8495575221239,
      "grad_norm": 7.96875,
      "learning_rate": 4.002777197786897e-05,
      "loss": 1.0056,
      "num_input_tokens_seen": 6537536,
      "step": 11790
    },
    {
      "epoch": 206.93805309734512,
      "grad_norm": 10.125,
      "learning_rate": 4.0019925029067326e-05,
      "loss": 1.0055,
      "num_input_tokens_seen": 6540048,
      "step": 11795
    },
    {
      "epoch": 207.01769911504425,
      "grad_norm": 7.46875,
      "learning_rate": 4.0012075764004495e-05,
      "loss": 0.8903,
      "num_input_tokens_seen": 6542352,
      "step": 11800
    },
    {
      "epoch": 207.01769911504425,
      "eval_loss": 1.0349314212799072,
      "eval_runtime": 0.93,
      "eval_samples_per_second": 26.882,
      "eval_steps_per_second": 13.979,
      "num_input_tokens_seen": 6542352,
      "step": 11800
    },
    {
      "epoch": 207.10619469026548,
      "grad_norm": 8.0,
      "learning_rate": 4.000422418389094e-05,
      "loss": 1.1224,
      "num_input_tokens_seen": 6545312,
      "step": 11805
    },
    {
      "epoch": 207.1946902654867,
      "grad_norm": 9.125,
      "learning_rate": 3.999637028993744e-05,
      "loss": 0.9819,
      "num_input_tokens_seen": 6548544,
      "step": 11810
    },
    {
      "epoch": 207.28318584070797,
      "grad_norm": 7.15625,
      "learning_rate": 3.99885140833552e-05,
      "loss": 0.982,
      "num_input_tokens_seen": 6551360,
      "step": 11815
    },
    {
      "epoch": 207.3716814159292,
      "grad_norm": 8.0,
      "learning_rate": 3.998065556535572e-05,
      "loss": 1.0802,
      "num_input_tokens_seen": 6553872,
      "step": 11820
    },
    {
      "epoch": 207.46017699115043,
      "grad_norm": 11.875,
      "learning_rate": 3.9972794737150895e-05,
      "loss": 1.0513,
      "num_input_tokens_seen": 6556256,
      "step": 11825
    },
    {
      "epoch": 207.5486725663717,
      "grad_norm": 7.78125,
      "learning_rate": 3.996493159995297e-05,
      "loss": 0.9363,
      "num_input_tokens_seen": 6559344,
      "step": 11830
    },
    {
      "epoch": 207.63716814159292,
      "grad_norm": 7.6875,
      "learning_rate": 3.995706615497453e-05,
      "loss": 1.0782,
      "num_input_tokens_seen": 6562288,
      "step": 11835
    },
    {
      "epoch": 207.72566371681415,
      "grad_norm": 8.6875,
      "learning_rate": 3.994919840342852e-05,
      "loss": 1.0379,
      "num_input_tokens_seen": 6565120,
      "step": 11840
    },
    {
      "epoch": 207.81415929203538,
      "grad_norm": 6.96875,
      "learning_rate": 3.994132834652825e-05,
      "loss": 1.0642,
      "num_input_tokens_seen": 6567568,
      "step": 11845
    },
    {
      "epoch": 207.90265486725664,
      "grad_norm": 9.125,
      "learning_rate": 3.99334559854874e-05,
      "loss": 1.078,
      "num_input_tokens_seen": 6570240,
      "step": 11850
    },
    {
      "epoch": 207.99115044247787,
      "grad_norm": 8.6875,
      "learning_rate": 3.9925581321519955e-05,
      "loss": 1.0683,
      "num_input_tokens_seen": 6573456,
      "step": 11855
    },
    {
      "epoch": 208.07079646017698,
      "grad_norm": 7.0,
      "learning_rate": 3.991770435584031e-05,
      "loss": 1.4463,
      "num_input_tokens_seen": 6575824,
      "step": 11860
    },
    {
      "epoch": 208.15929203539824,
      "grad_norm": 6.75,
      "learning_rate": 3.990982508966319e-05,
      "loss": 1.1083,
      "num_input_tokens_seen": 6578544,
      "step": 11865
    },
    {
      "epoch": 208.24778761061947,
      "grad_norm": 5.84375,
      "learning_rate": 3.990194352420367e-05,
      "loss": 1.0315,
      "num_input_tokens_seen": 6581552,
      "step": 11870
    },
    {
      "epoch": 208.3362831858407,
      "grad_norm": 7.90625,
      "learning_rate": 3.9894059660677184e-05,
      "loss": 1.1545,
      "num_input_tokens_seen": 6584432,
      "step": 11875
    },
    {
      "epoch": 208.42477876106196,
      "grad_norm": 6.53125,
      "learning_rate": 3.9886173500299526e-05,
      "loss": 1.0345,
      "num_input_tokens_seen": 6587088,
      "step": 11880
    },
    {
      "epoch": 208.5132743362832,
      "grad_norm": 10.375,
      "learning_rate": 3.987828504428685e-05,
      "loss": 1.0079,
      "num_input_tokens_seen": 6589984,
      "step": 11885
    },
    {
      "epoch": 208.60176991150442,
      "grad_norm": 8.5625,
      "learning_rate": 3.987039429385565e-05,
      "loss": 0.9237,
      "num_input_tokens_seen": 6592336,
      "step": 11890
    },
    {
      "epoch": 208.69026548672565,
      "grad_norm": 7.5,
      "learning_rate": 3.986250125022277e-05,
      "loss": 1.153,
      "num_input_tokens_seen": 6595248,
      "step": 11895
    },
    {
      "epoch": 208.7787610619469,
      "grad_norm": 7.0,
      "learning_rate": 3.985460591460544e-05,
      "loss": 0.8503,
      "num_input_tokens_seen": 6597744,
      "step": 11900
    },
    {
      "epoch": 208.86725663716814,
      "grad_norm": 10.25,
      "learning_rate": 3.984670828822118e-05,
      "loss": 1.2591,
      "num_input_tokens_seen": 6600752,
      "step": 11905
    },
    {
      "epoch": 208.95575221238937,
      "grad_norm": 7.5,
      "learning_rate": 3.983880837228794e-05,
      "loss": 1.0763,
      "num_input_tokens_seen": 6603840,
      "step": 11910
    },
    {
      "epoch": 209.0353982300885,
      "grad_norm": 9.6875,
      "learning_rate": 3.983090616802396e-05,
      "loss": 1.1691,
      "num_input_tokens_seen": 6606352,
      "step": 11915
    },
    {
      "epoch": 209.12389380530973,
      "grad_norm": 9.875,
      "learning_rate": 3.982300167664788e-05,
      "loss": 1.1237,
      "num_input_tokens_seen": 6608960,
      "step": 11920
    },
    {
      "epoch": 209.21238938053096,
      "grad_norm": 7.6875,
      "learning_rate": 3.981509489937868e-05,
      "loss": 0.8979,
      "num_input_tokens_seen": 6611632,
      "step": 11925
    },
    {
      "epoch": 209.30088495575222,
      "grad_norm": 11.0625,
      "learning_rate": 3.9807185837435643e-05,
      "loss": 1.1213,
      "num_input_tokens_seen": 6614464,
      "step": 11930
    },
    {
      "epoch": 209.38938053097345,
      "grad_norm": 8.5625,
      "learning_rate": 3.9799274492038484e-05,
      "loss": 0.9553,
      "num_input_tokens_seen": 6617216,
      "step": 11935
    },
    {
      "epoch": 209.47787610619469,
      "grad_norm": 8.3125,
      "learning_rate": 3.979136086440722e-05,
      "loss": 1.091,
      "num_input_tokens_seen": 6620304,
      "step": 11940
    },
    {
      "epoch": 209.56637168141592,
      "grad_norm": 8.8125,
      "learning_rate": 3.9783444955762226e-05,
      "loss": 1.1076,
      "num_input_tokens_seen": 6623024,
      "step": 11945
    },
    {
      "epoch": 209.65486725663717,
      "grad_norm": 6.0625,
      "learning_rate": 3.977552676732424e-05,
      "loss": 1.0088,
      "num_input_tokens_seen": 6625936,
      "step": 11950
    },
    {
      "epoch": 209.7433628318584,
      "grad_norm": 6.90625,
      "learning_rate": 3.976760630031435e-05,
      "loss": 1.1027,
      "num_input_tokens_seen": 6628352,
      "step": 11955
    },
    {
      "epoch": 209.83185840707964,
      "grad_norm": 8.8125,
      "learning_rate": 3.975968355595398e-05,
      "loss": 0.9794,
      "num_input_tokens_seen": 6631936,
      "step": 11960
    },
    {
      "epoch": 209.9203539823009,
      "grad_norm": 10.5,
      "learning_rate": 3.9751758535464935e-05,
      "loss": 1.086,
      "num_input_tokens_seen": 6634512,
      "step": 11965
    },
    {
      "epoch": 210.0,
      "grad_norm": 5.875,
      "learning_rate": 3.9743831240069326e-05,
      "loss": 0.9972,
      "num_input_tokens_seen": 6637056,
      "step": 11970
    },
    {
      "epoch": 210.08849557522123,
      "grad_norm": 8.25,
      "learning_rate": 3.9735901670989675e-05,
      "loss": 1.1055,
      "num_input_tokens_seen": 6639680,
      "step": 11975
    },
    {
      "epoch": 210.1769911504425,
      "grad_norm": 6.40625,
      "learning_rate": 3.97279698294488e-05,
      "loss": 0.9759,
      "num_input_tokens_seen": 6642272,
      "step": 11980
    },
    {
      "epoch": 210.26548672566372,
      "grad_norm": 9.125,
      "learning_rate": 3.9720035716669876e-05,
      "loss": 1.1672,
      "num_input_tokens_seen": 6645120,
      "step": 11985
    },
    {
      "epoch": 210.35398230088495,
      "grad_norm": 8.0625,
      "learning_rate": 3.9712099333876474e-05,
      "loss": 1.0281,
      "num_input_tokens_seen": 6648496,
      "step": 11990
    },
    {
      "epoch": 210.44247787610618,
      "grad_norm": 10.5625,
      "learning_rate": 3.9704160682292475e-05,
      "loss": 0.9969,
      "num_input_tokens_seen": 6651104,
      "step": 11995
    },
    {
      "epoch": 210.53097345132744,
      "grad_norm": 7.4375,
      "learning_rate": 3.9696219763142106e-05,
      "loss": 1.0844,
      "num_input_tokens_seen": 6654160,
      "step": 12000
    },
    {
      "epoch": 210.53097345132744,
      "eval_loss": 1.029395580291748,
      "eval_runtime": 0.9354,
      "eval_samples_per_second": 26.727,
      "eval_steps_per_second": 13.898,
      "num_input_tokens_seen": 6654160,
      "step": 12000
    },
    {
      "epoch": 210.61946902654867,
      "grad_norm": 8.6875,
      "learning_rate": 3.968827657764997e-05,
      "loss": 1.0921,
      "num_input_tokens_seen": 6656592,
      "step": 12005
    },
    {
      "epoch": 210.7079646017699,
      "grad_norm": 8.25,
      "learning_rate": 3.9680331127041e-05,
      "loss": 0.9847,
      "num_input_tokens_seen": 6659376,
      "step": 12010
    },
    {
      "epoch": 210.79646017699116,
      "grad_norm": 7.53125,
      "learning_rate": 3.9672383412540495e-05,
      "loss": 1.0667,
      "num_input_tokens_seen": 6662192,
      "step": 12015
    },
    {
      "epoch": 210.8849557522124,
      "grad_norm": 8.1875,
      "learning_rate": 3.966443343537407e-05,
      "loss": 1.0653,
      "num_input_tokens_seen": 6665328,
      "step": 12020
    },
    {
      "epoch": 210.97345132743362,
      "grad_norm": 8.6875,
      "learning_rate": 3.965648119676772e-05,
      "loss": 0.9834,
      "num_input_tokens_seen": 6668192,
      "step": 12025
    },
    {
      "epoch": 211.05309734513276,
      "grad_norm": 7.9375,
      "learning_rate": 3.96485266979478e-05,
      "loss": 1.2388,
      "num_input_tokens_seen": 6670240,
      "step": 12030
    },
    {
      "epoch": 211.141592920354,
      "grad_norm": 9.75,
      "learning_rate": 3.9640569940140974e-05,
      "loss": 1.1767,
      "num_input_tokens_seen": 6672688,
      "step": 12035
    },
    {
      "epoch": 211.23008849557522,
      "grad_norm": 10.8125,
      "learning_rate": 3.963261092457428e-05,
      "loss": 0.9736,
      "num_input_tokens_seen": 6675520,
      "step": 12040
    },
    {
      "epoch": 211.31858407079645,
      "grad_norm": 8.4375,
      "learning_rate": 3.962464965247509e-05,
      "loss": 0.9476,
      "num_input_tokens_seen": 6678480,
      "step": 12045
    },
    {
      "epoch": 211.4070796460177,
      "grad_norm": 11.125,
      "learning_rate": 3.9616686125071135e-05,
      "loss": 0.981,
      "num_input_tokens_seen": 6681344,
      "step": 12050
    },
    {
      "epoch": 211.49557522123894,
      "grad_norm": 7.15625,
      "learning_rate": 3.9608720343590506e-05,
      "loss": 0.8445,
      "num_input_tokens_seen": 6683920,
      "step": 12055
    },
    {
      "epoch": 211.58407079646017,
      "grad_norm": 7.90625,
      "learning_rate": 3.960075230926161e-05,
      "loss": 1.0802,
      "num_input_tokens_seen": 6686912,
      "step": 12060
    },
    {
      "epoch": 211.67256637168143,
      "grad_norm": 8.25,
      "learning_rate": 3.959278202331322e-05,
      "loss": 1.01,
      "num_input_tokens_seen": 6689952,
      "step": 12065
    },
    {
      "epoch": 211.76106194690266,
      "grad_norm": 6.8125,
      "learning_rate": 3.958480948697446e-05,
      "loss": 1.0078,
      "num_input_tokens_seen": 6692784,
      "step": 12070
    },
    {
      "epoch": 211.8495575221239,
      "grad_norm": 9.6875,
      "learning_rate": 3.95768347014748e-05,
      "loss": 1.2415,
      "num_input_tokens_seen": 6695792,
      "step": 12075
    },
    {
      "epoch": 211.93805309734512,
      "grad_norm": 9.6875,
      "learning_rate": 3.956885766804404e-05,
      "loss": 1.2364,
      "num_input_tokens_seen": 6698848,
      "step": 12080
    },
    {
      "epoch": 212.01769911504425,
      "grad_norm": 10.625,
      "learning_rate": 3.956087838791235e-05,
      "loss": 1.1205,
      "num_input_tokens_seen": 6701600,
      "step": 12085
    },
    {
      "epoch": 212.10619469026548,
      "grad_norm": 9.1875,
      "learning_rate": 3.955289686231022e-05,
      "loss": 1.1148,
      "num_input_tokens_seen": 6704640,
      "step": 12090
    },
    {
      "epoch": 212.1946902654867,
      "grad_norm": 7.625,
      "learning_rate": 3.9544913092468504e-05,
      "loss": 1.068,
      "num_input_tokens_seen": 6707216,
      "step": 12095
    },
    {
      "epoch": 212.28318584070797,
      "grad_norm": 7.625,
      "learning_rate": 3.9536927079618425e-05,
      "loss": 1.0766,
      "num_input_tokens_seen": 6710080,
      "step": 12100
    },
    {
      "epoch": 212.3716814159292,
      "grad_norm": 7.90625,
      "learning_rate": 3.9528938824991494e-05,
      "loss": 1.0498,
      "num_input_tokens_seen": 6713152,
      "step": 12105
    },
    {
      "epoch": 212.46017699115043,
      "grad_norm": 9.625,
      "learning_rate": 3.952094832981962e-05,
      "loss": 0.9807,
      "num_input_tokens_seen": 6716032,
      "step": 12110
    },
    {
      "epoch": 212.5486725663717,
      "grad_norm": 8.5625,
      "learning_rate": 3.951295559533503e-05,
      "loss": 0.9109,
      "num_input_tokens_seen": 6719264,
      "step": 12115
    },
    {
      "epoch": 212.63716814159292,
      "grad_norm": 9.3125,
      "learning_rate": 3.95049606227703e-05,
      "loss": 1.1747,
      "num_input_tokens_seen": 6721728,
      "step": 12120
    },
    {
      "epoch": 212.72566371681415,
      "grad_norm": 8.0625,
      "learning_rate": 3.949696341335838e-05,
      "loss": 0.9441,
      "num_input_tokens_seen": 6724592,
      "step": 12125
    },
    {
      "epoch": 212.81415929203538,
      "grad_norm": 7.0625,
      "learning_rate": 3.9488963968332503e-05,
      "loss": 1.1583,
      "num_input_tokens_seen": 6727072,
      "step": 12130
    },
    {
      "epoch": 212.90265486725664,
      "grad_norm": 6.9375,
      "learning_rate": 3.948096228892631e-05,
      "loss": 0.9582,
      "num_input_tokens_seen": 6729488,
      "step": 12135
    },
    {
      "epoch": 212.99115044247787,
      "grad_norm": 8.25,
      "learning_rate": 3.947295837637375e-05,
      "loss": 1.1818,
      "num_input_tokens_seen": 6732160,
      "step": 12140
    },
    {
      "epoch": 213.07079646017698,
      "grad_norm": 10.125,
      "learning_rate": 3.9464952231909135e-05,
      "loss": 0.9879,
      "num_input_tokens_seen": 6734880,
      "step": 12145
    },
    {
      "epoch": 213.15929203539824,
      "grad_norm": 6.3125,
      "learning_rate": 3.945694385676711e-05,
      "loss": 0.8247,
      "num_input_tokens_seen": 6738096,
      "step": 12150
    },
    {
      "epoch": 213.24778761061947,
      "grad_norm": 9.125,
      "learning_rate": 3.944893325218265e-05,
      "loss": 1.3326,
      "num_input_tokens_seen": 6740512,
      "step": 12155
    },
    {
      "epoch": 213.3362831858407,
      "grad_norm": 7.78125,
      "learning_rate": 3.944092041939112e-05,
      "loss": 0.9842,
      "num_input_tokens_seen": 6743584,
      "step": 12160
    },
    {
      "epoch": 213.42477876106196,
      "grad_norm": 7.28125,
      "learning_rate": 3.943290535962818e-05,
      "loss": 0.9244,
      "num_input_tokens_seen": 6746240,
      "step": 12165
    },
    {
      "epoch": 213.5132743362832,
      "grad_norm": 7.0,
      "learning_rate": 3.942488807412985e-05,
      "loss": 0.9712,
      "num_input_tokens_seen": 6749248,
      "step": 12170
    },
    {
      "epoch": 213.60176991150442,
      "grad_norm": 10.625,
      "learning_rate": 3.941686856413251e-05,
      "loss": 1.1707,
      "num_input_tokens_seen": 6751808,
      "step": 12175
    },
    {
      "epoch": 213.69026548672565,
      "grad_norm": 8.25,
      "learning_rate": 3.9408846830872874e-05,
      "loss": 1.0503,
      "num_input_tokens_seen": 6754688,
      "step": 12180
    },
    {
      "epoch": 213.7787610619469,
      "grad_norm": 7.65625,
      "learning_rate": 3.940082287558798e-05,
      "loss": 1.1369,
      "num_input_tokens_seen": 6757664,
      "step": 12185
    },
    {
      "epoch": 213.86725663716814,
      "grad_norm": 7.65625,
      "learning_rate": 3.939279669951522e-05,
      "loss": 0.9791,
      "num_input_tokens_seen": 6760176,
      "step": 12190
    },
    {
      "epoch": 213.95575221238937,
      "grad_norm": 11.25,
      "learning_rate": 3.938476830389234e-05,
      "loss": 1.1239,
      "num_input_tokens_seen": 6762656,
      "step": 12195
    },
    {
      "epoch": 214.0353982300885,
      "grad_norm": 9.75,
      "learning_rate": 3.937673768995742e-05,
      "loss": 1.1491,
      "num_input_tokens_seen": 6765224,
      "step": 12200
    },
    {
      "epoch": 214.0353982300885,
      "eval_loss": 1.0173370838165283,
      "eval_runtime": 0.9312,
      "eval_samples_per_second": 26.847,
      "eval_steps_per_second": 13.96,
      "num_input_tokens_seen": 6765224,
      "step": 12200
    },
    {
      "epoch": 214.12389380530973,
      "grad_norm": 8.375,
      "learning_rate": 3.936870485894888e-05,
      "loss": 0.9047,
      "num_input_tokens_seen": 6767848,
      "step": 12205
    },
    {
      "epoch": 214.21238938053096,
      "grad_norm": 9.375,
      "learning_rate": 3.9360669812105475e-05,
      "loss": 1.0439,
      "num_input_tokens_seen": 6770792,
      "step": 12210
    },
    {
      "epoch": 214.30088495575222,
      "grad_norm": 7.78125,
      "learning_rate": 3.9352632550666325e-05,
      "loss": 1.0383,
      "num_input_tokens_seen": 6773704,
      "step": 12215
    },
    {
      "epoch": 214.38938053097345,
      "grad_norm": 9.3125,
      "learning_rate": 3.9344593075870866e-05,
      "loss": 1.1732,
      "num_input_tokens_seen": 6776232,
      "step": 12220
    },
    {
      "epoch": 214.47787610619469,
      "grad_norm": 7.96875,
      "learning_rate": 3.933655138895889e-05,
      "loss": 1.0538,
      "num_input_tokens_seen": 6778952,
      "step": 12225
    },
    {
      "epoch": 214.56637168141592,
      "grad_norm": 6.90625,
      "learning_rate": 3.932850749117053e-05,
      "loss": 1.0913,
      "num_input_tokens_seen": 6781352,
      "step": 12230
    },
    {
      "epoch": 214.65486725663717,
      "grad_norm": 9.875,
      "learning_rate": 3.932046138374624e-05,
      "loss": 1.2191,
      "num_input_tokens_seen": 6783976,
      "step": 12235
    },
    {
      "epoch": 214.7433628318584,
      "grad_norm": 6.6875,
      "learning_rate": 3.9312413067926854e-05,
      "loss": 0.9822,
      "num_input_tokens_seen": 6786488,
      "step": 12240
    },
    {
      "epoch": 214.83185840707964,
      "grad_norm": 9.0,
      "learning_rate": 3.9304362544953506e-05,
      "loss": 1.1682,
      "num_input_tokens_seen": 6789288,
      "step": 12245
    },
    {
      "epoch": 214.9203539823009,
      "grad_norm": 9.0625,
      "learning_rate": 3.929630981606769e-05,
      "loss": 1.0142,
      "num_input_tokens_seen": 6792152,
      "step": 12250
    },
    {
      "epoch": 215.0,
      "grad_norm": 5.53125,
      "learning_rate": 3.928825488251124e-05,
      "loss": 0.6983,
      "num_input_tokens_seen": 6794728,
      "step": 12255
    },
    {
      "epoch": 215.08849557522123,
      "grad_norm": 7.8125,
      "learning_rate": 3.9280197745526344e-05,
      "loss": 0.9664,
      "num_input_tokens_seen": 6797480,
      "step": 12260
    },
    {
      "epoch": 215.1769911504425,
      "grad_norm": 11.3125,
      "learning_rate": 3.9272138406355495e-05,
      "loss": 1.2312,
      "num_input_tokens_seen": 6800120,
      "step": 12265
    },
    {
      "epoch": 215.26548672566372,
      "grad_norm": 8.875,
      "learning_rate": 3.926407686624154e-05,
      "loss": 1.3077,
      "num_input_tokens_seen": 6802440,
      "step": 12270
    },
    {
      "epoch": 215.35398230088495,
      "grad_norm": 8.625,
      "learning_rate": 3.9256013126427684e-05,
      "loss": 0.9398,
      "num_input_tokens_seen": 6805064,
      "step": 12275
    },
    {
      "epoch": 215.44247787610618,
      "grad_norm": 7.375,
      "learning_rate": 3.9247947188157455e-05,
      "loss": 0.9408,
      "num_input_tokens_seen": 6808680,
      "step": 12280
    },
    {
      "epoch": 215.53097345132744,
      "grad_norm": 7.71875,
      "learning_rate": 3.9239879052674715e-05,
      "loss": 0.9846,
      "num_input_tokens_seen": 6811768,
      "step": 12285
    },
    {
      "epoch": 215.61946902654867,
      "grad_norm": 8.3125,
      "learning_rate": 3.9231808721223673e-05,
      "loss": 1.0375,
      "num_input_tokens_seen": 6814872,
      "step": 12290
    },
    {
      "epoch": 215.7079646017699,
      "grad_norm": 10.4375,
      "learning_rate": 3.9223736195048886e-05,
      "loss": 1.0411,
      "num_input_tokens_seen": 6817416,
      "step": 12295
    },
    {
      "epoch": 215.79646017699116,
      "grad_norm": 9.6875,
      "learning_rate": 3.921566147539523e-05,
      "loss": 1.0284,
      "num_input_tokens_seen": 6820296,
      "step": 12300
    },
    {
      "epoch": 215.8849557522124,
      "grad_norm": 13.0625,
      "learning_rate": 3.920758456350792e-05,
      "loss": 1.1471,
      "num_input_tokens_seen": 6822984,
      "step": 12305
    },
    {
      "epoch": 215.97345132743362,
      "grad_norm": 7.78125,
      "learning_rate": 3.919950546063253e-05,
      "loss": 0.9261,
      "num_input_tokens_seen": 6825624,
      "step": 12310
    },
    {
      "epoch": 216.05309734513276,
      "grad_norm": 9.8125,
      "learning_rate": 3.919142416801496e-05,
      "loss": 1.1569,
      "num_input_tokens_seen": 6827880,
      "step": 12315
    },
    {
      "epoch": 216.141592920354,
      "grad_norm": 8.4375,
      "learning_rate": 3.918334068690144e-05,
      "loss": 1.1971,
      "num_input_tokens_seen": 6830664,
      "step": 12320
    },
    {
      "epoch": 216.23008849557522,
      "grad_norm": 5.34375,
      "learning_rate": 3.917525501853855e-05,
      "loss": 0.8593,
      "num_input_tokens_seen": 6833336,
      "step": 12325
    },
    {
      "epoch": 216.31858407079645,
      "grad_norm": 8.5,
      "learning_rate": 3.916716716417319e-05,
      "loss": 1.0971,
      "num_input_tokens_seen": 6836136,
      "step": 12330
    },
    {
      "epoch": 216.4070796460177,
      "grad_norm": 11.1875,
      "learning_rate": 3.915907712505263e-05,
      "loss": 1.1996,
      "num_input_tokens_seen": 6838856,
      "step": 12335
    },
    {
      "epoch": 216.49557522123894,
      "grad_norm": 6.6875,
      "learning_rate": 3.915098490242444e-05,
      "loss": 1.0007,
      "num_input_tokens_seen": 6841560,
      "step": 12340
    },
    {
      "epoch": 216.58407079646017,
      "grad_norm": 8.125,
      "learning_rate": 3.914289049753654e-05,
      "loss": 1.1192,
      "num_input_tokens_seen": 6844568,
      "step": 12345
    },
    {
      "epoch": 216.67256637168143,
      "grad_norm": 8.25,
      "learning_rate": 3.913479391163719e-05,
      "loss": 1.1708,
      "num_input_tokens_seen": 6847416,
      "step": 12350
    },
    {
      "epoch": 216.76106194690266,
      "grad_norm": 10.25,
      "learning_rate": 3.9126695145975e-05,
      "loss": 0.8088,
      "num_input_tokens_seen": 6850360,
      "step": 12355
    },
    {
      "epoch": 216.8495575221239,
      "grad_norm": 6.5625,
      "learning_rate": 3.911859420179889e-05,
      "loss": 1.1548,
      "num_input_tokens_seen": 6853416,
      "step": 12360
    },
    {
      "epoch": 216.93805309734512,
      "grad_norm": 7.28125,
      "learning_rate": 3.911049108035813e-05,
      "loss": 0.9929,
      "num_input_tokens_seen": 6855976,
      "step": 12365
    },
    {
      "epoch": 217.01769911504425,
      "grad_norm": 8.1875,
      "learning_rate": 3.910238578290232e-05,
      "loss": 1.0996,
      "num_input_tokens_seen": 6858376,
      "step": 12370
    },
    {
      "epoch": 217.10619469026548,
      "grad_norm": 9.25,
      "learning_rate": 3.90942783106814e-05,
      "loss": 1.2722,
      "num_input_tokens_seen": 6860920,
      "step": 12375
    },
    {
      "epoch": 217.1946902654867,
      "grad_norm": 8.875,
      "learning_rate": 3.908616866494564e-05,
      "loss": 1.1947,
      "num_input_tokens_seen": 6863544,
      "step": 12380
    },
    {
      "epoch": 217.28318584070797,
      "grad_norm": 4.875,
      "learning_rate": 3.907805684694566e-05,
      "loss": 0.8667,
      "num_input_tokens_seen": 6866520,
      "step": 12385
    },
    {
      "epoch": 217.3716814159292,
      "grad_norm": 6.28125,
      "learning_rate": 3.90699428579324e-05,
      "loss": 1.0287,
      "num_input_tokens_seen": 6868936,
      "step": 12390
    },
    {
      "epoch": 217.46017699115043,
      "grad_norm": 10.3125,
      "learning_rate": 3.906182669915713e-05,
      "loss": 1.0969,
      "num_input_tokens_seen": 6871912,
      "step": 12395
    },
    {
      "epoch": 217.5486725663717,
      "grad_norm": 7.125,
      "learning_rate": 3.9053708371871476e-05,
      "loss": 0.9916,
      "num_input_tokens_seen": 6874936,
      "step": 12400
    },
    {
      "epoch": 217.5486725663717,
      "eval_loss": 1.0220698118209839,
      "eval_runtime": 0.9407,
      "eval_samples_per_second": 26.575,
      "eval_steps_per_second": 13.819,
      "num_input_tokens_seen": 6874936,
      "step": 12400
    },
    {
      "epoch": 217.63716814159292,
      "grad_norm": 7.84375,
      "learning_rate": 3.904558787732738e-05,
      "loss": 1.0261,
      "num_input_tokens_seen": 6877528,
      "step": 12405
    },
    {
      "epoch": 217.72566371681415,
      "grad_norm": 8.375,
      "learning_rate": 3.9037465216777135e-05,
      "loss": 1.0878,
      "num_input_tokens_seen": 6880568,
      "step": 12410
    },
    {
      "epoch": 217.81415929203538,
      "grad_norm": 7.53125,
      "learning_rate": 3.902934039147334e-05,
      "loss": 0.9885,
      "num_input_tokens_seen": 6883256,
      "step": 12415
    },
    {
      "epoch": 217.90265486725664,
      "grad_norm": 7.1875,
      "learning_rate": 3.902121340266894e-05,
      "loss": 1.0132,
      "num_input_tokens_seen": 6886232,
      "step": 12420
    },
    {
      "epoch": 217.99115044247787,
      "grad_norm": 9.1875,
      "learning_rate": 3.9013084251617246e-05,
      "loss": 1.0679,
      "num_input_tokens_seen": 6889016,
      "step": 12425
    },
    {
      "epoch": 218.07079646017698,
      "grad_norm": 9.875,
      "learning_rate": 3.9004952939571865e-05,
      "loss": 1.137,
      "num_input_tokens_seen": 6891560,
      "step": 12430
    },
    {
      "epoch": 218.15929203539824,
      "grad_norm": 7.5625,
      "learning_rate": 3.899681946778673e-05,
      "loss": 0.8378,
      "num_input_tokens_seen": 6894104,
      "step": 12435
    },
    {
      "epoch": 218.24778761061947,
      "grad_norm": 8.4375,
      "learning_rate": 3.898868383751615e-05,
      "loss": 1.1602,
      "num_input_tokens_seen": 6896824,
      "step": 12440
    },
    {
      "epoch": 218.3362831858407,
      "grad_norm": 8.4375,
      "learning_rate": 3.8980546050014724e-05,
      "loss": 1.0022,
      "num_input_tokens_seen": 6899544,
      "step": 12445
    },
    {
      "epoch": 218.42477876106196,
      "grad_norm": 10.6875,
      "learning_rate": 3.897240610653741e-05,
      "loss": 1.1343,
      "num_input_tokens_seen": 6902840,
      "step": 12450
    },
    {
      "epoch": 218.5132743362832,
      "grad_norm": 8.0,
      "learning_rate": 3.896426400833948e-05,
      "loss": 0.9031,
      "num_input_tokens_seen": 6905544,
      "step": 12455
    },
    {
      "epoch": 218.60176991150442,
      "grad_norm": 8.375,
      "learning_rate": 3.895611975667656e-05,
      "loss": 1.1312,
      "num_input_tokens_seen": 6908216,
      "step": 12460
    },
    {
      "epoch": 218.69026548672565,
      "grad_norm": 10.25,
      "learning_rate": 3.8947973352804584e-05,
      "loss": 1.1001,
      "num_input_tokens_seen": 6911416,
      "step": 12465
    },
    {
      "epoch": 218.7787610619469,
      "grad_norm": 8.1875,
      "learning_rate": 3.893982479797984e-05,
      "loss": 0.9875,
      "num_input_tokens_seen": 6914152,
      "step": 12470
    },
    {
      "epoch": 218.86725663716814,
      "grad_norm": 8.5625,
      "learning_rate": 3.8931674093458926e-05,
      "loss": 1.0965,
      "num_input_tokens_seen": 6916984,
      "step": 12475
    },
    {
      "epoch": 218.95575221238937,
      "grad_norm": 8.9375,
      "learning_rate": 3.89235212404988e-05,
      "loss": 1.1742,
      "num_input_tokens_seen": 6919928,
      "step": 12480
    },
    {
      "epoch": 219.0353982300885,
      "grad_norm": 7.34375,
      "learning_rate": 3.891536624035672e-05,
      "loss": 0.9436,
      "num_input_tokens_seen": 6922448,
      "step": 12485
    },
    {
      "epoch": 219.12389380530973,
      "grad_norm": 9.5,
      "learning_rate": 3.8907209094290295e-05,
      "loss": 1.0203,
      "num_input_tokens_seen": 6925168,
      "step": 12490
    },
    {
      "epoch": 219.21238938053096,
      "grad_norm": 8.875,
      "learning_rate": 3.8899049803557466e-05,
      "loss": 1.1607,
      "num_input_tokens_seen": 6928256,
      "step": 12495
    },
    {
      "epoch": 219.30088495575222,
      "grad_norm": 8.625,
      "learning_rate": 3.889088836941648e-05,
      "loss": 0.9751,
      "num_input_tokens_seen": 6931088,
      "step": 12500
    },
    {
      "epoch": 219.38938053097345,
      "grad_norm": 7.65625,
      "learning_rate": 3.8882724793125946e-05,
      "loss": 1.0471,
      "num_input_tokens_seen": 6933936,
      "step": 12505
    },
    {
      "epoch": 219.47787610619469,
      "grad_norm": 8.25,
      "learning_rate": 3.8874559075944794e-05,
      "loss": 1.2153,
      "num_input_tokens_seen": 6936784,
      "step": 12510
    },
    {
      "epoch": 219.56637168141592,
      "grad_norm": 7.0,
      "learning_rate": 3.886639121913227e-05,
      "loss": 1.0226,
      "num_input_tokens_seen": 6939824,
      "step": 12515
    },
    {
      "epoch": 219.65486725663717,
      "grad_norm": 6.375,
      "learning_rate": 3.885822122394797e-05,
      "loss": 1.1754,
      "num_input_tokens_seen": 6942720,
      "step": 12520
    },
    {
      "epoch": 219.7433628318584,
      "grad_norm": 9.5625,
      "learning_rate": 3.8850049091651794e-05,
      "loss": 1.194,
      "num_input_tokens_seen": 6945344,
      "step": 12525
    },
    {
      "epoch": 219.83185840707964,
      "grad_norm": 8.625,
      "learning_rate": 3.8841874823504e-05,
      "loss": 0.9865,
      "num_input_tokens_seen": 6948080,
      "step": 12530
    },
    {
      "epoch": 219.9203539823009,
      "grad_norm": 8.0625,
      "learning_rate": 3.8833698420765157e-05,
      "loss": 0.9027,
      "num_input_tokens_seen": 6950528,
      "step": 12535
    },
    {
      "epoch": 220.0,
      "grad_norm": 8.625,
      "learning_rate": 3.882551988469618e-05,
      "loss": 0.9606,
      "num_input_tokens_seen": 6952792,
      "step": 12540
    },
    {
      "epoch": 220.08849557522123,
      "grad_norm": 8.625,
      "learning_rate": 3.881733921655829e-05,
      "loss": 0.9789,
      "num_input_tokens_seen": 6955480,
      "step": 12545
    },
    {
      "epoch": 220.1769911504425,
      "grad_norm": 8.3125,
      "learning_rate": 3.8809156417613054e-05,
      "loss": 1.1284,
      "num_input_tokens_seen": 6958152,
      "step": 12550
    },
    {
      "epoch": 220.26548672566372,
      "grad_norm": 7.84375,
      "learning_rate": 3.8800971489122364e-05,
      "loss": 1.2103,
      "num_input_tokens_seen": 6960792,
      "step": 12555
    },
    {
      "epoch": 220.35398230088495,
      "grad_norm": 10.125,
      "learning_rate": 3.8792784432348434e-05,
      "loss": 1.0374,
      "num_input_tokens_seen": 6964408,
      "step": 12560
    },
    {
      "epoch": 220.44247787610618,
      "grad_norm": 10.125,
      "learning_rate": 3.878459524855381e-05,
      "loss": 0.8469,
      "num_input_tokens_seen": 6967352,
      "step": 12565
    },
    {
      "epoch": 220.53097345132744,
      "grad_norm": 10.0625,
      "learning_rate": 3.8776403939001384e-05,
      "loss": 1.2037,
      "num_input_tokens_seen": 6969960,
      "step": 12570
    },
    {
      "epoch": 220.61946902654867,
      "grad_norm": 13.125,
      "learning_rate": 3.876821050495433e-05,
      "loss": 1.0432,
      "num_input_tokens_seen": 6972744,
      "step": 12575
    },
    {
      "epoch": 220.7079646017699,
      "grad_norm": 7.90625,
      "learning_rate": 3.87600149476762e-05,
      "loss": 1.1973,
      "num_input_tokens_seen": 6975784,
      "step": 12580
    },
    {
      "epoch": 220.79646017699116,
      "grad_norm": 6.09375,
      "learning_rate": 3.8751817268430843e-05,
      "loss": 0.7633,
      "num_input_tokens_seen": 6978648,
      "step": 12585
    },
    {
      "epoch": 220.8849557522124,
      "grad_norm": 8.0625,
      "learning_rate": 3.8743617468482464e-05,
      "loss": 1.1976,
      "num_input_tokens_seen": 6981176,
      "step": 12590
    },
    {
      "epoch": 220.97345132743362,
      "grad_norm": 8.9375,
      "learning_rate": 3.8735415549095535e-05,
      "loss": 1.0704,
      "num_input_tokens_seen": 6983976,
      "step": 12595
    },
    {
      "epoch": 221.05309734513276,
      "grad_norm": 7.4375,
      "learning_rate": 3.8727211511534934e-05,
      "loss": 1.0089,
      "num_input_tokens_seen": 6986248,
      "step": 12600
    },
    {
      "epoch": 221.05309734513276,
      "eval_loss": 1.0246447324752808,
      "eval_runtime": 0.934,
      "eval_samples_per_second": 26.766,
      "eval_steps_per_second": 13.919,
      "num_input_tokens_seen": 6986248,
      "step": 12600
    },
    {
      "epoch": 221.141592920354,
      "grad_norm": 8.5,
      "learning_rate": 3.8719005357065804e-05,
      "loss": 1.0293,
      "num_input_tokens_seen": 6989032,
      "step": 12605
    },
    {
      "epoch": 221.23008849557522,
      "grad_norm": 8.0625,
      "learning_rate": 3.8710797086953645e-05,
      "loss": 1.0106,
      "num_input_tokens_seen": 6991976,
      "step": 12610
    },
    {
      "epoch": 221.31858407079645,
      "grad_norm": 7.90625,
      "learning_rate": 3.870258670246427e-05,
      "loss": 1.0662,
      "num_input_tokens_seen": 6994824,
      "step": 12615
    },
    {
      "epoch": 221.4070796460177,
      "grad_norm": 9.0625,
      "learning_rate": 3.869437420486384e-05,
      "loss": 1.0651,
      "num_input_tokens_seen": 6997672,
      "step": 12620
    },
    {
      "epoch": 221.49557522123894,
      "grad_norm": 10.625,
      "learning_rate": 3.8686159595418805e-05,
      "loss": 1.1979,
      "num_input_tokens_seen": 7000648,
      "step": 12625
    },
    {
      "epoch": 221.58407079646017,
      "grad_norm": 7.40625,
      "learning_rate": 3.867794287539597e-05,
      "loss": 0.812,
      "num_input_tokens_seen": 7003576,
      "step": 12630
    },
    {
      "epoch": 221.67256637168143,
      "grad_norm": 9.75,
      "learning_rate": 3.866972404606245e-05,
      "loss": 1.1077,
      "num_input_tokens_seen": 7006104,
      "step": 12635
    },
    {
      "epoch": 221.76106194690266,
      "grad_norm": 8.8125,
      "learning_rate": 3.866150310868571e-05,
      "loss": 1.1016,
      "num_input_tokens_seen": 7009320,
      "step": 12640
    },
    {
      "epoch": 221.8495575221239,
      "grad_norm": 7.8125,
      "learning_rate": 3.8653280064533506e-05,
      "loss": 0.9995,
      "num_input_tokens_seen": 7012232,
      "step": 12645
    },
    {
      "epoch": 221.93805309734512,
      "grad_norm": 8.0,
      "learning_rate": 3.864505491487394e-05,
      "loss": 1.1176,
      "num_input_tokens_seen": 7014584,
      "step": 12650
    },
    {
      "epoch": 222.01769911504425,
      "grad_norm": 7.90625,
      "learning_rate": 3.8636827660975414e-05,
      "loss": 1.1204,
      "num_input_tokens_seen": 7017128,
      "step": 12655
    },
    {
      "epoch": 222.10619469026548,
      "grad_norm": 7.6875,
      "learning_rate": 3.862859830410671e-05,
      "loss": 1.0456,
      "num_input_tokens_seen": 7019960,
      "step": 12660
    },
    {
      "epoch": 222.1946902654867,
      "grad_norm": 8.0625,
      "learning_rate": 3.862036684553688e-05,
      "loss": 0.9755,
      "num_input_tokens_seen": 7022520,
      "step": 12665
    },
    {
      "epoch": 222.28318584070797,
      "grad_norm": 7.53125,
      "learning_rate": 3.8612133286535314e-05,
      "loss": 0.9732,
      "num_input_tokens_seen": 7025656,
      "step": 12670
    },
    {
      "epoch": 222.3716814159292,
      "grad_norm": 8.4375,
      "learning_rate": 3.860389762837173e-05,
      "loss": 1.0127,
      "num_input_tokens_seen": 7028792,
      "step": 12675
    },
    {
      "epoch": 222.46017699115043,
      "grad_norm": 8.125,
      "learning_rate": 3.859565987231618e-05,
      "loss": 1.0853,
      "num_input_tokens_seen": 7031752,
      "step": 12680
    },
    {
      "epoch": 222.5486725663717,
      "grad_norm": 9.75,
      "learning_rate": 3.858742001963902e-05,
      "loss": 1.1346,
      "num_input_tokens_seen": 7034520,
      "step": 12685
    },
    {
      "epoch": 222.63716814159292,
      "grad_norm": 7.71875,
      "learning_rate": 3.857917807161094e-05,
      "loss": 1.0108,
      "num_input_tokens_seen": 7037496,
      "step": 12690
    },
    {
      "epoch": 222.72566371681415,
      "grad_norm": 8.125,
      "learning_rate": 3.857093402950296e-05,
      "loss": 0.9916,
      "num_input_tokens_seen": 7040040,
      "step": 12695
    },
    {
      "epoch": 222.81415929203538,
      "grad_norm": 9.0,
      "learning_rate": 3.8562687894586414e-05,
      "loss": 0.9516,
      "num_input_tokens_seen": 7042552,
      "step": 12700
    },
    {
      "epoch": 222.90265486725664,
      "grad_norm": 9.875,
      "learning_rate": 3.8554439668132946e-05,
      "loss": 1.2324,
      "num_input_tokens_seen": 7045480,
      "step": 12705
    },
    {
      "epoch": 222.99115044247787,
      "grad_norm": 7.375,
      "learning_rate": 3.854618935141455e-05,
      "loss": 1.191,
      "num_input_tokens_seen": 7048040,
      "step": 12710
    },
    {
      "epoch": 223.07079646017698,
      "grad_norm": 9.375,
      "learning_rate": 3.8537936945703525e-05,
      "loss": 1.1363,
      "num_input_tokens_seen": 7050456,
      "step": 12715
    },
    {
      "epoch": 223.15929203539824,
      "grad_norm": 9.4375,
      "learning_rate": 3.852968245227249e-05,
      "loss": 0.9962,
      "num_input_tokens_seen": 7053224,
      "step": 12720
    },
    {
      "epoch": 223.24778761061947,
      "grad_norm": 10.125,
      "learning_rate": 3.85214258723944e-05,
      "loss": 1.1436,
      "num_input_tokens_seen": 7055688,
      "step": 12725
    },
    {
      "epoch": 223.3362831858407,
      "grad_norm": 7.75,
      "learning_rate": 3.8513167207342524e-05,
      "loss": 1.2102,
      "num_input_tokens_seen": 7058312,
      "step": 12730
    },
    {
      "epoch": 223.42477876106196,
      "grad_norm": 12.6875,
      "learning_rate": 3.850490645839044e-05,
      "loss": 0.9964,
      "num_input_tokens_seen": 7061576,
      "step": 12735
    },
    {
      "epoch": 223.5132743362832,
      "grad_norm": 8.375,
      "learning_rate": 3.849664362681207e-05,
      "loss": 1.0011,
      "num_input_tokens_seen": 7064632,
      "step": 12740
    },
    {
      "epoch": 223.60176991150442,
      "grad_norm": 6.15625,
      "learning_rate": 3.848837871388165e-05,
      "loss": 0.9013,
      "num_input_tokens_seen": 7067512,
      "step": 12745
    },
    {
      "epoch": 223.69026548672565,
      "grad_norm": 9.375,
      "learning_rate": 3.848011172087371e-05,
      "loss": 1.1136,
      "num_input_tokens_seen": 7070312,
      "step": 12750
    },
    {
      "epoch": 223.7787610619469,
      "grad_norm": 9.125,
      "learning_rate": 3.847184264906315e-05,
      "loss": 0.9751,
      "num_input_tokens_seen": 7073272,
      "step": 12755
    },
    {
      "epoch": 223.86725663716814,
      "grad_norm": 8.25,
      "learning_rate": 3.846357149972516e-05,
      "loss": 1.0186,
      "num_input_tokens_seen": 7076088,
      "step": 12760
    },
    {
      "epoch": 223.95575221238937,
      "grad_norm": 9.0,
      "learning_rate": 3.8455298274135246e-05,
      "loss": 1.155,
      "num_input_tokens_seen": 7078408,
      "step": 12765
    },
    {
      "epoch": 224.0353982300885,
      "grad_norm": 10.0,
      "learning_rate": 3.8447022973569254e-05,
      "loss": 1.088,
      "num_input_tokens_seen": 7080800,
      "step": 12770
    },
    {
      "epoch": 224.12389380530973,
      "grad_norm": 6.84375,
      "learning_rate": 3.843874559930332e-05,
      "loss": 0.9569,
      "num_input_tokens_seen": 7084032,
      "step": 12775
    },
    {
      "epoch": 224.21238938053096,
      "grad_norm": 9.375,
      "learning_rate": 3.843046615261394e-05,
      "loss": 0.8921,
      "num_input_tokens_seen": 7086992,
      "step": 12780
    },
    {
      "epoch": 224.30088495575222,
      "grad_norm": 8.9375,
      "learning_rate": 3.842218463477791e-05,
      "loss": 1.0573,
      "num_input_tokens_seen": 7089888,
      "step": 12785
    },
    {
      "epoch": 224.38938053097345,
      "grad_norm": 8.5625,
      "learning_rate": 3.841390104707233e-05,
      "loss": 1.0486,
      "num_input_tokens_seen": 7092656,
      "step": 12790
    },
    {
      "epoch": 224.47787610619469,
      "grad_norm": 9.875,
      "learning_rate": 3.8405615390774643e-05,
      "loss": 1.0303,
      "num_input_tokens_seen": 7095120,
      "step": 12795
    },
    {
      "epoch": 224.56637168141592,
      "grad_norm": 8.0,
      "learning_rate": 3.839732766716259e-05,
      "loss": 0.9634,
      "num_input_tokens_seen": 7097808,
      "step": 12800
    },
    {
      "epoch": 224.56637168141592,
      "eval_loss": 1.0277111530303955,
      "eval_runtime": 0.9359,
      "eval_samples_per_second": 26.712,
      "eval_steps_per_second": 13.89,
      "num_input_tokens_seen": 7097808,
      "step": 12800
    },
    {
      "epoch": 224.65486725663717,
      "grad_norm": 6.75,
      "learning_rate": 3.838903787751425e-05,
      "loss": 1.2229,
      "num_input_tokens_seen": 7100720,
      "step": 12805
    },
    {
      "epoch": 224.7433628318584,
      "grad_norm": 12.25,
      "learning_rate": 3.838074602310802e-05,
      "loss": 1.0871,
      "num_input_tokens_seen": 7103552,
      "step": 12810
    },
    {
      "epoch": 224.83185840707964,
      "grad_norm": 9.9375,
      "learning_rate": 3.837245210522258e-05,
      "loss": 1.2754,
      "num_input_tokens_seen": 7106112,
      "step": 12815
    },
    {
      "epoch": 224.9203539823009,
      "grad_norm": 8.4375,
      "learning_rate": 3.8364156125136996e-05,
      "loss": 1.1087,
      "num_input_tokens_seen": 7108736,
      "step": 12820
    },
    {
      "epoch": 225.0,
      "grad_norm": 6.71875,
      "learning_rate": 3.835585808413059e-05,
      "loss": 0.7855,
      "num_input_tokens_seen": 7111520,
      "step": 12825
    },
    {
      "epoch": 225.08849557522123,
      "grad_norm": 8.375,
      "learning_rate": 3.8347557983483024e-05,
      "loss": 1.1041,
      "num_input_tokens_seen": 7114384,
      "step": 12830
    },
    {
      "epoch": 225.1769911504425,
      "grad_norm": 7.375,
      "learning_rate": 3.833925582447428e-05,
      "loss": 1.1246,
      "num_input_tokens_seen": 7117072,
      "step": 12835
    },
    {
      "epoch": 225.26548672566372,
      "grad_norm": 7.4375,
      "learning_rate": 3.8330951608384656e-05,
      "loss": 0.8898,
      "num_input_tokens_seen": 7119808,
      "step": 12840
    },
    {
      "epoch": 225.35398230088495,
      "grad_norm": 8.4375,
      "learning_rate": 3.832264533649477e-05,
      "loss": 1.0554,
      "num_input_tokens_seen": 7122880,
      "step": 12845
    },
    {
      "epoch": 225.44247787610618,
      "grad_norm": 10.1875,
      "learning_rate": 3.8314337010085555e-05,
      "loss": 1.2508,
      "num_input_tokens_seen": 7126048,
      "step": 12850
    },
    {
      "epoch": 225.53097345132744,
      "grad_norm": 8.9375,
      "learning_rate": 3.830602663043824e-05,
      "loss": 1.0253,
      "num_input_tokens_seen": 7128816,
      "step": 12855
    },
    {
      "epoch": 225.61946902654867,
      "grad_norm": 7.6875,
      "learning_rate": 3.8297714198834414e-05,
      "loss": 0.9847,
      "num_input_tokens_seen": 7131728,
      "step": 12860
    },
    {
      "epoch": 225.7079646017699,
      "grad_norm": 10.875,
      "learning_rate": 3.828939971655595e-05,
      "loss": 0.9102,
      "num_input_tokens_seen": 7134448,
      "step": 12865
    },
    {
      "epoch": 225.79646017699116,
      "grad_norm": 7.6875,
      "learning_rate": 3.828108318488505e-05,
      "loss": 0.9833,
      "num_input_tokens_seen": 7137152,
      "step": 12870
    },
    {
      "epoch": 225.8849557522124,
      "grad_norm": 7.03125,
      "learning_rate": 3.8272764605104216e-05,
      "loss": 1.245,
      "num_input_tokens_seen": 7140256,
      "step": 12875
    },
    {
      "epoch": 225.97345132743362,
      "grad_norm": 7.78125,
      "learning_rate": 3.826444397849628e-05,
      "loss": 1.0038,
      "num_input_tokens_seen": 7142560,
      "step": 12880
    },
    {
      "epoch": 226.05309734513276,
      "grad_norm": 7.84375,
      "learning_rate": 3.825612130634439e-05,
      "loss": 1.0236,
      "num_input_tokens_seen": 7144672,
      "step": 12885
    },
    {
      "epoch": 226.141592920354,
      "grad_norm": 7.875,
      "learning_rate": 3.824779658993202e-05,
      "loss": 1.1873,
      "num_input_tokens_seen": 7147584,
      "step": 12890
    },
    {
      "epoch": 226.23008849557522,
      "grad_norm": 7.46875,
      "learning_rate": 3.823946983054292e-05,
      "loss": 0.983,
      "num_input_tokens_seen": 7150272,
      "step": 12895
    },
    {
      "epoch": 226.31858407079645,
      "grad_norm": 9.5,
      "learning_rate": 3.82311410294612e-05,
      "loss": 1.1938,
      "num_input_tokens_seen": 7153152,
      "step": 12900
    },
    {
      "epoch": 226.4070796460177,
      "grad_norm": 5.59375,
      "learning_rate": 3.822281018797127e-05,
      "loss": 1.0398,
      "num_input_tokens_seen": 7155584,
      "step": 12905
    },
    {
      "epoch": 226.49557522123894,
      "grad_norm": 8.4375,
      "learning_rate": 3.821447730735783e-05,
      "loss": 0.9769,
      "num_input_tokens_seen": 7158176,
      "step": 12910
    },
    {
      "epoch": 226.58407079646017,
      "grad_norm": 9.625,
      "learning_rate": 3.820614238890592e-05,
      "loss": 1.045,
      "num_input_tokens_seen": 7161056,
      "step": 12915
    },
    {
      "epoch": 226.67256637168143,
      "grad_norm": 7.875,
      "learning_rate": 3.819780543390091e-05,
      "loss": 1.0476,
      "num_input_tokens_seen": 7164176,
      "step": 12920
    },
    {
      "epoch": 226.76106194690266,
      "grad_norm": 9.0,
      "learning_rate": 3.818946644362844e-05,
      "loss": 1.1027,
      "num_input_tokens_seen": 7166752,
      "step": 12925
    },
    {
      "epoch": 226.8495575221239,
      "grad_norm": 6.25,
      "learning_rate": 3.81811254193745e-05,
      "loss": 1.0028,
      "num_input_tokens_seen": 7169600,
      "step": 12930
    },
    {
      "epoch": 226.93805309734512,
      "grad_norm": 8.375,
      "learning_rate": 3.8172782362425366e-05,
      "loss": 1.1358,
      "num_input_tokens_seen": 7172416,
      "step": 12935
    },
    {
      "epoch": 227.01769911504425,
      "grad_norm": 8.5625,
      "learning_rate": 3.816443727406765e-05,
      "loss": 0.981,
      "num_input_tokens_seen": 7175232,
      "step": 12940
    },
    {
      "epoch": 227.10619469026548,
      "grad_norm": 9.0,
      "learning_rate": 3.815609015558829e-05,
      "loss": 1.0729,
      "num_input_tokens_seen": 7178160,
      "step": 12945
    },
    {
      "epoch": 227.1946902654867,
      "grad_norm": 8.4375,
      "learning_rate": 3.814774100827448e-05,
      "loss": 1.0421,
      "num_input_tokens_seen": 7181136,
      "step": 12950
    },
    {
      "epoch": 227.28318584070797,
      "grad_norm": 9.5,
      "learning_rate": 3.813938983341379e-05,
      "loss": 1.0308,
      "num_input_tokens_seen": 7183952,
      "step": 12955
    },
    {
      "epoch": 227.3716814159292,
      "grad_norm": 8.9375,
      "learning_rate": 3.813103663229407e-05,
      "loss": 1.1111,
      "num_input_tokens_seen": 7186480,
      "step": 12960
    },
    {
      "epoch": 227.46017699115043,
      "grad_norm": 8.25,
      "learning_rate": 3.812268140620349e-05,
      "loss": 1.0412,
      "num_input_tokens_seen": 7189344,
      "step": 12965
    },
    {
      "epoch": 227.5486725663717,
      "grad_norm": 7.9375,
      "learning_rate": 3.811432415643051e-05,
      "loss": 0.9502,
      "num_input_tokens_seen": 7192000,
      "step": 12970
    },
    {
      "epoch": 227.63716814159292,
      "grad_norm": 6.8125,
      "learning_rate": 3.8105964884263954e-05,
      "loss": 1.0041,
      "num_input_tokens_seen": 7195088,
      "step": 12975
    },
    {
      "epoch": 227.72566371681415,
      "grad_norm": 6.84375,
      "learning_rate": 3.809760359099291e-05,
      "loss": 1.1481,
      "num_input_tokens_seen": 7198032,
      "step": 12980
    },
    {
      "epoch": 227.81415929203538,
      "grad_norm": 8.6875,
      "learning_rate": 3.8089240277906804e-05,
      "loss": 0.9994,
      "num_input_tokens_seen": 7200832,
      "step": 12985
    },
    {
      "epoch": 227.90265486725664,
      "grad_norm": 8.4375,
      "learning_rate": 3.808087494629535e-05,
      "loss": 1.0341,
      "num_input_tokens_seen": 7203776,
      "step": 12990
    },
    {
      "epoch": 227.99115044247787,
      "grad_norm": 12.4375,
      "learning_rate": 3.8072507597448595e-05,
      "loss": 1.1922,
      "num_input_tokens_seen": 7206336,
      "step": 12995
    },
    {
      "epoch": 228.07079646017698,
      "grad_norm": 14.875,
      "learning_rate": 3.806413823265689e-05,
      "loss": 1.348,
      "num_input_tokens_seen": 7208392,
      "step": 13000
    },
    {
      "epoch": 228.07079646017698,
      "eval_loss": 1.0306082963943481,
      "eval_runtime": 0.9312,
      "eval_samples_per_second": 26.847,
      "eval_steps_per_second": 13.96,
      "num_input_tokens_seen": 7208392,
      "step": 13000
    },
    {
      "epoch": 228.15929203539824,
      "grad_norm": 8.5625,
      "learning_rate": 3.805576685321089e-05,
      "loss": 1.2336,
      "num_input_tokens_seen": 7210872,
      "step": 13005
    },
    {
      "epoch": 228.24778761061947,
      "grad_norm": 8.25,
      "learning_rate": 3.804739346040158e-05,
      "loss": 1.2037,
      "num_input_tokens_seen": 7213528,
      "step": 13010
    },
    {
      "epoch": 228.3362831858407,
      "grad_norm": 10.125,
      "learning_rate": 3.8039018055520234e-05,
      "loss": 1.1865,
      "num_input_tokens_seen": 7216200,
      "step": 13015
    },
    {
      "epoch": 228.42477876106196,
      "grad_norm": 8.5625,
      "learning_rate": 3.803064063985844e-05,
      "loss": 0.9586,
      "num_input_tokens_seen": 7218696,
      "step": 13020
    },
    {
      "epoch": 228.5132743362832,
      "grad_norm": 7.53125,
      "learning_rate": 3.802226121470811e-05,
      "loss": 1.0252,
      "num_input_tokens_seen": 7221608,
      "step": 13025
    },
    {
      "epoch": 228.60176991150442,
      "grad_norm": 9.625,
      "learning_rate": 3.801387978136145e-05,
      "loss": 1.2546,
      "num_input_tokens_seen": 7224408,
      "step": 13030
    },
    {
      "epoch": 228.69026548672565,
      "grad_norm": 7.40625,
      "learning_rate": 3.800549634111099e-05,
      "loss": 0.7514,
      "num_input_tokens_seen": 7227512,
      "step": 13035
    },
    {
      "epoch": 228.7787610619469,
      "grad_norm": 7.25,
      "learning_rate": 3.799711089524955e-05,
      "loss": 0.9508,
      "num_input_tokens_seen": 7230216,
      "step": 13040
    },
    {
      "epoch": 228.86725663716814,
      "grad_norm": 6.03125,
      "learning_rate": 3.7988723445070285e-05,
      "loss": 0.9398,
      "num_input_tokens_seen": 7233064,
      "step": 13045
    },
    {
      "epoch": 228.95575221238937,
      "grad_norm": 7.375,
      "learning_rate": 3.798033399186663e-05,
      "loss": 0.9114,
      "num_input_tokens_seen": 7235992,
      "step": 13050
    },
    {
      "epoch": 229.0353982300885,
      "grad_norm": 8.0625,
      "learning_rate": 3.797194253693237e-05,
      "loss": 0.9502,
      "num_input_tokens_seen": 7238392,
      "step": 13055
    },
    {
      "epoch": 229.12389380530973,
      "grad_norm": 8.1875,
      "learning_rate": 3.796354908156153e-05,
      "loss": 1.1017,
      "num_input_tokens_seen": 7240984,
      "step": 13060
    },
    {
      "epoch": 229.21238938053096,
      "grad_norm": 8.9375,
      "learning_rate": 3.795515362704853e-05,
      "loss": 1.2812,
      "num_input_tokens_seen": 7243864,
      "step": 13065
    },
    {
      "epoch": 229.30088495575222,
      "grad_norm": 8.3125,
      "learning_rate": 3.794675617468803e-05,
      "loss": 0.9756,
      "num_input_tokens_seen": 7246184,
      "step": 13070
    },
    {
      "epoch": 229.38938053097345,
      "grad_norm": 12.5625,
      "learning_rate": 3.793835672577503e-05,
      "loss": 0.9522,
      "num_input_tokens_seen": 7249224,
      "step": 13075
    },
    {
      "epoch": 229.47787610619469,
      "grad_norm": 7.3125,
      "learning_rate": 3.7929955281604826e-05,
      "loss": 1.0721,
      "num_input_tokens_seen": 7251896,
      "step": 13080
    },
    {
      "epoch": 229.56637168141592,
      "grad_norm": 7.40625,
      "learning_rate": 3.7921551843473036e-05,
      "loss": 1.1757,
      "num_input_tokens_seen": 7254824,
      "step": 13085
    },
    {
      "epoch": 229.65486725663717,
      "grad_norm": 7.71875,
      "learning_rate": 3.791314641267557e-05,
      "loss": 0.953,
      "num_input_tokens_seen": 7257656,
      "step": 13090
    },
    {
      "epoch": 229.7433628318584,
      "grad_norm": 9.75,
      "learning_rate": 3.790473899050864e-05,
      "loss": 0.9738,
      "num_input_tokens_seen": 7260728,
      "step": 13095
    },
    {
      "epoch": 229.83185840707964,
      "grad_norm": 8.8125,
      "learning_rate": 3.7896329578268794e-05,
      "loss": 0.9674,
      "num_input_tokens_seen": 7263944,
      "step": 13100
    },
    {
      "epoch": 229.9203539823009,
      "grad_norm": 8.1875,
      "learning_rate": 3.7887918177252855e-05,
      "loss": 1.1226,
      "num_input_tokens_seen": 7266696,
      "step": 13105
    },
    {
      "epoch": 230.0,
      "grad_norm": 8.625,
      "learning_rate": 3.787950478875798e-05,
      "loss": 1.1544,
      "num_input_tokens_seen": 7269064,
      "step": 13110
    },
    {
      "epoch": 230.08849557522123,
      "grad_norm": 12.9375,
      "learning_rate": 3.787108941408162e-05,
      "loss": 1.0258,
      "num_input_tokens_seen": 7271656,
      "step": 13115
    },
    {
      "epoch": 230.1769911504425,
      "grad_norm": 13.4375,
      "learning_rate": 3.786267205452151e-05,
      "loss": 1.0635,
      "num_input_tokens_seen": 7274200,
      "step": 13120
    },
    {
      "epoch": 230.26548672566372,
      "grad_norm": 10.1875,
      "learning_rate": 3.785425271137573e-05,
      "loss": 1.2456,
      "num_input_tokens_seen": 7276632,
      "step": 13125
    },
    {
      "epoch": 230.35398230088495,
      "grad_norm": 9.0625,
      "learning_rate": 3.7845831385942655e-05,
      "loss": 1.0689,
      "num_input_tokens_seen": 7279624,
      "step": 13130
    },
    {
      "epoch": 230.44247787610618,
      "grad_norm": 7.46875,
      "learning_rate": 3.7837408079520944e-05,
      "loss": 0.9676,
      "num_input_tokens_seen": 7282920,
      "step": 13135
    },
    {
      "epoch": 230.53097345132744,
      "grad_norm": 8.1875,
      "learning_rate": 3.782898279340957e-05,
      "loss": 1.2863,
      "num_input_tokens_seen": 7285560,
      "step": 13140
    },
    {
      "epoch": 230.61946902654867,
      "grad_norm": 10.1875,
      "learning_rate": 3.782055552890784e-05,
      "loss": 0.9464,
      "num_input_tokens_seen": 7288488,
      "step": 13145
    },
    {
      "epoch": 230.7079646017699,
      "grad_norm": 9.5,
      "learning_rate": 3.781212628731534e-05,
      "loss": 0.985,
      "num_input_tokens_seen": 7290952,
      "step": 13150
    },
    {
      "epoch": 230.79646017699116,
      "grad_norm": 6.75,
      "learning_rate": 3.7803695069931946e-05,
      "loss": 0.9639,
      "num_input_tokens_seen": 7293896,
      "step": 13155
    },
    {
      "epoch": 230.8849557522124,
      "grad_norm": 8.375,
      "learning_rate": 3.779526187805789e-05,
      "loss": 0.9164,
      "num_input_tokens_seen": 7297320,
      "step": 13160
    },
    {
      "epoch": 230.97345132743362,
      "grad_norm": 8.625,
      "learning_rate": 3.778682671299364e-05,
      "loss": 1.1137,
      "num_input_tokens_seen": 7299992,
      "step": 13165
    },
    {
      "epoch": 231.05309734513276,
      "grad_norm": 9.5,
      "learning_rate": 3.777838957604003e-05,
      "loss": 1.2781,
      "num_input_tokens_seen": 7302264,
      "step": 13170
    },
    {
      "epoch": 231.141592920354,
      "grad_norm": 9.1875,
      "learning_rate": 3.776995046849816e-05,
      "loss": 1.1023,
      "num_input_tokens_seen": 7305128,
      "step": 13175
    },
    {
      "epoch": 231.23008849557522,
      "grad_norm": 7.84375,
      "learning_rate": 3.776150939166945e-05,
      "loss": 0.9664,
      "num_input_tokens_seen": 7307672,
      "step": 13180
    },
    {
      "epoch": 231.31858407079645,
      "grad_norm": 14.25,
      "learning_rate": 3.775306634685562e-05,
      "loss": 1.1286,
      "num_input_tokens_seen": 7310520,
      "step": 13185
    },
    {
      "epoch": 231.4070796460177,
      "grad_norm": 8.8125,
      "learning_rate": 3.7744621335358696e-05,
      "loss": 1.0051,
      "num_input_tokens_seen": 7313208,
      "step": 13190
    },
    {
      "epoch": 231.49557522123894,
      "grad_norm": 8.0625,
      "learning_rate": 3.7736174358481e-05,
      "loss": 1.2717,
      "num_input_tokens_seen": 7315544,
      "step": 13195
    },
    {
      "epoch": 231.58407079646017,
      "grad_norm": 6.96875,
      "learning_rate": 3.7727725417525175e-05,
      "loss": 1.1051,
      "num_input_tokens_seen": 7318456,
      "step": 13200
    },
    {
      "epoch": 231.58407079646017,
      "eval_loss": 1.0326834917068481,
      "eval_runtime": 0.934,
      "eval_samples_per_second": 26.766,
      "eval_steps_per_second": 13.918,
      "num_input_tokens_seen": 7318456,
      "step": 13200
    },
    {
      "epoch": 231.67256637168143,
      "grad_norm": 10.125,
      "learning_rate": 3.771927451379414e-05,
      "loss": 0.9772,
      "num_input_tokens_seen": 7321272,
      "step": 13205
    },
    {
      "epoch": 231.76106194690266,
      "grad_norm": 7.75,
      "learning_rate": 3.7710821648591135e-05,
      "loss": 0.9328,
      "num_input_tokens_seen": 7324376,
      "step": 13210
    },
    {
      "epoch": 231.8495575221239,
      "grad_norm": 8.3125,
      "learning_rate": 3.7702366823219694e-05,
      "loss": 0.8839,
      "num_input_tokens_seen": 7327848,
      "step": 13215
    },
    {
      "epoch": 231.93805309734512,
      "grad_norm": 8.4375,
      "learning_rate": 3.769391003898366e-05,
      "loss": 1.0763,
      "num_input_tokens_seen": 7330472,
      "step": 13220
    },
    {
      "epoch": 232.01769911504425,
      "grad_norm": 8.25,
      "learning_rate": 3.768545129718718e-05,
      "loss": 1.5366,
      "num_input_tokens_seen": 7332696,
      "step": 13225
    },
    {
      "epoch": 232.10619469026548,
      "grad_norm": 9.75,
      "learning_rate": 3.7676990599134686e-05,
      "loss": 1.1693,
      "num_input_tokens_seen": 7335256,
      "step": 13230
    },
    {
      "epoch": 232.1946902654867,
      "grad_norm": 7.78125,
      "learning_rate": 3.766852794613095e-05,
      "loss": 0.9565,
      "num_input_tokens_seen": 7338152,
      "step": 13235
    },
    {
      "epoch": 232.28318584070797,
      "grad_norm": 8.75,
      "learning_rate": 3.766006333948099e-05,
      "loss": 1.0365,
      "num_input_tokens_seen": 7341224,
      "step": 13240
    },
    {
      "epoch": 232.3716814159292,
      "grad_norm": 7.375,
      "learning_rate": 3.765159678049017e-05,
      "loss": 0.9164,
      "num_input_tokens_seen": 7343944,
      "step": 13245
    },
    {
      "epoch": 232.46017699115043,
      "grad_norm": 9.375,
      "learning_rate": 3.7643128270464134e-05,
      "loss": 1.1403,
      "num_input_tokens_seen": 7347080,
      "step": 13250
    },
    {
      "epoch": 232.5486725663717,
      "grad_norm": 6.625,
      "learning_rate": 3.763465781070884e-05,
      "loss": 0.9964,
      "num_input_tokens_seen": 7350184,
      "step": 13255
    },
    {
      "epoch": 232.63716814159292,
      "grad_norm": 8.75,
      "learning_rate": 3.762618540253052e-05,
      "loss": 1.0908,
      "num_input_tokens_seen": 7352824,
      "step": 13260
    },
    {
      "epoch": 232.72566371681415,
      "grad_norm": 9.0,
      "learning_rate": 3.761771104723576e-05,
      "loss": 1.0914,
      "num_input_tokens_seen": 7355704,
      "step": 13265
    },
    {
      "epoch": 232.81415929203538,
      "grad_norm": 7.1875,
      "learning_rate": 3.7609234746131386e-05,
      "loss": 1.1079,
      "num_input_tokens_seen": 7358056,
      "step": 13270
    },
    {
      "epoch": 232.90265486725664,
      "grad_norm": 7.21875,
      "learning_rate": 3.7600756500524556e-05,
      "loss": 0.9331,
      "num_input_tokens_seen": 7361064,
      "step": 13275
    },
    {
      "epoch": 232.99115044247787,
      "grad_norm": 7.28125,
      "learning_rate": 3.759227631172271e-05,
      "loss": 1.1192,
      "num_input_tokens_seen": 7363832,
      "step": 13280
    },
    {
      "epoch": 233.07079646017698,
      "grad_norm": 11.1875,
      "learning_rate": 3.758379418103363e-05,
      "loss": 1.0388,
      "num_input_tokens_seen": 7365880,
      "step": 13285
    },
    {
      "epoch": 233.15929203539824,
      "grad_norm": 8.5,
      "learning_rate": 3.757531010976534e-05,
      "loss": 1.036,
      "num_input_tokens_seen": 7369144,
      "step": 13290
    },
    {
      "epoch": 233.24778761061947,
      "grad_norm": 8.0,
      "learning_rate": 3.75668240992262e-05,
      "loss": 0.8878,
      "num_input_tokens_seen": 7372360,
      "step": 13295
    },
    {
      "epoch": 233.3362831858407,
      "grad_norm": 7.4375,
      "learning_rate": 3.7558336150724865e-05,
      "loss": 0.8333,
      "num_input_tokens_seen": 7375352,
      "step": 13300
    },
    {
      "epoch": 233.42477876106196,
      "grad_norm": 8.875,
      "learning_rate": 3.754984626557028e-05,
      "loss": 1.1722,
      "num_input_tokens_seen": 7377864,
      "step": 13305
    },
    {
      "epoch": 233.5132743362832,
      "grad_norm": 7.28125,
      "learning_rate": 3.754135444507168e-05,
      "loss": 0.9715,
      "num_input_tokens_seen": 7380152,
      "step": 13310
    },
    {
      "epoch": 233.60176991150442,
      "grad_norm": 10.625,
      "learning_rate": 3.753286069053863e-05,
      "loss": 1.2223,
      "num_input_tokens_seen": 7382888,
      "step": 13315
    },
    {
      "epoch": 233.69026548672565,
      "grad_norm": 7.53125,
      "learning_rate": 3.7524365003280945e-05,
      "loss": 1.0037,
      "num_input_tokens_seen": 7385992,
      "step": 13320
    },
    {
      "epoch": 233.7787610619469,
      "grad_norm": 8.4375,
      "learning_rate": 3.75158673846088e-05,
      "loss": 1.1318,
      "num_input_tokens_seen": 7388824,
      "step": 13325
    },
    {
      "epoch": 233.86725663716814,
      "grad_norm": 9.875,
      "learning_rate": 3.750736783583262e-05,
      "loss": 1.0592,
      "num_input_tokens_seen": 7391448,
      "step": 13330
    },
    {
      "epoch": 233.95575221238937,
      "grad_norm": 9.6875,
      "learning_rate": 3.7498866358263144e-05,
      "loss": 1.1409,
      "num_input_tokens_seen": 7394024,
      "step": 13335
    },
    {
      "epoch": 234.0353982300885,
      "grad_norm": 7.65625,
      "learning_rate": 3.74903629532114e-05,
      "loss": 0.9175,
      "num_input_tokens_seen": 7396328,
      "step": 13340
    },
    {
      "epoch": 234.12389380530973,
      "grad_norm": 8.9375,
      "learning_rate": 3.748185762198873e-05,
      "loss": 1.0034,
      "num_input_tokens_seen": 7399336,
      "step": 13345
    },
    {
      "epoch": 234.21238938053096,
      "grad_norm": 7.78125,
      "learning_rate": 3.747335036590676e-05,
      "loss": 0.9346,
      "num_input_tokens_seen": 7402120,
      "step": 13350
    },
    {
      "epoch": 234.30088495575222,
      "grad_norm": 7.75,
      "learning_rate": 3.7464841186277405e-05,
      "loss": 0.9544,
      "num_input_tokens_seen": 7404616,
      "step": 13355
    },
    {
      "epoch": 234.38938053097345,
      "grad_norm": 6.65625,
      "learning_rate": 3.7456330084412896e-05,
      "loss": 1.1117,
      "num_input_tokens_seen": 7407288,
      "step": 13360
    },
    {
      "epoch": 234.47787610619469,
      "grad_norm": 10.0,
      "learning_rate": 3.744781706162576e-05,
      "loss": 1.1579,
      "num_input_tokens_seen": 7410456,
      "step": 13365
    },
    {
      "epoch": 234.56637168141592,
      "grad_norm": 6.625,
      "learning_rate": 3.743930211922879e-05,
      "loss": 1.0472,
      "num_input_tokens_seen": 7413448,
      "step": 13370
    },
    {
      "epoch": 234.65486725663717,
      "grad_norm": 9.25,
      "learning_rate": 3.743078525853513e-05,
      "loss": 0.9783,
      "num_input_tokens_seen": 7416088,
      "step": 13375
    },
    {
      "epoch": 234.7433628318584,
      "grad_norm": 8.125,
      "learning_rate": 3.7422266480858154e-05,
      "loss": 1.0603,
      "num_input_tokens_seen": 7419240,
      "step": 13380
    },
    {
      "epoch": 234.83185840707964,
      "grad_norm": 9.0625,
      "learning_rate": 3.741374578751158e-05,
      "loss": 1.2413,
      "num_input_tokens_seen": 7421720,
      "step": 13385
    },
    {
      "epoch": 234.9203539823009,
      "grad_norm": 12.9375,
      "learning_rate": 3.740522317980941e-05,
      "loss": 1.0527,
      "num_input_tokens_seen": 7424632,
      "step": 13390
    },
    {
      "epoch": 235.0,
      "grad_norm": 21.75,
      "learning_rate": 3.739669865906593e-05,
      "loss": 1.2228,
      "num_input_tokens_seen": 7426880,
      "step": 13395
    },
    {
      "epoch": 235.08849557522123,
      "grad_norm": 8.5,
      "learning_rate": 3.738817222659573e-05,
      "loss": 1.0819,
      "num_input_tokens_seen": 7430160,
      "step": 13400
    },
    {
      "epoch": 235.08849557522123,
      "eval_loss": 1.026598572731018,
      "eval_runtime": 0.9334,
      "eval_samples_per_second": 26.784,
      "eval_steps_per_second": 13.928,
      "num_input_tokens_seen": 7430160,
      "step": 13400
    },
    {
      "epoch": 235.1769911504425,
      "grad_norm": 7.40625,
      "learning_rate": 3.73796438837137e-05,
      "loss": 1.0957,
      "num_input_tokens_seen": 7433104,
      "step": 13405
    },
    {
      "epoch": 235.26548672566372,
      "grad_norm": 6.9375,
      "learning_rate": 3.7371113631735e-05,
      "loss": 0.8586,
      "num_input_tokens_seen": 7436112,
      "step": 13410
    },
    {
      "epoch": 235.35398230088495,
      "grad_norm": 10.9375,
      "learning_rate": 3.736258147197512e-05,
      "loss": 1.1126,
      "num_input_tokens_seen": 7439072,
      "step": 13415
    },
    {
      "epoch": 235.44247787610618,
      "grad_norm": 6.375,
      "learning_rate": 3.735404740574981e-05,
      "loss": 0.9951,
      "num_input_tokens_seen": 7441888,
      "step": 13420
    },
    {
      "epoch": 235.53097345132744,
      "grad_norm": 11.0,
      "learning_rate": 3.7345511434375145e-05,
      "loss": 1.0723,
      "num_input_tokens_seen": 7444320,
      "step": 13425
    },
    {
      "epoch": 235.61946902654867,
      "grad_norm": 8.0625,
      "learning_rate": 3.733697355916748e-05,
      "loss": 0.9136,
      "num_input_tokens_seen": 7446688,
      "step": 13430
    },
    {
      "epoch": 235.7079646017699,
      "grad_norm": 8.75,
      "learning_rate": 3.732843378144345e-05,
      "loss": 1.1392,
      "num_input_tokens_seen": 7449312,
      "step": 13435
    },
    {
      "epoch": 235.79646017699116,
      "grad_norm": 7.0,
      "learning_rate": 3.7319892102519995e-05,
      "loss": 0.9696,
      "num_input_tokens_seen": 7452240,
      "step": 13440
    },
    {
      "epoch": 235.8849557522124,
      "grad_norm": 10.0,
      "learning_rate": 3.731134852371436e-05,
      "loss": 1.1027,
      "num_input_tokens_seen": 7455152,
      "step": 13445
    },
    {
      "epoch": 235.97345132743362,
      "grad_norm": 8.375,
      "learning_rate": 3.730280304634408e-05,
      "loss": 1.1754,
      "num_input_tokens_seen": 7458160,
      "step": 13450
    },
    {
      "epoch": 236.05309734513276,
      "grad_norm": 8.6875,
      "learning_rate": 3.729425567172696e-05,
      "loss": 1.1775,
      "num_input_tokens_seen": 7460488,
      "step": 13455
    },
    {
      "epoch": 236.141592920354,
      "grad_norm": 9.4375,
      "learning_rate": 3.728570640118111e-05,
      "loss": 1.0097,
      "num_input_tokens_seen": 7463656,
      "step": 13460
    },
    {
      "epoch": 236.23008849557522,
      "grad_norm": 8.0625,
      "learning_rate": 3.727715523602494e-05,
      "loss": 0.9736,
      "num_input_tokens_seen": 7466392,
      "step": 13465
    },
    {
      "epoch": 236.31858407079645,
      "grad_norm": 10.0,
      "learning_rate": 3.726860217757715e-05,
      "loss": 1.0069,
      "num_input_tokens_seen": 7469480,
      "step": 13470
    },
    {
      "epoch": 236.4070796460177,
      "grad_norm": 8.1875,
      "learning_rate": 3.726004722715673e-05,
      "loss": 1.1106,
      "num_input_tokens_seen": 7471816,
      "step": 13475
    },
    {
      "epoch": 236.49557522123894,
      "grad_norm": 8.75,
      "learning_rate": 3.725149038608296e-05,
      "loss": 0.9599,
      "num_input_tokens_seen": 7475080,
      "step": 13480
    },
    {
      "epoch": 236.58407079646017,
      "grad_norm": 9.375,
      "learning_rate": 3.7242931655675404e-05,
      "loss": 1.1241,
      "num_input_tokens_seen": 7477576,
      "step": 13485
    },
    {
      "epoch": 236.67256637168143,
      "grad_norm": 7.96875,
      "learning_rate": 3.7234371037253937e-05,
      "loss": 0.9912,
      "num_input_tokens_seen": 7480232,
      "step": 13490
    },
    {
      "epoch": 236.76106194690266,
      "grad_norm": 8.6875,
      "learning_rate": 3.7225808532138705e-05,
      "loss": 1.0689,
      "num_input_tokens_seen": 7483000,
      "step": 13495
    },
    {
      "epoch": 236.8495575221239,
      "grad_norm": 8.375,
      "learning_rate": 3.721724414165016e-05,
      "loss": 1.1699,
      "num_input_tokens_seen": 7485800,
      "step": 13500
    },
    {
      "epoch": 236.93805309734512,
      "grad_norm": 6.5625,
      "learning_rate": 3.720867786710904e-05,
      "loss": 1.1001,
      "num_input_tokens_seen": 7488376,
      "step": 13505
    },
    {
      "epoch": 237.01769911504425,
      "grad_norm": 8.5,
      "learning_rate": 3.7200109709836366e-05,
      "loss": 1.0101,
      "num_input_tokens_seen": 7490768,
      "step": 13510
    },
    {
      "epoch": 237.10619469026548,
      "grad_norm": 9.0625,
      "learning_rate": 3.7191539671153465e-05,
      "loss": 1.2894,
      "num_input_tokens_seen": 7493680,
      "step": 13515
    },
    {
      "epoch": 237.1946902654867,
      "grad_norm": 10.4375,
      "learning_rate": 3.718296775238193e-05,
      "loss": 1.0213,
      "num_input_tokens_seen": 7496544,
      "step": 13520
    },
    {
      "epoch": 237.28318584070797,
      "grad_norm": 13.1875,
      "learning_rate": 3.7174393954843675e-05,
      "loss": 1.1887,
      "num_input_tokens_seen": 7499488,
      "step": 13525
    },
    {
      "epoch": 237.3716814159292,
      "grad_norm": 9.3125,
      "learning_rate": 3.716581827986087e-05,
      "loss": 0.9765,
      "num_input_tokens_seen": 7502784,
      "step": 13530
    },
    {
      "epoch": 237.46017699115043,
      "grad_norm": 8.5,
      "learning_rate": 3.7157240728756004e-05,
      "loss": 0.965,
      "num_input_tokens_seen": 7505440,
      "step": 13535
    },
    {
      "epoch": 237.5486725663717,
      "grad_norm": 8.75,
      "learning_rate": 3.714866130285184e-05,
      "loss": 0.9785,
      "num_input_tokens_seen": 7507888,
      "step": 13540
    },
    {
      "epoch": 237.63716814159292,
      "grad_norm": 5.65625,
      "learning_rate": 3.714008000347143e-05,
      "loss": 0.9958,
      "num_input_tokens_seen": 7510848,
      "step": 13545
    },
    {
      "epoch": 237.72566371681415,
      "grad_norm": 6.8125,
      "learning_rate": 3.7131496831938126e-05,
      "loss": 0.9696,
      "num_input_tokens_seen": 7513696,
      "step": 13550
    },
    {
      "epoch": 237.81415929203538,
      "grad_norm": 8.625,
      "learning_rate": 3.7122911789575565e-05,
      "loss": 1.0428,
      "num_input_tokens_seen": 7516864,
      "step": 13555
    },
    {
      "epoch": 237.90265486725664,
      "grad_norm": 7.84375,
      "learning_rate": 3.711432487770765e-05,
      "loss": 1.0762,
      "num_input_tokens_seen": 7519376,
      "step": 13560
    },
    {
      "epoch": 237.99115044247787,
      "grad_norm": 8.25,
      "learning_rate": 3.710573609765861e-05,
      "loss": 1.0612,
      "num_input_tokens_seen": 7521920,
      "step": 13565
    },
    {
      "epoch": 238.07079646017698,
      "grad_norm": 7.40625,
      "learning_rate": 3.709714545075292e-05,
      "loss": 0.843,
      "num_input_tokens_seen": 7524136,
      "step": 13570
    },
    {
      "epoch": 238.15929203539824,
      "grad_norm": 7.53125,
      "learning_rate": 3.708855293831538e-05,
      "loss": 1.0334,
      "num_input_tokens_seen": 7526696,
      "step": 13575
    },
    {
      "epoch": 238.24778761061947,
      "grad_norm": 10.8125,
      "learning_rate": 3.707995856167107e-05,
      "loss": 1.2072,
      "num_input_tokens_seen": 7529496,
      "step": 13580
    },
    {
      "epoch": 238.3362831858407,
      "grad_norm": 9.5,
      "learning_rate": 3.707136232214534e-05,
      "loss": 1.0706,
      "num_input_tokens_seen": 7532568,
      "step": 13585
    },
    {
      "epoch": 238.42477876106196,
      "grad_norm": 8.25,
      "learning_rate": 3.7062764221063844e-05,
      "loss": 1.2919,
      "num_input_tokens_seen": 7535112,
      "step": 13590
    },
    {
      "epoch": 238.5132743362832,
      "grad_norm": 8.5,
      "learning_rate": 3.705416425975252e-05,
      "loss": 0.933,
      "num_input_tokens_seen": 7537656,
      "step": 13595
    },
    {
      "epoch": 238.60176991150442,
      "grad_norm": 15.1875,
      "learning_rate": 3.704556243953758e-05,
      "loss": 1.094,
      "num_input_tokens_seen": 7540344,
      "step": 13600
    },
    {
      "epoch": 238.60176991150442,
      "eval_loss": 1.0253781080245972,
      "eval_runtime": 0.9371,
      "eval_samples_per_second": 26.677,
      "eval_steps_per_second": 13.872,
      "num_input_tokens_seen": 7540344,
      "step": 13600
    },
    {
      "epoch": 238.69026548672565,
      "grad_norm": 8.375,
      "learning_rate": 3.7036958761745535e-05,
      "loss": 0.8944,
      "num_input_tokens_seen": 7543016,
      "step": 13605
    },
    {
      "epoch": 238.7787610619469,
      "grad_norm": 8.1875,
      "learning_rate": 3.702835322770318e-05,
      "loss": 1.2253,
      "num_input_tokens_seen": 7545752,
      "step": 13610
    },
    {
      "epoch": 238.86725663716814,
      "grad_norm": 4.8125,
      "learning_rate": 3.701974583873761e-05,
      "loss": 1.0402,
      "num_input_tokens_seen": 7548632,
      "step": 13615
    },
    {
      "epoch": 238.95575221238937,
      "grad_norm": 8.625,
      "learning_rate": 3.701113659617618e-05,
      "loss": 0.9423,
      "num_input_tokens_seen": 7551544,
      "step": 13620
    },
    {
      "epoch": 239.0353982300885,
      "grad_norm": 8.875,
      "learning_rate": 3.7002525501346535e-05,
      "loss": 0.8714,
      "num_input_tokens_seen": 7554152,
      "step": 13625
    },
    {
      "epoch": 239.12389380530973,
      "grad_norm": 8.4375,
      "learning_rate": 3.699391255557664e-05,
      "loss": 1.0119,
      "num_input_tokens_seen": 7556808,
      "step": 13630
    },
    {
      "epoch": 239.21238938053096,
      "grad_norm": 10.1875,
      "learning_rate": 3.69852977601947e-05,
      "loss": 1.1741,
      "num_input_tokens_seen": 7560120,
      "step": 13635
    },
    {
      "epoch": 239.30088495575222,
      "grad_norm": 7.53125,
      "learning_rate": 3.697668111652922e-05,
      "loss": 1.0954,
      "num_input_tokens_seen": 7562696,
      "step": 13640
    },
    {
      "epoch": 239.38938053097345,
      "grad_norm": 9.0,
      "learning_rate": 3.6968062625909005e-05,
      "loss": 1.1986,
      "num_input_tokens_seen": 7565608,
      "step": 13645
    },
    {
      "epoch": 239.47787610619469,
      "grad_norm": 11.4375,
      "learning_rate": 3.6959442289663135e-05,
      "loss": 1.1014,
      "num_input_tokens_seen": 7568216,
      "step": 13650
    },
    {
      "epoch": 239.56637168141592,
      "grad_norm": 7.25,
      "learning_rate": 3.695082010912098e-05,
      "loss": 1.1505,
      "num_input_tokens_seen": 7570824,
      "step": 13655
    },
    {
      "epoch": 239.65486725663717,
      "grad_norm": 10.625,
      "learning_rate": 3.694219608561217e-05,
      "loss": 0.9143,
      "num_input_tokens_seen": 7574072,
      "step": 13660
    },
    {
      "epoch": 239.7433628318584,
      "grad_norm": 8.25,
      "learning_rate": 3.693357022046665e-05,
      "loss": 0.9331,
      "num_input_tokens_seen": 7576808,
      "step": 13665
    },
    {
      "epoch": 239.83185840707964,
      "grad_norm": 7.3125,
      "learning_rate": 3.6924942515014644e-05,
      "loss": 1.1164,
      "num_input_tokens_seen": 7579512,
      "step": 13670
    },
    {
      "epoch": 239.9203539823009,
      "grad_norm": 8.75,
      "learning_rate": 3.691631297058664e-05,
      "loss": 0.893,
      "num_input_tokens_seen": 7581896,
      "step": 13675
    },
    {
      "epoch": 240.0,
      "grad_norm": 13.0,
      "learning_rate": 3.6907681588513424e-05,
      "loss": 0.9807,
      "num_input_tokens_seen": 7584296,
      "step": 13680
    },
    {
      "epoch": 240.08849557522123,
      "grad_norm": 9.8125,
      "learning_rate": 3.689904837012606e-05,
      "loss": 1.2315,
      "num_input_tokens_seen": 7587432,
      "step": 13685
    },
    {
      "epoch": 240.1769911504425,
      "grad_norm": 7.75,
      "learning_rate": 3.689041331675591e-05,
      "loss": 0.9253,
      "num_input_tokens_seen": 7589928,
      "step": 13690
    },
    {
      "epoch": 240.26548672566372,
      "grad_norm": 7.25,
      "learning_rate": 3.688177642973461e-05,
      "loss": 1.1806,
      "num_input_tokens_seen": 7592664,
      "step": 13695
    },
    {
      "epoch": 240.35398230088495,
      "grad_norm": 11.125,
      "learning_rate": 3.687313771039406e-05,
      "loss": 0.9531,
      "num_input_tokens_seen": 7595416,
      "step": 13700
    },
    {
      "epoch": 240.44247787610618,
      "grad_norm": 8.9375,
      "learning_rate": 3.686449716006647e-05,
      "loss": 0.9585,
      "num_input_tokens_seen": 7598568,
      "step": 13705
    },
    {
      "epoch": 240.53097345132744,
      "grad_norm": 7.28125,
      "learning_rate": 3.685585478008432e-05,
      "loss": 1.0352,
      "num_input_tokens_seen": 7601592,
      "step": 13710
    },
    {
      "epoch": 240.61946902654867,
      "grad_norm": 8.1875,
      "learning_rate": 3.6847210571780364e-05,
      "loss": 0.926,
      "num_input_tokens_seen": 7604344,
      "step": 13715
    },
    {
      "epoch": 240.7079646017699,
      "grad_norm": 7.1875,
      "learning_rate": 3.683856453648767e-05,
      "loss": 0.9835,
      "num_input_tokens_seen": 7607144,
      "step": 13720
    },
    {
      "epoch": 240.79646017699116,
      "grad_norm": 8.125,
      "learning_rate": 3.682991667553954e-05,
      "loss": 1.0265,
      "num_input_tokens_seen": 7609960,
      "step": 13725
    },
    {
      "epoch": 240.8849557522124,
      "grad_norm": 8.3125,
      "learning_rate": 3.6821266990269606e-05,
      "loss": 1.2586,
      "num_input_tokens_seen": 7612584,
      "step": 13730
    },
    {
      "epoch": 240.97345132743362,
      "grad_norm": 10.625,
      "learning_rate": 3.681261548201174e-05,
      "loss": 1.0498,
      "num_input_tokens_seen": 7615368,
      "step": 13735
    },
    {
      "epoch": 241.05309734513276,
      "grad_norm": 8.9375,
      "learning_rate": 3.6803962152100125e-05,
      "loss": 1.0849,
      "num_input_tokens_seen": 7617512,
      "step": 13740
    },
    {
      "epoch": 241.141592920354,
      "grad_norm": 9.125,
      "learning_rate": 3.67953070018692e-05,
      "loss": 1.0517,
      "num_input_tokens_seen": 7620024,
      "step": 13745
    },
    {
      "epoch": 241.23008849557522,
      "grad_norm": 5.78125,
      "learning_rate": 3.678665003265371e-05,
      "loss": 0.9722,
      "num_input_tokens_seen": 7622552,
      "step": 13750
    },
    {
      "epoch": 241.31858407079645,
      "grad_norm": 11.0,
      "learning_rate": 3.677799124578867e-05,
      "loss": 1.095,
      "num_input_tokens_seen": 7625784,
      "step": 13755
    },
    {
      "epoch": 241.4070796460177,
      "grad_norm": 8.5,
      "learning_rate": 3.676933064260937e-05,
      "loss": 1.2597,
      "num_input_tokens_seen": 7628600,
      "step": 13760
    },
    {
      "epoch": 241.49557522123894,
      "grad_norm": 8.375,
      "learning_rate": 3.6760668224451365e-05,
      "loss": 1.0129,
      "num_input_tokens_seen": 7632152,
      "step": 13765
    },
    {
      "epoch": 241.58407079646017,
      "grad_norm": 12.75,
      "learning_rate": 3.675200399265054e-05,
      "loss": 0.837,
      "num_input_tokens_seen": 7634712,
      "step": 13770
    },
    {
      "epoch": 241.67256637168143,
      "grad_norm": 11.875,
      "learning_rate": 3.6743337948543014e-05,
      "loss": 1.1825,
      "num_input_tokens_seen": 7637736,
      "step": 13775
    },
    {
      "epoch": 241.76106194690266,
      "grad_norm": 8.0625,
      "learning_rate": 3.6734670093465204e-05,
      "loss": 1.0239,
      "num_input_tokens_seen": 7640648,
      "step": 13780
    },
    {
      "epoch": 241.8495575221239,
      "grad_norm": 6.5625,
      "learning_rate": 3.672600042875379e-05,
      "loss": 1.149,
      "num_input_tokens_seen": 7643048,
      "step": 13785
    },
    {
      "epoch": 241.93805309734512,
      "grad_norm": 7.625,
      "learning_rate": 3.671732895574575e-05,
      "loss": 1.0207,
      "num_input_tokens_seen": 7645928,
      "step": 13790
    },
    {
      "epoch": 242.01769911504425,
      "grad_norm": 9.625,
      "learning_rate": 3.670865567577834e-05,
      "loss": 1.2699,
      "num_input_tokens_seen": 7648232,
      "step": 13795
    },
    {
      "epoch": 242.10619469026548,
      "grad_norm": 10.0,
      "learning_rate": 3.669998059018909e-05,
      "loss": 1.0909,
      "num_input_tokens_seen": 7650824,
      "step": 13800
    },
    {
      "epoch": 242.10619469026548,
      "eval_loss": 1.0268114805221558,
      "eval_runtime": 0.9322,
      "eval_samples_per_second": 26.817,
      "eval_steps_per_second": 13.945,
      "num_input_tokens_seen": 7650824,
      "step": 13800
    },
    {
      "epoch": 242.1946902654867,
      "grad_norm": 8.375,
      "learning_rate": 3.6691303700315796e-05,
      "loss": 1.011,
      "num_input_tokens_seen": 7653640,
      "step": 13805
    },
    {
      "epoch": 242.28318584070797,
      "grad_norm": 8.25,
      "learning_rate": 3.668262500749655e-05,
      "loss": 1.0379,
      "num_input_tokens_seen": 7656520,
      "step": 13810
    },
    {
      "epoch": 242.3716814159292,
      "grad_norm": 7.53125,
      "learning_rate": 3.667394451306971e-05,
      "loss": 1.1593,
      "num_input_tokens_seen": 7659080,
      "step": 13815
    },
    {
      "epoch": 242.46017699115043,
      "grad_norm": 11.3125,
      "learning_rate": 3.666526221837393e-05,
      "loss": 0.98,
      "num_input_tokens_seen": 7661624,
      "step": 13820
    },
    {
      "epoch": 242.5486725663717,
      "grad_norm": 7.1875,
      "learning_rate": 3.665657812474812e-05,
      "loss": 1.1106,
      "num_input_tokens_seen": 7664152,
      "step": 13825
    },
    {
      "epoch": 242.63716814159292,
      "grad_norm": 7.9375,
      "learning_rate": 3.664789223353147e-05,
      "loss": 0.9555,
      "num_input_tokens_seen": 7667112,
      "step": 13830
    },
    {
      "epoch": 242.72566371681415,
      "grad_norm": 8.5625,
      "learning_rate": 3.663920454606347e-05,
      "loss": 1.0591,
      "num_input_tokens_seen": 7670312,
      "step": 13835
    },
    {
      "epoch": 242.81415929203538,
      "grad_norm": 8.0625,
      "learning_rate": 3.6630515063683856e-05,
      "loss": 1.0349,
      "num_input_tokens_seen": 7672984,
      "step": 13840
    },
    {
      "epoch": 242.90265486725664,
      "grad_norm": 11.625,
      "learning_rate": 3.662182378773267e-05,
      "loss": 0.9487,
      "num_input_tokens_seen": 7675784,
      "step": 13845
    },
    {
      "epoch": 242.99115044247787,
      "grad_norm": 5.84375,
      "learning_rate": 3.66131307195502e-05,
      "loss": 1.125,
      "num_input_tokens_seen": 7679416,
      "step": 13850
    },
    {
      "epoch": 243.07079646017698,
      "grad_norm": 9.375,
      "learning_rate": 3.6604435860477034e-05,
      "loss": 1.246,
      "num_input_tokens_seen": 7681424,
      "step": 13855
    },
    {
      "epoch": 243.15929203539824,
      "grad_norm": 11.4375,
      "learning_rate": 3.6595739211854025e-05,
      "loss": 1.1039,
      "num_input_tokens_seen": 7684064,
      "step": 13860
    },
    {
      "epoch": 243.24778761061947,
      "grad_norm": 7.90625,
      "learning_rate": 3.658704077502231e-05,
      "loss": 0.9102,
      "num_input_tokens_seen": 7686736,
      "step": 13865
    },
    {
      "epoch": 243.3362831858407,
      "grad_norm": 8.0,
      "learning_rate": 3.65783405513233e-05,
      "loss": 1.2983,
      "num_input_tokens_seen": 7689328,
      "step": 13870
    },
    {
      "epoch": 243.42477876106196,
      "grad_norm": 8.1875,
      "learning_rate": 3.656963854209867e-05,
      "loss": 1.1029,
      "num_input_tokens_seen": 7692128,
      "step": 13875
    },
    {
      "epoch": 243.5132743362832,
      "grad_norm": 7.03125,
      "learning_rate": 3.656093474869038e-05,
      "loss": 1.0284,
      "num_input_tokens_seen": 7695056,
      "step": 13880
    },
    {
      "epoch": 243.60176991150442,
      "grad_norm": 9.125,
      "learning_rate": 3.655222917244068e-05,
      "loss": 1.0037,
      "num_input_tokens_seen": 7697904,
      "step": 13885
    },
    {
      "epoch": 243.69026548672565,
      "grad_norm": 10.25,
      "learning_rate": 3.6543521814692054e-05,
      "loss": 1.0308,
      "num_input_tokens_seen": 7700960,
      "step": 13890
    },
    {
      "epoch": 243.7787610619469,
      "grad_norm": 6.375,
      "learning_rate": 3.653481267678731e-05,
      "loss": 0.9049,
      "num_input_tokens_seen": 7703920,
      "step": 13895
    },
    {
      "epoch": 243.86725663716814,
      "grad_norm": 7.4375,
      "learning_rate": 3.652610176006949e-05,
      "loss": 1.0916,
      "num_input_tokens_seen": 7706880,
      "step": 13900
    },
    {
      "epoch": 243.95575221238937,
      "grad_norm": 8.4375,
      "learning_rate": 3.6517389065881925e-05,
      "loss": 1.0318,
      "num_input_tokens_seen": 7709904,
      "step": 13905
    },
    {
      "epoch": 244.0353982300885,
      "grad_norm": 7.6875,
      "learning_rate": 3.650867459556824e-05,
      "loss": 0.9591,
      "num_input_tokens_seen": 7712288,
      "step": 13910
    },
    {
      "epoch": 244.12389380530973,
      "grad_norm": 9.5625,
      "learning_rate": 3.64999583504723e-05,
      "loss": 1.0077,
      "num_input_tokens_seen": 7714688,
      "step": 13915
    },
    {
      "epoch": 244.21238938053096,
      "grad_norm": 8.0625,
      "learning_rate": 3.649124033193827e-05,
      "loss": 0.9679,
      "num_input_tokens_seen": 7717312,
      "step": 13920
    },
    {
      "epoch": 244.30088495575222,
      "grad_norm": 7.75,
      "learning_rate": 3.648252054131057e-05,
      "loss": 1.1678,
      "num_input_tokens_seen": 7720064,
      "step": 13925
    },
    {
      "epoch": 244.38938053097345,
      "grad_norm": 9.625,
      "learning_rate": 3.647379897993391e-05,
      "loss": 1.0596,
      "num_input_tokens_seen": 7723104,
      "step": 13930
    },
    {
      "epoch": 244.47787610619469,
      "grad_norm": 8.3125,
      "learning_rate": 3.646507564915325e-05,
      "loss": 1.1122,
      "num_input_tokens_seen": 7726144,
      "step": 13935
    },
    {
      "epoch": 244.56637168141592,
      "grad_norm": 7.84375,
      "learning_rate": 3.645635055031385e-05,
      "loss": 1.0837,
      "num_input_tokens_seen": 7728848,
      "step": 13940
    },
    {
      "epoch": 244.65486725663717,
      "grad_norm": 8.625,
      "learning_rate": 3.6447623684761224e-05,
      "loss": 1.0206,
      "num_input_tokens_seen": 7731904,
      "step": 13945
    },
    {
      "epoch": 244.7433628318584,
      "grad_norm": 7.75,
      "learning_rate": 3.643889505384117e-05,
      "loss": 1.099,
      "num_input_tokens_seen": 7734640,
      "step": 13950
    },
    {
      "epoch": 244.83185840707964,
      "grad_norm": 6.9375,
      "learning_rate": 3.6430164658899744e-05,
      "loss": 1.0228,
      "num_input_tokens_seen": 7737280,
      "step": 13955
    },
    {
      "epoch": 244.9203539823009,
      "grad_norm": 8.4375,
      "learning_rate": 3.642143250128329e-05,
      "loss": 0.9748,
      "num_input_tokens_seen": 7739984,
      "step": 13960
    },
    {
      "epoch": 245.0,
      "grad_norm": 9.4375,
      "learning_rate": 3.641269858233841e-05,
      "loss": 1.2243,
      "num_input_tokens_seen": 7742192,
      "step": 13965
    },
    {
      "epoch": 245.08849557522123,
      "grad_norm": 6.625,
      "learning_rate": 3.640396290341199e-05,
      "loss": 1.0124,
      "num_input_tokens_seen": 7745280,
      "step": 13970
    },
    {
      "epoch": 245.1769911504425,
      "grad_norm": 9.4375,
      "learning_rate": 3.639522546585118e-05,
      "loss": 0.9964,
      "num_input_tokens_seen": 7748128,
      "step": 13975
    },
    {
      "epoch": 245.26548672566372,
      "grad_norm": 7.0625,
      "learning_rate": 3.6386486271003404e-05,
      "loss": 0.894,
      "num_input_tokens_seen": 7751104,
      "step": 13980
    },
    {
      "epoch": 245.35398230088495,
      "grad_norm": 7.75,
      "learning_rate": 3.6377745320216346e-05,
      "loss": 1.0209,
      "num_input_tokens_seen": 7754224,
      "step": 13985
    },
    {
      "epoch": 245.44247787610618,
      "grad_norm": 7.75,
      "learning_rate": 3.636900261483798e-05,
      "loss": 1.1004,
      "num_input_tokens_seen": 7756912,
      "step": 13990
    },
    {
      "epoch": 245.53097345132744,
      "grad_norm": 8.0,
      "learning_rate": 3.636025815621654e-05,
      "loss": 1.3321,
      "num_input_tokens_seen": 7759440,
      "step": 13995
    },
    {
      "epoch": 245.61946902654867,
      "grad_norm": 6.125,
      "learning_rate": 3.635151194570054e-05,
      "loss": 1.1066,
      "num_input_tokens_seen": 7761968,
      "step": 14000
    },
    {
      "epoch": 245.61946902654867,
      "eval_loss": 1.0255075693130493,
      "eval_runtime": 0.9353,
      "eval_samples_per_second": 26.728,
      "eval_steps_per_second": 13.899,
      "num_input_tokens_seen": 7761968,
      "step": 14000
    },
    {
      "epoch": 245.7079646017699,
      "grad_norm": 7.21875,
      "learning_rate": 3.634276398463873e-05,
      "loss": 1.0345,
      "num_input_tokens_seen": 7764896,
      "step": 14005
    },
    {
      "epoch": 245.79646017699116,
      "grad_norm": 6.96875,
      "learning_rate": 3.633401427438018e-05,
      "loss": 1.0486,
      "num_input_tokens_seen": 7767424,
      "step": 14010
    },
    {
      "epoch": 245.8849557522124,
      "grad_norm": 8.8125,
      "learning_rate": 3.63252628162742e-05,
      "loss": 1.0072,
      "num_input_tokens_seen": 7769744,
      "step": 14015
    },
    {
      "epoch": 245.97345132743362,
      "grad_norm": 7.0,
      "learning_rate": 3.6316509611670364e-05,
      "loss": 1.0666,
      "num_input_tokens_seen": 7772880,
      "step": 14020
    },
    {
      "epoch": 246.05309734513276,
      "grad_norm": 8.6875,
      "learning_rate": 3.630775466191854e-05,
      "loss": 1.0495,
      "num_input_tokens_seen": 7775344,
      "step": 14025
    },
    {
      "epoch": 246.141592920354,
      "grad_norm": 7.90625,
      "learning_rate": 3.629899796836884e-05,
      "loss": 0.9377,
      "num_input_tokens_seen": 7778176,
      "step": 14030
    },
    {
      "epoch": 246.23008849557522,
      "grad_norm": 9.6875,
      "learning_rate": 3.6290239532371666e-05,
      "loss": 1.1872,
      "num_input_tokens_seen": 7781040,
      "step": 14035
    },
    {
      "epoch": 246.31858407079645,
      "grad_norm": 8.1875,
      "learning_rate": 3.628147935527767e-05,
      "loss": 0.98,
      "num_input_tokens_seen": 7784112,
      "step": 14040
    },
    {
      "epoch": 246.4070796460177,
      "grad_norm": 9.3125,
      "learning_rate": 3.627271743843779e-05,
      "loss": 1.1237,
      "num_input_tokens_seen": 7786896,
      "step": 14045
    },
    {
      "epoch": 246.49557522123894,
      "grad_norm": 7.375,
      "learning_rate": 3.626395378320321e-05,
      "loss": 1.0191,
      "num_input_tokens_seen": 7789264,
      "step": 14050
    },
    {
      "epoch": 246.58407079646017,
      "grad_norm": 8.625,
      "learning_rate": 3.625518839092541e-05,
      "loss": 1.3032,
      "num_input_tokens_seen": 7792048,
      "step": 14055
    },
    {
      "epoch": 246.67256637168143,
      "grad_norm": 8.4375,
      "learning_rate": 3.624642126295612e-05,
      "loss": 0.9972,
      "num_input_tokens_seen": 7795136,
      "step": 14060
    },
    {
      "epoch": 246.76106194690266,
      "grad_norm": 8.875,
      "learning_rate": 3.6237652400647345e-05,
      "loss": 1.0545,
      "num_input_tokens_seen": 7798240,
      "step": 14065
    },
    {
      "epoch": 246.8495575221239,
      "grad_norm": 11.75,
      "learning_rate": 3.622888180535134e-05,
      "loss": 1.0529,
      "num_input_tokens_seen": 7800784,
      "step": 14070
    },
    {
      "epoch": 246.93805309734512,
      "grad_norm": 9.1875,
      "learning_rate": 3.6220109478420655e-05,
      "loss": 0.9907,
      "num_input_tokens_seen": 7803664,
      "step": 14075
    },
    {
      "epoch": 247.01769911504425,
      "grad_norm": 8.1875,
      "learning_rate": 3.6211335421208084e-05,
      "loss": 0.9622,
      "num_input_tokens_seen": 7805936,
      "step": 14080
    },
    {
      "epoch": 247.10619469026548,
      "grad_norm": 8.0,
      "learning_rate": 3.62025596350667e-05,
      "loss": 1.0769,
      "num_input_tokens_seen": 7808912,
      "step": 14085
    },
    {
      "epoch": 247.1946902654867,
      "grad_norm": 8.3125,
      "learning_rate": 3.619378212134984e-05,
      "loss": 1.1883,
      "num_input_tokens_seen": 7811584,
      "step": 14090
    },
    {
      "epoch": 247.28318584070797,
      "grad_norm": 11.125,
      "learning_rate": 3.618500288141111e-05,
      "loss": 1.0518,
      "num_input_tokens_seen": 7814112,
      "step": 14095
    },
    {
      "epoch": 247.3716814159292,
      "grad_norm": 7.0625,
      "learning_rate": 3.617622191660438e-05,
      "loss": 0.805,
      "num_input_tokens_seen": 7816704,
      "step": 14100
    },
    {
      "epoch": 247.46017699115043,
      "grad_norm": 7.40625,
      "learning_rate": 3.616743922828377e-05,
      "loss": 0.9946,
      "num_input_tokens_seen": 7819744,
      "step": 14105
    },
    {
      "epoch": 247.5486725663717,
      "grad_norm": 7.53125,
      "learning_rate": 3.615865481780371e-05,
      "loss": 1.1527,
      "num_input_tokens_seen": 7822096,
      "step": 14110
    },
    {
      "epoch": 247.63716814159292,
      "grad_norm": 7.09375,
      "learning_rate": 3.614986868651883e-05,
      "loss": 0.9943,
      "num_input_tokens_seen": 7825584,
      "step": 14115
    },
    {
      "epoch": 247.72566371681415,
      "grad_norm": 8.25,
      "learning_rate": 3.614108083578409e-05,
      "loss": 0.9374,
      "num_input_tokens_seen": 7828416,
      "step": 14120
    },
    {
      "epoch": 247.81415929203538,
      "grad_norm": 7.4375,
      "learning_rate": 3.613229126695467e-05,
      "loss": 1.1244,
      "num_input_tokens_seen": 7831152,
      "step": 14125
    },
    {
      "epoch": 247.90265486725664,
      "grad_norm": 9.4375,
      "learning_rate": 3.612349998138605e-05,
      "loss": 1.1923,
      "num_input_tokens_seen": 7833744,
      "step": 14130
    },
    {
      "epoch": 247.99115044247787,
      "grad_norm": 12.8125,
      "learning_rate": 3.6114706980433946e-05,
      "loss": 1.0653,
      "num_input_tokens_seen": 7836736,
      "step": 14135
    },
    {
      "epoch": 248.07079646017698,
      "grad_norm": 7.78125,
      "learning_rate": 3.610591226545435e-05,
      "loss": 1.2335,
      "num_input_tokens_seen": 7839336,
      "step": 14140
    },
    {
      "epoch": 248.15929203539824,
      "grad_norm": 8.0625,
      "learning_rate": 3.6097115837803505e-05,
      "loss": 0.789,
      "num_input_tokens_seen": 7841752,
      "step": 14145
    },
    {
      "epoch": 248.24778761061947,
      "grad_norm": 9.0625,
      "learning_rate": 3.608831769883795e-05,
      "loss": 1.2398,
      "num_input_tokens_seen": 7845096,
      "step": 14150
    },
    {
      "epoch": 248.3362831858407,
      "grad_norm": 7.96875,
      "learning_rate": 3.607951784991446e-05,
      "loss": 1.155,
      "num_input_tokens_seen": 7847896,
      "step": 14155
    },
    {
      "epoch": 248.42477876106196,
      "grad_norm": 7.8125,
      "learning_rate": 3.6070716292390085e-05,
      "loss": 1.1467,
      "num_input_tokens_seen": 7850392,
      "step": 14160
    },
    {
      "epoch": 248.5132743362832,
      "grad_norm": 7.6875,
      "learning_rate": 3.606191302762213e-05,
      "loss": 0.8922,
      "num_input_tokens_seen": 7853496,
      "step": 14165
    },
    {
      "epoch": 248.60176991150442,
      "grad_norm": 9.5625,
      "learning_rate": 3.605310805696818e-05,
      "loss": 0.9655,
      "num_input_tokens_seen": 7855928,
      "step": 14170
    },
    {
      "epoch": 248.69026548672565,
      "grad_norm": 7.65625,
      "learning_rate": 3.6044301381786067e-05,
      "loss": 1.0734,
      "num_input_tokens_seen": 7858888,
      "step": 14175
    },
    {
      "epoch": 248.7787610619469,
      "grad_norm": 8.5625,
      "learning_rate": 3.6035493003433883e-05,
      "loss": 1.2682,
      "num_input_tokens_seen": 7861464,
      "step": 14180
    },
    {
      "epoch": 248.86725663716814,
      "grad_norm": 8.3125,
      "learning_rate": 3.6026682923269994e-05,
      "loss": 0.9728,
      "num_input_tokens_seen": 7864504,
      "step": 14185
    },
    {
      "epoch": 248.95575221238937,
      "grad_norm": 8.5,
      "learning_rate": 3.6017871142653034e-05,
      "loss": 1.0935,
      "num_input_tokens_seen": 7867144,
      "step": 14190
    },
    {
      "epoch": 249.0353982300885,
      "grad_norm": 7.1875,
      "learning_rate": 3.600905766294189e-05,
      "loss": 0.9182,
      "num_input_tokens_seen": 7869960,
      "step": 14195
    },
    {
      "epoch": 249.12389380530973,
      "grad_norm": 6.09375,
      "learning_rate": 3.60002424854957e-05,
      "loss": 0.8353,
      "num_input_tokens_seen": 7872968,
      "step": 14200
    },
    {
      "epoch": 249.12389380530973,
      "eval_loss": 1.0351446866989136,
      "eval_runtime": 0.9357,
      "eval_samples_per_second": 26.719,
      "eval_steps_per_second": 13.894,
      "num_input_tokens_seen": 7872968,
      "step": 14200
    },
    {
      "epoch": 249.21238938053096,
      "grad_norm": 9.3125,
      "learning_rate": 3.5991425611673876e-05,
      "loss": 1.07,
      "num_input_tokens_seen": 7875416,
      "step": 14205
    },
    {
      "epoch": 249.30088495575222,
      "grad_norm": 10.8125,
      "learning_rate": 3.5982607042836105e-05,
      "loss": 1.1574,
      "num_input_tokens_seen": 7878216,
      "step": 14210
    },
    {
      "epoch": 249.38938053097345,
      "grad_norm": 8.1875,
      "learning_rate": 3.597378678034231e-05,
      "loss": 1.1969,
      "num_input_tokens_seen": 7880792,
      "step": 14215
    },
    {
      "epoch": 249.47787610619469,
      "grad_norm": 8.0625,
      "learning_rate": 3.596496482555269e-05,
      "loss": 0.9485,
      "num_input_tokens_seen": 7883320,
      "step": 14220
    },
    {
      "epoch": 249.56637168141592,
      "grad_norm": 9.25,
      "learning_rate": 3.595614117982769e-05,
      "loss": 1.0041,
      "num_input_tokens_seen": 7886296,
      "step": 14225
    },
    {
      "epoch": 249.65486725663717,
      "grad_norm": 9.25,
      "learning_rate": 3.594731584452805e-05,
      "loss": 0.9668,
      "num_input_tokens_seen": 7889224,
      "step": 14230
    },
    {
      "epoch": 249.7433628318584,
      "grad_norm": 6.6875,
      "learning_rate": 3.593848882101472e-05,
      "loss": 1.0789,
      "num_input_tokens_seen": 7892360,
      "step": 14235
    },
    {
      "epoch": 249.83185840707964,
      "grad_norm": 11.3125,
      "learning_rate": 3.592966011064896e-05,
      "loss": 1.3576,
      "num_input_tokens_seen": 7894776,
      "step": 14240
    },
    {
      "epoch": 249.9203539823009,
      "grad_norm": 8.375,
      "learning_rate": 3.592082971479226e-05,
      "loss": 1.0848,
      "num_input_tokens_seen": 7897576,
      "step": 14245
    },
    {
      "epoch": 250.0,
      "grad_norm": 8.875,
      "learning_rate": 3.5911997634806385e-05,
      "loss": 1.0199,
      "num_input_tokens_seen": 7899888,
      "step": 14250
    },
    {
      "epoch": 250.08849557522123,
      "grad_norm": 7.9375,
      "learning_rate": 3.5903163872053336e-05,
      "loss": 1.2544,
      "num_input_tokens_seen": 7902352,
      "step": 14255
    },
    {
      "epoch": 250.1769911504425,
      "grad_norm": 6.71875,
      "learning_rate": 3.58943284278954e-05,
      "loss": 0.932,
      "num_input_tokens_seen": 7905120,
      "step": 14260
    },
    {
      "epoch": 250.26548672566372,
      "grad_norm": 8.25,
      "learning_rate": 3.588549130369512e-05,
      "loss": 1.0678,
      "num_input_tokens_seen": 7908432,
      "step": 14265
    },
    {
      "epoch": 250.35398230088495,
      "grad_norm": 7.6875,
      "learning_rate": 3.5876652500815274e-05,
      "loss": 0.9534,
      "num_input_tokens_seen": 7911344,
      "step": 14270
    },
    {
      "epoch": 250.44247787610618,
      "grad_norm": 7.6875,
      "learning_rate": 3.586781202061894e-05,
      "loss": 1.111,
      "num_input_tokens_seen": 7914048,
      "step": 14275
    },
    {
      "epoch": 250.53097345132744,
      "grad_norm": 8.3125,
      "learning_rate": 3.585896986446942e-05,
      "loss": 0.9775,
      "num_input_tokens_seen": 7916592,
      "step": 14280
    },
    {
      "epoch": 250.61946902654867,
      "grad_norm": 9.9375,
      "learning_rate": 3.585012603373028e-05,
      "loss": 1.2279,
      "num_input_tokens_seen": 7919232,
      "step": 14285
    },
    {
      "epoch": 250.7079646017699,
      "grad_norm": 8.3125,
      "learning_rate": 3.584128052976535e-05,
      "loss": 0.7976,
      "num_input_tokens_seen": 7922208,
      "step": 14290
    },
    {
      "epoch": 250.79646017699116,
      "grad_norm": 10.25,
      "learning_rate": 3.5832433353938724e-05,
      "loss": 1.048,
      "num_input_tokens_seen": 7924944,
      "step": 14295
    },
    {
      "epoch": 250.8849557522124,
      "grad_norm": 8.4375,
      "learning_rate": 3.5823584507614746e-05,
      "loss": 1.0794,
      "num_input_tokens_seen": 7927856,
      "step": 14300
    },
    {
      "epoch": 250.97345132743362,
      "grad_norm": 7.875,
      "learning_rate": 3.581473399215802e-05,
      "loss": 1.1408,
      "num_input_tokens_seen": 7930464,
      "step": 14305
    },
    {
      "epoch": 251.05309734513276,
      "grad_norm": 8.3125,
      "learning_rate": 3.580588180893341e-05,
      "loss": 1.0118,
      "num_input_tokens_seen": 7932832,
      "step": 14310
    },
    {
      "epoch": 251.141592920354,
      "grad_norm": 10.3125,
      "learning_rate": 3.579702795930602e-05,
      "loss": 1.0515,
      "num_input_tokens_seen": 7935760,
      "step": 14315
    },
    {
      "epoch": 251.23008849557522,
      "grad_norm": 8.6875,
      "learning_rate": 3.578817244464125e-05,
      "loss": 1.0928,
      "num_input_tokens_seen": 7938304,
      "step": 14320
    },
    {
      "epoch": 251.31858407079645,
      "grad_norm": 11.4375,
      "learning_rate": 3.577931526630471e-05,
      "loss": 0.9547,
      "num_input_tokens_seen": 7941408,
      "step": 14325
    },
    {
      "epoch": 251.4070796460177,
      "grad_norm": 7.90625,
      "learning_rate": 3.577045642566229e-05,
      "loss": 1.2576,
      "num_input_tokens_seen": 7943984,
      "step": 14330
    },
    {
      "epoch": 251.49557522123894,
      "grad_norm": 7.5625,
      "learning_rate": 3.576159592408014e-05,
      "loss": 0.8286,
      "num_input_tokens_seen": 7946848,
      "step": 14335
    },
    {
      "epoch": 251.58407079646017,
      "grad_norm": 8.5,
      "learning_rate": 3.575273376292466e-05,
      "loss": 1.1266,
      "num_input_tokens_seen": 7949792,
      "step": 14340
    },
    {
      "epoch": 251.67256637168143,
      "grad_norm": 6.875,
      "learning_rate": 3.574386994356251e-05,
      "loss": 0.9269,
      "num_input_tokens_seen": 7952672,
      "step": 14345
    },
    {
      "epoch": 251.76106194690266,
      "grad_norm": 10.25,
      "learning_rate": 3.573500446736059e-05,
      "loss": 1.1016,
      "num_input_tokens_seen": 7955328,
      "step": 14350
    },
    {
      "epoch": 251.8495575221239,
      "grad_norm": 7.84375,
      "learning_rate": 3.5726137335686094e-05,
      "loss": 1.1187,
      "num_input_tokens_seen": 7958016,
      "step": 14355
    },
    {
      "epoch": 251.93805309734512,
      "grad_norm": 7.8125,
      "learning_rate": 3.571726854990642e-05,
      "loss": 1.0833,
      "num_input_tokens_seen": 7960864,
      "step": 14360
    },
    {
      "epoch": 252.01769911504425,
      "grad_norm": 7.0,
      "learning_rate": 3.570839811138925e-05,
      "loss": 1.0815,
      "num_input_tokens_seen": 7963640,
      "step": 14365
    },
    {
      "epoch": 252.10619469026548,
      "grad_norm": 8.375,
      "learning_rate": 3.569952602150252e-05,
      "loss": 0.993,
      "num_input_tokens_seen": 7966520,
      "step": 14370
    },
    {
      "epoch": 252.1946902654867,
      "grad_norm": 8.5,
      "learning_rate": 3.569065228161442e-05,
      "loss": 0.972,
      "num_input_tokens_seen": 7969256,
      "step": 14375
    },
    {
      "epoch": 252.28318584070797,
      "grad_norm": 12.5625,
      "learning_rate": 3.5681776893093395e-05,
      "loss": 1.0188,
      "num_input_tokens_seen": 7971912,
      "step": 14380
    },
    {
      "epoch": 252.3716814159292,
      "grad_norm": 7.96875,
      "learning_rate": 3.5672899857308134e-05,
      "loss": 1.1972,
      "num_input_tokens_seen": 7974136,
      "step": 14385
    },
    {
      "epoch": 252.46017699115043,
      "grad_norm": 9.9375,
      "learning_rate": 3.566402117562759e-05,
      "loss": 1.289,
      "num_input_tokens_seen": 7977400,
      "step": 14390
    },
    {
      "epoch": 252.5486725663717,
      "grad_norm": 7.5,
      "learning_rate": 3.565514084942097e-05,
      "loss": 1.0833,
      "num_input_tokens_seen": 7980472,
      "step": 14395
    },
    {
      "epoch": 252.63716814159292,
      "grad_norm": 8.25,
      "learning_rate": 3.564625888005773e-05,
      "loss": 0.9608,
      "num_input_tokens_seen": 7983464,
      "step": 14400
    },
    {
      "epoch": 252.63716814159292,
      "eval_loss": 1.0306404829025269,
      "eval_runtime": 0.9331,
      "eval_samples_per_second": 26.791,
      "eval_steps_per_second": 13.931,
      "num_input_tokens_seen": 7983464,
      "step": 14400
    },
    {
      "epoch": 252.72566371681415,
      "grad_norm": 8.5625,
      "learning_rate": 3.563737526890759e-05,
      "loss": 0.9094,
      "num_input_tokens_seen": 7987048,
      "step": 14405
    },
    {
      "epoch": 252.81415929203538,
      "grad_norm": 8.875,
      "learning_rate": 3.562849001734049e-05,
      "loss": 1.0935,
      "num_input_tokens_seen": 7989624,
      "step": 14410
    },
    {
      "epoch": 252.90265486725664,
      "grad_norm": 7.3125,
      "learning_rate": 3.561960312672667e-05,
      "loss": 1.0428,
      "num_input_tokens_seen": 7992248,
      "step": 14415
    },
    {
      "epoch": 252.99115044247787,
      "grad_norm": 8.1875,
      "learning_rate": 3.5610714598436596e-05,
      "loss": 1.1347,
      "num_input_tokens_seen": 7994536,
      "step": 14420
    },
    {
      "epoch": 253.07079646017698,
      "grad_norm": 9.0,
      "learning_rate": 3.5601824433840986e-05,
      "loss": 1.0111,
      "num_input_tokens_seen": 7996672,
      "step": 14425
    },
    {
      "epoch": 253.15929203539824,
      "grad_norm": 8.5,
      "learning_rate": 3.559293263431082e-05,
      "loss": 1.1098,
      "num_input_tokens_seen": 7999520,
      "step": 14430
    },
    {
      "epoch": 253.24778761061947,
      "grad_norm": 9.75,
      "learning_rate": 3.558403920121732e-05,
      "loss": 1.011,
      "num_input_tokens_seen": 8001968,
      "step": 14435
    },
    {
      "epoch": 253.3362831858407,
      "grad_norm": 8.375,
      "learning_rate": 3.557514413593197e-05,
      "loss": 0.9439,
      "num_input_tokens_seen": 8004672,
      "step": 14440
    },
    {
      "epoch": 253.42477876106196,
      "grad_norm": 6.6875,
      "learning_rate": 3.55662474398265e-05,
      "loss": 1.1234,
      "num_input_tokens_seen": 8007392,
      "step": 14445
    },
    {
      "epoch": 253.5132743362832,
      "grad_norm": 7.40625,
      "learning_rate": 3.555734911427288e-05,
      "loss": 1.0517,
      "num_input_tokens_seen": 8010416,
      "step": 14450
    },
    {
      "epoch": 253.60176991150442,
      "grad_norm": 10.625,
      "learning_rate": 3.5548449160643363e-05,
      "loss": 0.8112,
      "num_input_tokens_seen": 8013488,
      "step": 14455
    },
    {
      "epoch": 253.69026548672565,
      "grad_norm": 7.15625,
      "learning_rate": 3.553954758031043e-05,
      "loss": 1.2108,
      "num_input_tokens_seen": 8015984,
      "step": 14460
    },
    {
      "epoch": 253.7787610619469,
      "grad_norm": 10.8125,
      "learning_rate": 3.5530644374646815e-05,
      "loss": 1.1003,
      "num_input_tokens_seen": 8019440,
      "step": 14465
    },
    {
      "epoch": 253.86725663716814,
      "grad_norm": 10.3125,
      "learning_rate": 3.552173954502549e-05,
      "loss": 1.0856,
      "num_input_tokens_seen": 8022304,
      "step": 14470
    },
    {
      "epoch": 253.95575221238937,
      "grad_norm": 6.84375,
      "learning_rate": 3.55128330928197e-05,
      "loss": 1.0043,
      "num_input_tokens_seen": 8024768,
      "step": 14475
    },
    {
      "epoch": 254.0353982300885,
      "grad_norm": 9.875,
      "learning_rate": 3.550392501940294e-05,
      "loss": 1.2193,
      "num_input_tokens_seen": 8027248,
      "step": 14480
    },
    {
      "epoch": 254.12389380530973,
      "grad_norm": 8.5,
      "learning_rate": 3.5495015326148945e-05,
      "loss": 0.9998,
      "num_input_tokens_seen": 8030064,
      "step": 14485
    },
    {
      "epoch": 254.21238938053096,
      "grad_norm": 11.0,
      "learning_rate": 3.548610401443169e-05,
      "loss": 1.0997,
      "num_input_tokens_seen": 8032304,
      "step": 14490
    },
    {
      "epoch": 254.30088495575222,
      "grad_norm": 5.90625,
      "learning_rate": 3.547719108562543e-05,
      "loss": 0.7768,
      "num_input_tokens_seen": 8035264,
      "step": 14495
    },
    {
      "epoch": 254.38938053097345,
      "grad_norm": 6.0625,
      "learning_rate": 3.546827654110464e-05,
      "loss": 0.9246,
      "num_input_tokens_seen": 8038896,
      "step": 14500
    },
    {
      "epoch": 254.47787610619469,
      "grad_norm": 8.0625,
      "learning_rate": 3.545936038224405e-05,
      "loss": 1.1019,
      "num_input_tokens_seen": 8041424,
      "step": 14505
    },
    {
      "epoch": 254.56637168141592,
      "grad_norm": 8.0,
      "learning_rate": 3.545044261041864e-05,
      "loss": 1.0271,
      "num_input_tokens_seen": 8044256,
      "step": 14510
    },
    {
      "epoch": 254.65486725663717,
      "grad_norm": 8.875,
      "learning_rate": 3.5441523227003657e-05,
      "loss": 1.0952,
      "num_input_tokens_seen": 8046736,
      "step": 14515
    },
    {
      "epoch": 254.7433628318584,
      "grad_norm": 8.5,
      "learning_rate": 3.543260223337459e-05,
      "loss": 1.1583,
      "num_input_tokens_seen": 8049568,
      "step": 14520
    },
    {
      "epoch": 254.83185840707964,
      "grad_norm": 8.5,
      "learning_rate": 3.542367963090714e-05,
      "loss": 1.1403,
      "num_input_tokens_seen": 8052352,
      "step": 14525
    },
    {
      "epoch": 254.9203539823009,
      "grad_norm": 9.6875,
      "learning_rate": 3.5414755420977295e-05,
      "loss": 1.0842,
      "num_input_tokens_seen": 8055536,
      "step": 14530
    },
    {
      "epoch": 255.0,
      "grad_norm": 17.25,
      "learning_rate": 3.54058296049613e-05,
      "loss": 1.1473,
      "num_input_tokens_seen": 8057952,
      "step": 14535
    },
    {
      "epoch": 255.08849557522123,
      "grad_norm": 9.125,
      "learning_rate": 3.53969021842356e-05,
      "loss": 1.0053,
      "num_input_tokens_seen": 8060608,
      "step": 14540
    },
    {
      "epoch": 255.1769911504425,
      "grad_norm": 10.0625,
      "learning_rate": 3.5387973160176926e-05,
      "loss": 1.023,
      "num_input_tokens_seen": 8063216,
      "step": 14545
    },
    {
      "epoch": 255.26548672566372,
      "grad_norm": 6.4375,
      "learning_rate": 3.537904253416224e-05,
      "loss": 1.0541,
      "num_input_tokens_seen": 8065472,
      "step": 14550
    },
    {
      "epoch": 255.35398230088495,
      "grad_norm": 8.5,
      "learning_rate": 3.537011030756878e-05,
      "loss": 1.2093,
      "num_input_tokens_seen": 8068304,
      "step": 14555
    },
    {
      "epoch": 255.44247787610618,
      "grad_norm": 8.375,
      "learning_rate": 3.536117648177399e-05,
      "loss": 1.0596,
      "num_input_tokens_seen": 8071696,
      "step": 14560
    },
    {
      "epoch": 255.53097345132744,
      "grad_norm": 8.4375,
      "learning_rate": 3.535224105815558e-05,
      "loss": 0.9657,
      "num_input_tokens_seen": 8074112,
      "step": 14565
    },
    {
      "epoch": 255.61946902654867,
      "grad_norm": 11.6875,
      "learning_rate": 3.5343304038091494e-05,
      "loss": 1.0889,
      "num_input_tokens_seen": 8077200,
      "step": 14570
    },
    {
      "epoch": 255.7079646017699,
      "grad_norm": 6.84375,
      "learning_rate": 3.5334365422959955e-05,
      "loss": 0.8068,
      "num_input_tokens_seen": 8079984,
      "step": 14575
    },
    {
      "epoch": 255.79646017699116,
      "grad_norm": 8.1875,
      "learning_rate": 3.5325425214139396e-05,
      "loss": 1.3524,
      "num_input_tokens_seen": 8082672,
      "step": 14580
    },
    {
      "epoch": 255.8849557522124,
      "grad_norm": 8.375,
      "learning_rate": 3.531648341300851e-05,
      "loss": 0.8576,
      "num_input_tokens_seen": 8085328,
      "step": 14585
    },
    {
      "epoch": 255.97345132743362,
      "grad_norm": 8.1875,
      "learning_rate": 3.530754002094623e-05,
      "loss": 1.1035,
      "num_input_tokens_seen": 8088608,
      "step": 14590
    },
    {
      "epoch": 256.05309734513276,
      "grad_norm": 10.0625,
      "learning_rate": 3.529859503933175e-05,
      "loss": 1.1801,
      "num_input_tokens_seen": 8090576,
      "step": 14595
    },
    {
      "epoch": 256.14159292035396,
      "grad_norm": 8.0625,
      "learning_rate": 3.52896484695445e-05,
      "loss": 0.9857,
      "num_input_tokens_seen": 8093616,
      "step": 14600
    },
    {
      "epoch": 256.14159292035396,
      "eval_loss": 1.0302517414093018,
      "eval_runtime": 0.9394,
      "eval_samples_per_second": 26.614,
      "eval_steps_per_second": 13.839,
      "num_input_tokens_seen": 8093616,
      "step": 14600
    },
    {
      "epoch": 256.2300884955752,
      "grad_norm": 10.75,
      "learning_rate": 3.528070031296414e-05,
      "loss": 1.2441,
      "num_input_tokens_seen": 8096320,
      "step": 14605
    },
    {
      "epoch": 256.3185840707965,
      "grad_norm": 8.4375,
      "learning_rate": 3.5271750570970605e-05,
      "loss": 0.8944,
      "num_input_tokens_seen": 8099040,
      "step": 14610
    },
    {
      "epoch": 256.4070796460177,
      "grad_norm": 7.34375,
      "learning_rate": 3.526279924494405e-05,
      "loss": 0.8587,
      "num_input_tokens_seen": 8101968,
      "step": 14615
    },
    {
      "epoch": 256.49557522123894,
      "grad_norm": 11.5,
      "learning_rate": 3.5253846336264874e-05,
      "loss": 1.2656,
      "num_input_tokens_seen": 8105024,
      "step": 14620
    },
    {
      "epoch": 256.5840707964602,
      "grad_norm": 8.875,
      "learning_rate": 3.5244891846313736e-05,
      "loss": 1.0967,
      "num_input_tokens_seen": 8107616,
      "step": 14625
    },
    {
      "epoch": 256.6725663716814,
      "grad_norm": 5.625,
      "learning_rate": 3.5235935776471527e-05,
      "loss": 0.8777,
      "num_input_tokens_seen": 8110496,
      "step": 14630
    },
    {
      "epoch": 256.76106194690266,
      "grad_norm": 7.84375,
      "learning_rate": 3.522697812811939e-05,
      "loss": 1.0061,
      "num_input_tokens_seen": 8112880,
      "step": 14635
    },
    {
      "epoch": 256.8495575221239,
      "grad_norm": 7.875,
      "learning_rate": 3.521801890263871e-05,
      "loss": 1.1241,
      "num_input_tokens_seen": 8115872,
      "step": 14640
    },
    {
      "epoch": 256.9380530973451,
      "grad_norm": 8.4375,
      "learning_rate": 3.5209058101411114e-05,
      "loss": 1.0254,
      "num_input_tokens_seen": 8119152,
      "step": 14645
    },
    {
      "epoch": 257.01769911504425,
      "grad_norm": 8.625,
      "learning_rate": 3.520009572581845e-05,
      "loss": 1.2241,
      "num_input_tokens_seen": 8121336,
      "step": 14650
    },
    {
      "epoch": 257.1061946902655,
      "grad_norm": 5.96875,
      "learning_rate": 3.519113177724285e-05,
      "loss": 1.0165,
      "num_input_tokens_seen": 8123960,
      "step": 14655
    },
    {
      "epoch": 257.1946902654867,
      "grad_norm": 8.25,
      "learning_rate": 3.5182166257066656e-05,
      "loss": 1.2603,
      "num_input_tokens_seen": 8126616,
      "step": 14660
    },
    {
      "epoch": 257.283185840708,
      "grad_norm": 8.1875,
      "learning_rate": 3.517319916667247e-05,
      "loss": 0.9509,
      "num_input_tokens_seen": 8129240,
      "step": 14665
    },
    {
      "epoch": 257.37168141592923,
      "grad_norm": 8.0625,
      "learning_rate": 3.516423050744313e-05,
      "loss": 0.7683,
      "num_input_tokens_seen": 8132328,
      "step": 14670
    },
    {
      "epoch": 257.46017699115043,
      "grad_norm": 10.875,
      "learning_rate": 3.5155260280761704e-05,
      "loss": 1.3205,
      "num_input_tokens_seen": 8135176,
      "step": 14675
    },
    {
      "epoch": 257.5486725663717,
      "grad_norm": 8.3125,
      "learning_rate": 3.514628848801154e-05,
      "loss": 1.0248,
      "num_input_tokens_seen": 8138168,
      "step": 14680
    },
    {
      "epoch": 257.6371681415929,
      "grad_norm": 11.25,
      "learning_rate": 3.5137315130576174e-05,
      "loss": 1.0845,
      "num_input_tokens_seen": 8140728,
      "step": 14685
    },
    {
      "epoch": 257.72566371681415,
      "grad_norm": 8.75,
      "learning_rate": 3.512834020983942e-05,
      "loss": 1.034,
      "num_input_tokens_seen": 8143768,
      "step": 14690
    },
    {
      "epoch": 257.8141592920354,
      "grad_norm": 6.21875,
      "learning_rate": 3.5119363727185334e-05,
      "loss": 1.0023,
      "num_input_tokens_seen": 8146600,
      "step": 14695
    },
    {
      "epoch": 257.9026548672566,
      "grad_norm": 9.75,
      "learning_rate": 3.511038568399819e-05,
      "loss": 0.9944,
      "num_input_tokens_seen": 8149336,
      "step": 14700
    },
    {
      "epoch": 257.9911504424779,
      "grad_norm": 9.75,
      "learning_rate": 3.510140608166251e-05,
      "loss": 1.0756,
      "num_input_tokens_seen": 8152648,
      "step": 14705
    },
    {
      "epoch": 258.070796460177,
      "grad_norm": 8.5625,
      "learning_rate": 3.509242492156308e-05,
      "loss": 1.1516,
      "num_input_tokens_seen": 8154576,
      "step": 14710
    },
    {
      "epoch": 258.1592920353982,
      "grad_norm": 9.125,
      "learning_rate": 3.5083442205084896e-05,
      "loss": 1.0585,
      "num_input_tokens_seen": 8157472,
      "step": 14715
    },
    {
      "epoch": 258.24778761061947,
      "grad_norm": 9.0625,
      "learning_rate": 3.507445793361321e-05,
      "loss": 1.0215,
      "num_input_tokens_seen": 8160128,
      "step": 14720
    },
    {
      "epoch": 258.3362831858407,
      "grad_norm": 7.875,
      "learning_rate": 3.5065472108533505e-05,
      "loss": 1.0307,
      "num_input_tokens_seen": 8163152,
      "step": 14725
    },
    {
      "epoch": 258.42477876106193,
      "grad_norm": 11.5625,
      "learning_rate": 3.5056484731231504e-05,
      "loss": 1.1843,
      "num_input_tokens_seen": 8165600,
      "step": 14730
    },
    {
      "epoch": 258.5132743362832,
      "grad_norm": 8.25,
      "learning_rate": 3.504749580309319e-05,
      "loss": 1.0273,
      "num_input_tokens_seen": 8169008,
      "step": 14735
    },
    {
      "epoch": 258.60176991150445,
      "grad_norm": 7.28125,
      "learning_rate": 3.5038505325504753e-05,
      "loss": 1.0435,
      "num_input_tokens_seen": 8172400,
      "step": 14740
    },
    {
      "epoch": 258.69026548672565,
      "grad_norm": 8.5,
      "learning_rate": 3.502951329985264e-05,
      "loss": 1.0935,
      "num_input_tokens_seen": 8175232,
      "step": 14745
    },
    {
      "epoch": 258.7787610619469,
      "grad_norm": 6.875,
      "learning_rate": 3.502051972752354e-05,
      "loss": 1.0136,
      "num_input_tokens_seen": 8177584,
      "step": 14750
    },
    {
      "epoch": 258.86725663716817,
      "grad_norm": 6.8125,
      "learning_rate": 3.5011524609904374e-05,
      "loss": 1.1429,
      "num_input_tokens_seen": 8180352,
      "step": 14755
    },
    {
      "epoch": 258.95575221238937,
      "grad_norm": 9.25,
      "learning_rate": 3.50025279483823e-05,
      "loss": 1.052,
      "num_input_tokens_seen": 8183232,
      "step": 14760
    },
    {
      "epoch": 259.0353982300885,
      "grad_norm": 8.375,
      "learning_rate": 3.499352974434472e-05,
      "loss": 0.912,
      "num_input_tokens_seen": 8185632,
      "step": 14765
    },
    {
      "epoch": 259.12389380530976,
      "grad_norm": 9.25,
      "learning_rate": 3.498452999917926e-05,
      "loss": 1.2234,
      "num_input_tokens_seen": 8188528,
      "step": 14770
    },
    {
      "epoch": 259.21238938053096,
      "grad_norm": 9.375,
      "learning_rate": 3.4975528714273795e-05,
      "loss": 1.0092,
      "num_input_tokens_seen": 8191264,
      "step": 14775
    },
    {
      "epoch": 259.3008849557522,
      "grad_norm": 8.4375,
      "learning_rate": 3.4966525891016454e-05,
      "loss": 1.0181,
      "num_input_tokens_seen": 8194240,
      "step": 14780
    },
    {
      "epoch": 259.3893805309734,
      "grad_norm": 7.53125,
      "learning_rate": 3.495752153079557e-05,
      "loss": 1.074,
      "num_input_tokens_seen": 8196608,
      "step": 14785
    },
    {
      "epoch": 259.4778761061947,
      "grad_norm": 7.625,
      "learning_rate": 3.494851563499974e-05,
      "loss": 1.2119,
      "num_input_tokens_seen": 8199056,
      "step": 14790
    },
    {
      "epoch": 259.56637168141594,
      "grad_norm": 8.125,
      "learning_rate": 3.493950820501777e-05,
      "loss": 0.9471,
      "num_input_tokens_seen": 8201776,
      "step": 14795
    },
    {
      "epoch": 259.65486725663715,
      "grad_norm": 8.5625,
      "learning_rate": 3.493049924223872e-05,
      "loss": 1.0436,
      "num_input_tokens_seen": 8204560,
      "step": 14800
    },
    {
      "epoch": 259.65486725663715,
      "eval_loss": 1.0242397785186768,
      "eval_runtime": 0.9382,
      "eval_samples_per_second": 26.647,
      "eval_steps_per_second": 13.857,
      "num_input_tokens_seen": 8204560,
      "step": 14800
    },
    {
      "epoch": 259.7433628318584,
      "grad_norm": 11.0625,
      "learning_rate": 3.49214887480519e-05,
      "loss": 1.0632,
      "num_input_tokens_seen": 8207104,
      "step": 14805
    },
    {
      "epoch": 259.83185840707966,
      "grad_norm": 7.28125,
      "learning_rate": 3.4912476723846834e-05,
      "loss": 0.9904,
      "num_input_tokens_seen": 8210144,
      "step": 14810
    },
    {
      "epoch": 259.92035398230087,
      "grad_norm": 8.25,
      "learning_rate": 3.490346317101328e-05,
      "loss": 1.0487,
      "num_input_tokens_seen": 8213200,
      "step": 14815
    },
    {
      "epoch": 260.0,
      "grad_norm": 6.59375,
      "learning_rate": 3.4894448090941266e-05,
      "loss": 0.8863,
      "num_input_tokens_seen": 8215984,
      "step": 14820
    },
    {
      "epoch": 260.08849557522126,
      "grad_norm": 7.28125,
      "learning_rate": 3.488543148502101e-05,
      "loss": 0.973,
      "num_input_tokens_seen": 8218864,
      "step": 14825
    },
    {
      "epoch": 260.17699115044246,
      "grad_norm": 7.625,
      "learning_rate": 3.487641335464299e-05,
      "loss": 1.0479,
      "num_input_tokens_seen": 8221392,
      "step": 14830
    },
    {
      "epoch": 260.2654867256637,
      "grad_norm": 8.625,
      "learning_rate": 3.4867393701197914e-05,
      "loss": 1.0904,
      "num_input_tokens_seen": 8224288,
      "step": 14835
    },
    {
      "epoch": 260.353982300885,
      "grad_norm": 6.375,
      "learning_rate": 3.485837252607673e-05,
      "loss": 0.964,
      "num_input_tokens_seen": 8226864,
      "step": 14840
    },
    {
      "epoch": 260.4424778761062,
      "grad_norm": 7.1875,
      "learning_rate": 3.4849349830670615e-05,
      "loss": 0.9333,
      "num_input_tokens_seen": 8229584,
      "step": 14845
    },
    {
      "epoch": 260.53097345132744,
      "grad_norm": 8.8125,
      "learning_rate": 3.4840325616370976e-05,
      "loss": 0.984,
      "num_input_tokens_seen": 8232192,
      "step": 14850
    },
    {
      "epoch": 260.6194690265487,
      "grad_norm": 6.40625,
      "learning_rate": 3.483129988456947e-05,
      "loss": 0.9637,
      "num_input_tokens_seen": 8235536,
      "step": 14855
    },
    {
      "epoch": 260.7079646017699,
      "grad_norm": 7.71875,
      "learning_rate": 3.482227263665797e-05,
      "loss": 1.105,
      "num_input_tokens_seen": 8238048,
      "step": 14860
    },
    {
      "epoch": 260.79646017699116,
      "grad_norm": 12.875,
      "learning_rate": 3.48132438740286e-05,
      "loss": 1.0346,
      "num_input_tokens_seen": 8241328,
      "step": 14865
    },
    {
      "epoch": 260.88495575221236,
      "grad_norm": 9.125,
      "learning_rate": 3.48042135980737e-05,
      "loss": 1.2581,
      "num_input_tokens_seen": 8243792,
      "step": 14870
    },
    {
      "epoch": 260.9734513274336,
      "grad_norm": 9.3125,
      "learning_rate": 3.479518181018586e-05,
      "loss": 1.2744,
      "num_input_tokens_seen": 8246512,
      "step": 14875
    },
    {
      "epoch": 261.05309734513276,
      "grad_norm": 7.65625,
      "learning_rate": 3.4786148511757886e-05,
      "loss": 1.1015,
      "num_input_tokens_seen": 8249144,
      "step": 14880
    },
    {
      "epoch": 261.14159292035396,
      "grad_norm": 8.0625,
      "learning_rate": 3.477711370418284e-05,
      "loss": 0.9184,
      "num_input_tokens_seen": 8252264,
      "step": 14885
    },
    {
      "epoch": 261.2300884955752,
      "grad_norm": 10.5,
      "learning_rate": 3.476807738885399e-05,
      "loss": 1.1384,
      "num_input_tokens_seen": 8255688,
      "step": 14890
    },
    {
      "epoch": 261.3185840707965,
      "grad_norm": 7.84375,
      "learning_rate": 3.475903956716485e-05,
      "loss": 1.0156,
      "num_input_tokens_seen": 8258840,
      "step": 14895
    },
    {
      "epoch": 261.4070796460177,
      "grad_norm": 9.375,
      "learning_rate": 3.475000024050917e-05,
      "loss": 1.1623,
      "num_input_tokens_seen": 8261144,
      "step": 14900
    },
    {
      "epoch": 261.49557522123894,
      "grad_norm": 12.5,
      "learning_rate": 3.4740959410280926e-05,
      "loss": 1.1742,
      "num_input_tokens_seen": 8264152,
      "step": 14905
    },
    {
      "epoch": 261.5840707964602,
      "grad_norm": 8.3125,
      "learning_rate": 3.4731917077874324e-05,
      "loss": 1.1461,
      "num_input_tokens_seen": 8266712,
      "step": 14910
    },
    {
      "epoch": 261.6725663716814,
      "grad_norm": 6.4375,
      "learning_rate": 3.4722873244683816e-05,
      "loss": 0.9575,
      "num_input_tokens_seen": 8269208,
      "step": 14915
    },
    {
      "epoch": 261.76106194690266,
      "grad_norm": 8.5625,
      "learning_rate": 3.4713827912104065e-05,
      "loss": 0.8856,
      "num_input_tokens_seen": 8272152,
      "step": 14920
    },
    {
      "epoch": 261.8495575221239,
      "grad_norm": 8.3125,
      "learning_rate": 3.470478108152998e-05,
      "loss": 0.9948,
      "num_input_tokens_seen": 8274696,
      "step": 14925
    },
    {
      "epoch": 261.9380530973451,
      "grad_norm": 8.4375,
      "learning_rate": 3.4695732754356695e-05,
      "loss": 1.1637,
      "num_input_tokens_seen": 8277576,
      "step": 14930
    },
    {
      "epoch": 262.01769911504425,
      "grad_norm": 8.1875,
      "learning_rate": 3.4686682931979576e-05,
      "loss": 1.0296,
      "num_input_tokens_seen": 8279792,
      "step": 14935
    },
    {
      "epoch": 262.1061946902655,
      "grad_norm": 6.09375,
      "learning_rate": 3.467763161579422e-05,
      "loss": 0.9039,
      "num_input_tokens_seen": 8282528,
      "step": 14940
    },
    {
      "epoch": 262.1946902654867,
      "grad_norm": 8.0625,
      "learning_rate": 3.466857880719645e-05,
      "loss": 1.0334,
      "num_input_tokens_seen": 8285648,
      "step": 14945
    },
    {
      "epoch": 262.283185840708,
      "grad_norm": 7.40625,
      "learning_rate": 3.465952450758233e-05,
      "loss": 0.9119,
      "num_input_tokens_seen": 8288480,
      "step": 14950
    },
    {
      "epoch": 262.37168141592923,
      "grad_norm": 6.5,
      "learning_rate": 3.4650468718348126e-05,
      "loss": 1.0479,
      "num_input_tokens_seen": 8291232,
      "step": 14955
    },
    {
      "epoch": 262.46017699115043,
      "grad_norm": 7.875,
      "learning_rate": 3.464141144089038e-05,
      "loss": 1.1732,
      "num_input_tokens_seen": 8294080,
      "step": 14960
    },
    {
      "epoch": 262.5486725663717,
      "grad_norm": 9.5625,
      "learning_rate": 3.463235267660583e-05,
      "loss": 1.0786,
      "num_input_tokens_seen": 8297088,
      "step": 14965
    },
    {
      "epoch": 262.6371681415929,
      "grad_norm": 8.25,
      "learning_rate": 3.462329242689145e-05,
      "loss": 1.0393,
      "num_input_tokens_seen": 8300160,
      "step": 14970
    },
    {
      "epoch": 262.72566371681415,
      "grad_norm": 8.0625,
      "learning_rate": 3.461423069314444e-05,
      "loss": 1.052,
      "num_input_tokens_seen": 8302640,
      "step": 14975
    },
    {
      "epoch": 262.8141592920354,
      "grad_norm": 12.75,
      "learning_rate": 3.460516747676224e-05,
      "loss": 1.0114,
      "num_input_tokens_seen": 8305440,
      "step": 14980
    },
    {
      "epoch": 262.9026548672566,
      "grad_norm": 10.8125,
      "learning_rate": 3.459610277914251e-05,
      "loss": 1.3089,
      "num_input_tokens_seen": 8308464,
      "step": 14985
    },
    {
      "epoch": 262.9911504424779,
      "grad_norm": 7.0625,
      "learning_rate": 3.458703660168314e-05,
      "loss": 0.9891,
      "num_input_tokens_seen": 8310768,
      "step": 14990
    },
    {
      "epoch": 263.070796460177,
      "grad_norm": 7.3125,
      "learning_rate": 3.457796894578224e-05,
      "loss": 1.2368,
      "num_input_tokens_seen": 8313464,
      "step": 14995
    },
    {
      "epoch": 263.1592920353982,
      "grad_norm": 8.0,
      "learning_rate": 3.456889981283817e-05,
      "loss": 1.0435,
      "num_input_tokens_seen": 8315912,
      "step": 15000
    },
    {
      "epoch": 263.1592920353982,
      "eval_loss": 1.029687523841858,
      "eval_runtime": 0.9355,
      "eval_samples_per_second": 26.723,
      "eval_steps_per_second": 13.896,
      "num_input_tokens_seen": 8315912,
      "step": 15000
    },
    {
      "epoch": 263.24778761061947,
      "grad_norm": 8.75,
      "learning_rate": 3.45598292042495e-05,
      "loss": 1.1355,
      "num_input_tokens_seen": 8318648,
      "step": 15005
    },
    {
      "epoch": 263.3362831858407,
      "grad_norm": 9.25,
      "learning_rate": 3.4550757121415035e-05,
      "loss": 1.0737,
      "num_input_tokens_seen": 8321528,
      "step": 15010
    },
    {
      "epoch": 263.42477876106193,
      "grad_norm": 8.5625,
      "learning_rate": 3.454168356573378e-05,
      "loss": 1.1055,
      "num_input_tokens_seen": 8324264,
      "step": 15015
    },
    {
      "epoch": 263.5132743362832,
      "grad_norm": 7.75,
      "learning_rate": 3.453260853860503e-05,
      "loss": 0.9774,
      "num_input_tokens_seen": 8327096,
      "step": 15020
    },
    {
      "epoch": 263.60176991150445,
      "grad_norm": 7.9375,
      "learning_rate": 3.452353204142824e-05,
      "loss": 1.0123,
      "num_input_tokens_seen": 8330120,
      "step": 15025
    },
    {
      "epoch": 263.69026548672565,
      "grad_norm": 9.125,
      "learning_rate": 3.4514454075603136e-05,
      "loss": 0.9749,
      "num_input_tokens_seen": 8332776,
      "step": 15030
    },
    {
      "epoch": 263.7787610619469,
      "grad_norm": 7.9375,
      "learning_rate": 3.450537464252964e-05,
      "loss": 0.9661,
      "num_input_tokens_seen": 8336120,
      "step": 15035
    },
    {
      "epoch": 263.86725663716817,
      "grad_norm": 8.625,
      "learning_rate": 3.4496293743607925e-05,
      "loss": 1.2408,
      "num_input_tokens_seen": 8338488,
      "step": 15040
    },
    {
      "epoch": 263.95575221238937,
      "grad_norm": 8.375,
      "learning_rate": 3.448721138023838e-05,
      "loss": 0.994,
      "num_input_tokens_seen": 8341176,
      "step": 15045
    },
    {
      "epoch": 264.0353982300885,
      "grad_norm": 6.90625,
      "learning_rate": 3.447812755382162e-05,
      "loss": 1.1085,
      "num_input_tokens_seen": 8343288,
      "step": 15050
    },
    {
      "epoch": 264.12389380530976,
      "grad_norm": 9.0625,
      "learning_rate": 3.446904226575847e-05,
      "loss": 0.9733,
      "num_input_tokens_seen": 8346104,
      "step": 15055
    },
    {
      "epoch": 264.21238938053096,
      "grad_norm": 6.84375,
      "learning_rate": 3.445995551745002e-05,
      "loss": 1.0651,
      "num_input_tokens_seen": 8349000,
      "step": 15060
    },
    {
      "epoch": 264.3008849557522,
      "grad_norm": 8.875,
      "learning_rate": 3.445086731029753e-05,
      "loss": 1.0261,
      "num_input_tokens_seen": 8351512,
      "step": 15065
    },
    {
      "epoch": 264.3893805309734,
      "grad_norm": 6.375,
      "learning_rate": 3.444177764570255e-05,
      "loss": 0.9481,
      "num_input_tokens_seen": 8354328,
      "step": 15070
    },
    {
      "epoch": 264.4778761061947,
      "grad_norm": 9.375,
      "learning_rate": 3.44326865250668e-05,
      "loss": 1.085,
      "num_input_tokens_seen": 8357112,
      "step": 15075
    },
    {
      "epoch": 264.56637168141594,
      "grad_norm": 7.4375,
      "learning_rate": 3.442359394979225e-05,
      "loss": 1.0824,
      "num_input_tokens_seen": 8359656,
      "step": 15080
    },
    {
      "epoch": 264.65486725663715,
      "grad_norm": 8.875,
      "learning_rate": 3.441449992128108e-05,
      "loss": 1.1726,
      "num_input_tokens_seen": 8362056,
      "step": 15085
    },
    {
      "epoch": 264.7433628318584,
      "grad_norm": 7.03125,
      "learning_rate": 3.440540444093573e-05,
      "loss": 1.0257,
      "num_input_tokens_seen": 8364824,
      "step": 15090
    },
    {
      "epoch": 264.83185840707966,
      "grad_norm": 6.4375,
      "learning_rate": 3.43963075101588e-05,
      "loss": 1.1307,
      "num_input_tokens_seen": 8367816,
      "step": 15095
    },
    {
      "epoch": 264.92035398230087,
      "grad_norm": 10.75,
      "learning_rate": 3.438720913035318e-05,
      "loss": 1.1883,
      "num_input_tokens_seen": 8370600,
      "step": 15100
    },
    {
      "epoch": 265.0,
      "grad_norm": 8.1875,
      "learning_rate": 3.437810930292195e-05,
      "loss": 0.9897,
      "num_input_tokens_seen": 8373448,
      "step": 15105
    },
    {
      "epoch": 265.08849557522126,
      "grad_norm": 9.375,
      "learning_rate": 3.43690080292684e-05,
      "loss": 1.0069,
      "num_input_tokens_seen": 8375960,
      "step": 15110
    },
    {
      "epoch": 265.17699115044246,
      "grad_norm": 9.625,
      "learning_rate": 3.435990531079608e-05,
      "loss": 1.0998,
      "num_input_tokens_seen": 8378584,
      "step": 15115
    },
    {
      "epoch": 265.2654867256637,
      "grad_norm": 12.4375,
      "learning_rate": 3.435080114890874e-05,
      "loss": 1.1625,
      "num_input_tokens_seen": 8381112,
      "step": 15120
    },
    {
      "epoch": 265.353982300885,
      "grad_norm": 7.75,
      "learning_rate": 3.434169554501035e-05,
      "loss": 1.0431,
      "num_input_tokens_seen": 8383944,
      "step": 15125
    },
    {
      "epoch": 265.4424778761062,
      "grad_norm": 9.6875,
      "learning_rate": 3.433258850050511e-05,
      "loss": 1.0892,
      "num_input_tokens_seen": 8387128,
      "step": 15130
    },
    {
      "epoch": 265.53097345132744,
      "grad_norm": 9.25,
      "learning_rate": 3.4323480016797446e-05,
      "loss": 1.0257,
      "num_input_tokens_seen": 8389800,
      "step": 15135
    },
    {
      "epoch": 265.6194690265487,
      "grad_norm": 8.375,
      "learning_rate": 3.4314370095291995e-05,
      "loss": 1.1183,
      "num_input_tokens_seen": 8392424,
      "step": 15140
    },
    {
      "epoch": 265.7079646017699,
      "grad_norm": 7.65625,
      "learning_rate": 3.430525873739363e-05,
      "loss": 1.0018,
      "num_input_tokens_seen": 8395032,
      "step": 15145
    },
    {
      "epoch": 265.79646017699116,
      "grad_norm": 9.25,
      "learning_rate": 3.429614594450743e-05,
      "loss": 0.979,
      "num_input_tokens_seen": 8397752,
      "step": 15150
    },
    {
      "epoch": 265.88495575221236,
      "grad_norm": 9.0625,
      "learning_rate": 3.428703171803869e-05,
      "loss": 1.1356,
      "num_input_tokens_seen": 8401048,
      "step": 15155
    },
    {
      "epoch": 265.9734513274336,
      "grad_norm": 5.375,
      "learning_rate": 3.4277916059392964e-05,
      "loss": 0.8559,
      "num_input_tokens_seen": 8404280,
      "step": 15160
    },
    {
      "epoch": 266.05309734513276,
      "grad_norm": 9.5,
      "learning_rate": 3.426879896997598e-05,
      "loss": 1.029,
      "num_input_tokens_seen": 8406656,
      "step": 15165
    },
    {
      "epoch": 266.14159292035396,
      "grad_norm": 10.0,
      "learning_rate": 3.425968045119372e-05,
      "loss": 0.9032,
      "num_input_tokens_seen": 8409536,
      "step": 15170
    },
    {
      "epoch": 266.2300884955752,
      "grad_norm": 8.9375,
      "learning_rate": 3.425056050445237e-05,
      "loss": 1.0541,
      "num_input_tokens_seen": 8412192,
      "step": 15175
    },
    {
      "epoch": 266.3185840707965,
      "grad_norm": 8.4375,
      "learning_rate": 3.4241439131158336e-05,
      "loss": 0.9801,
      "num_input_tokens_seen": 8414912,
      "step": 15180
    },
    {
      "epoch": 266.4070796460177,
      "grad_norm": 8.25,
      "learning_rate": 3.423231633271825e-05,
      "loss": 1.2469,
      "num_input_tokens_seen": 8417520,
      "step": 15185
    },
    {
      "epoch": 266.49557522123894,
      "grad_norm": 7.34375,
      "learning_rate": 3.4223192110538985e-05,
      "loss": 1.0404,
      "num_input_tokens_seen": 8420880,
      "step": 15190
    },
    {
      "epoch": 266.5840707964602,
      "grad_norm": 8.3125,
      "learning_rate": 3.4214066466027575e-05,
      "loss": 0.9663,
      "num_input_tokens_seen": 8423856,
      "step": 15195
    },
    {
      "epoch": 266.6725663716814,
      "grad_norm": 7.78125,
      "learning_rate": 3.4204939400591325e-05,
      "loss": 1.0667,
      "num_input_tokens_seen": 8426448,
      "step": 15200
    },
    {
      "epoch": 266.6725663716814,
      "eval_loss": 1.0314768552780151,
      "eval_runtime": 0.9334,
      "eval_samples_per_second": 26.783,
      "eval_steps_per_second": 13.927,
      "num_input_tokens_seen": 8426448,
      "step": 15200
    },
    {
      "epoch": 266.76106194690266,
      "grad_norm": 8.5,
      "learning_rate": 3.419581091563775e-05,
      "loss": 1.2602,
      "num_input_tokens_seen": 8429120,
      "step": 15205
    },
    {
      "epoch": 266.8495575221239,
      "grad_norm": 7.34375,
      "learning_rate": 3.418668101257456e-05,
      "loss": 1.0702,
      "num_input_tokens_seen": 8431824,
      "step": 15210
    },
    {
      "epoch": 266.9380530973451,
      "grad_norm": 9.25,
      "learning_rate": 3.417754969280971e-05,
      "loss": 1.1223,
      "num_input_tokens_seen": 8434848,
      "step": 15215
    },
    {
      "epoch": 267.01769911504425,
      "grad_norm": 7.96875,
      "learning_rate": 3.416841695775137e-05,
      "loss": 0.9061,
      "num_input_tokens_seen": 8437512,
      "step": 15220
    },
    {
      "epoch": 267.1061946902655,
      "grad_norm": 6.375,
      "learning_rate": 3.415928280880792e-05,
      "loss": 0.8575,
      "num_input_tokens_seen": 8440728,
      "step": 15225
    },
    {
      "epoch": 267.1946902654867,
      "grad_norm": 12.9375,
      "learning_rate": 3.4150147247387965e-05,
      "loss": 1.0202,
      "num_input_tokens_seen": 8443352,
      "step": 15230
    },
    {
      "epoch": 267.283185840708,
      "grad_norm": 8.8125,
      "learning_rate": 3.4141010274900306e-05,
      "loss": 1.1556,
      "num_input_tokens_seen": 8446360,
      "step": 15235
    },
    {
      "epoch": 267.37168141592923,
      "grad_norm": 8.1875,
      "learning_rate": 3.413187189275399e-05,
      "loss": 1.039,
      "num_input_tokens_seen": 8448984,
      "step": 15240
    },
    {
      "epoch": 267.46017699115043,
      "grad_norm": 7.90625,
      "learning_rate": 3.4122732102358265e-05,
      "loss": 0.9948,
      "num_input_tokens_seen": 8451992,
      "step": 15245
    },
    {
      "epoch": 267.5486725663717,
      "grad_norm": 6.65625,
      "learning_rate": 3.411359090512261e-05,
      "loss": 1.0915,
      "num_input_tokens_seen": 8454120,
      "step": 15250
    },
    {
      "epoch": 267.6371681415929,
      "grad_norm": 8.125,
      "learning_rate": 3.410444830245672e-05,
      "loss": 1.0531,
      "num_input_tokens_seen": 8456904,
      "step": 15255
    },
    {
      "epoch": 267.72566371681415,
      "grad_norm": 8.9375,
      "learning_rate": 3.409530429577048e-05,
      "loss": 1.2728,
      "num_input_tokens_seen": 8459592,
      "step": 15260
    },
    {
      "epoch": 267.8141592920354,
      "grad_norm": 8.9375,
      "learning_rate": 3.408615888647402e-05,
      "loss": 1.0493,
      "num_input_tokens_seen": 8462472,
      "step": 15265
    },
    {
      "epoch": 267.9026548672566,
      "grad_norm": 9.25,
      "learning_rate": 3.4077012075977675e-05,
      "loss": 1.0768,
      "num_input_tokens_seen": 8465096,
      "step": 15270
    },
    {
      "epoch": 267.9911504424779,
      "grad_norm": 7.21875,
      "learning_rate": 3.4067863865692e-05,
      "loss": 0.937,
      "num_input_tokens_seen": 8468200,
      "step": 15275
    },
    {
      "epoch": 268.070796460177,
      "grad_norm": 8.5625,
      "learning_rate": 3.4058714257027755e-05,
      "loss": 1.1151,
      "num_input_tokens_seen": 8470832,
      "step": 15280
    },
    {
      "epoch": 268.1592920353982,
      "grad_norm": 11.3125,
      "learning_rate": 3.404956325139594e-05,
      "loss": 1.0379,
      "num_input_tokens_seen": 8473504,
      "step": 15285
    },
    {
      "epoch": 268.24778761061947,
      "grad_norm": 8.25,
      "learning_rate": 3.404041085020775e-05,
      "loss": 1.0449,
      "num_input_tokens_seen": 8476352,
      "step": 15290
    },
    {
      "epoch": 268.3362831858407,
      "grad_norm": 6.34375,
      "learning_rate": 3.403125705487459e-05,
      "loss": 1.1285,
      "num_input_tokens_seen": 8479120,
      "step": 15295
    },
    {
      "epoch": 268.42477876106193,
      "grad_norm": 9.4375,
      "learning_rate": 3.402210186680811e-05,
      "loss": 1.0693,
      "num_input_tokens_seen": 8481904,
      "step": 15300
    },
    {
      "epoch": 268.5132743362832,
      "grad_norm": 7.1875,
      "learning_rate": 3.4012945287420137e-05,
      "loss": 1.0877,
      "num_input_tokens_seen": 8484624,
      "step": 15305
    },
    {
      "epoch": 268.60176991150445,
      "grad_norm": 8.8125,
      "learning_rate": 3.400378731812274e-05,
      "loss": 1.0158,
      "num_input_tokens_seen": 8487408,
      "step": 15310
    },
    {
      "epoch": 268.69026548672565,
      "grad_norm": 12.625,
      "learning_rate": 3.399462796032817e-05,
      "loss": 0.9978,
      "num_input_tokens_seen": 8490016,
      "step": 15315
    },
    {
      "epoch": 268.7787610619469,
      "grad_norm": 9.9375,
      "learning_rate": 3.3985467215448954e-05,
      "loss": 1.2899,
      "num_input_tokens_seen": 8492864,
      "step": 15320
    },
    {
      "epoch": 268.86725663716817,
      "grad_norm": 8.75,
      "learning_rate": 3.3976305084897776e-05,
      "loss": 1.1784,
      "num_input_tokens_seen": 8495776,
      "step": 15325
    },
    {
      "epoch": 268.95575221238937,
      "grad_norm": 7.375,
      "learning_rate": 3.3967141570087544e-05,
      "loss": 0.8345,
      "num_input_tokens_seen": 8498256,
      "step": 15330
    },
    {
      "epoch": 269.0353982300885,
      "grad_norm": 10.0625,
      "learning_rate": 3.39579766724314e-05,
      "loss": 1.003,
      "num_input_tokens_seen": 8500496,
      "step": 15335
    },
    {
      "epoch": 269.12389380530976,
      "grad_norm": 11.6875,
      "learning_rate": 3.3948810393342677e-05,
      "loss": 1.1384,
      "num_input_tokens_seen": 8503024,
      "step": 15340
    },
    {
      "epoch": 269.21238938053096,
      "grad_norm": 8.6875,
      "learning_rate": 3.3939642734234936e-05,
      "loss": 1.105,
      "num_input_tokens_seen": 8506112,
      "step": 15345
    },
    {
      "epoch": 269.3008849557522,
      "grad_norm": 7.65625,
      "learning_rate": 3.393047369652194e-05,
      "loss": 1.0855,
      "num_input_tokens_seen": 8508768,
      "step": 15350
    },
    {
      "epoch": 269.3893805309734,
      "grad_norm": 8.25,
      "learning_rate": 3.3921303281617664e-05,
      "loss": 1.09,
      "num_input_tokens_seen": 8512016,
      "step": 15355
    },
    {
      "epoch": 269.4778761061947,
      "grad_norm": 9.0,
      "learning_rate": 3.391213149093632e-05,
      "loss": 1.1374,
      "num_input_tokens_seen": 8514800,
      "step": 15360
    },
    {
      "epoch": 269.56637168141594,
      "grad_norm": 8.6875,
      "learning_rate": 3.3902958325892303e-05,
      "loss": 1.1364,
      "num_input_tokens_seen": 8517248,
      "step": 15365
    },
    {
      "epoch": 269.65486725663715,
      "grad_norm": 9.9375,
      "learning_rate": 3.389378378790023e-05,
      "loss": 0.8636,
      "num_input_tokens_seen": 8520048,
      "step": 15370
    },
    {
      "epoch": 269.7433628318584,
      "grad_norm": 8.125,
      "learning_rate": 3.388460787837493e-05,
      "loss": 1.0394,
      "num_input_tokens_seen": 8522960,
      "step": 15375
    },
    {
      "epoch": 269.83185840707966,
      "grad_norm": 8.0625,
      "learning_rate": 3.387543059873145e-05,
      "loss": 0.9929,
      "num_input_tokens_seen": 8526064,
      "step": 15380
    },
    {
      "epoch": 269.92035398230087,
      "grad_norm": 8.1875,
      "learning_rate": 3.386625195038503e-05,
      "loss": 1.0092,
      "num_input_tokens_seen": 8528544,
      "step": 15385
    },
    {
      "epoch": 270.0,
      "grad_norm": 3.375,
      "learning_rate": 3.3857071934751136e-05,
      "loss": 0.8201,
      "num_input_tokens_seen": 8531216,
      "step": 15390
    },
    {
      "epoch": 270.08849557522126,
      "grad_norm": 13.1875,
      "learning_rate": 3.384789055324544e-05,
      "loss": 1.1527,
      "num_input_tokens_seen": 8533696,
      "step": 15395
    },
    {
      "epoch": 270.17699115044246,
      "grad_norm": 7.65625,
      "learning_rate": 3.3838707807283843e-05,
      "loss": 1.046,
      "num_input_tokens_seen": 8536288,
      "step": 15400
    },
    {
      "epoch": 270.17699115044246,
      "eval_loss": 1.0310862064361572,
      "eval_runtime": 0.9369,
      "eval_samples_per_second": 26.684,
      "eval_steps_per_second": 13.875,
      "num_input_tokens_seen": 8536288,
      "step": 15400
    },
    {
      "epoch": 270.2654867256637,
      "grad_norm": 7.9375,
      "learning_rate": 3.382952369828243e-05,
      "loss": 0.8661,
      "num_input_tokens_seen": 8539584,
      "step": 15405
    },
    {
      "epoch": 270.353982300885,
      "grad_norm": 7.0,
      "learning_rate": 3.38203382276575e-05,
      "loss": 0.9573,
      "num_input_tokens_seen": 8542128,
      "step": 15410
    },
    {
      "epoch": 270.4424778761062,
      "grad_norm": 11.0625,
      "learning_rate": 3.381115139682557e-05,
      "loss": 1.072,
      "num_input_tokens_seen": 8545088,
      "step": 15415
    },
    {
      "epoch": 270.53097345132744,
      "grad_norm": 7.9375,
      "learning_rate": 3.3801963207203366e-05,
      "loss": 1.1145,
      "num_input_tokens_seen": 8548080,
      "step": 15420
    },
    {
      "epoch": 270.6194690265487,
      "grad_norm": 8.5625,
      "learning_rate": 3.379277366020782e-05,
      "loss": 1.0124,
      "num_input_tokens_seen": 8551104,
      "step": 15425
    },
    {
      "epoch": 270.7079646017699,
      "grad_norm": 9.625,
      "learning_rate": 3.3783582757256085e-05,
      "loss": 1.0462,
      "num_input_tokens_seen": 8553520,
      "step": 15430
    },
    {
      "epoch": 270.79646017699116,
      "grad_norm": 8.5625,
      "learning_rate": 3.3774390499765504e-05,
      "loss": 1.2056,
      "num_input_tokens_seen": 8556160,
      "step": 15435
    },
    {
      "epoch": 270.88495575221236,
      "grad_norm": 7.59375,
      "learning_rate": 3.376519688915364e-05,
      "loss": 1.1425,
      "num_input_tokens_seen": 8559200,
      "step": 15440
    },
    {
      "epoch": 270.9734513274336,
      "grad_norm": 7.875,
      "learning_rate": 3.3756001926838273e-05,
      "loss": 1.041,
      "num_input_tokens_seen": 8562320,
      "step": 15445
    },
    {
      "epoch": 271.05309734513276,
      "grad_norm": 7.71875,
      "learning_rate": 3.374680561423737e-05,
      "loss": 0.9009,
      "num_input_tokens_seen": 8565056,
      "step": 15450
    },
    {
      "epoch": 271.14159292035396,
      "grad_norm": 8.3125,
      "learning_rate": 3.373760795276912e-05,
      "loss": 1.2361,
      "num_input_tokens_seen": 8567968,
      "step": 15455
    },
    {
      "epoch": 271.2300884955752,
      "grad_norm": 10.8125,
      "learning_rate": 3.372840894385192e-05,
      "loss": 1.2689,
      "num_input_tokens_seen": 8571216,
      "step": 15460
    },
    {
      "epoch": 271.3185840707965,
      "grad_norm": 8.125,
      "learning_rate": 3.3719208588904375e-05,
      "loss": 1.1114,
      "num_input_tokens_seen": 8574128,
      "step": 15465
    },
    {
      "epoch": 271.4070796460177,
      "grad_norm": 9.8125,
      "learning_rate": 3.371000688934529e-05,
      "loss": 1.0516,
      "num_input_tokens_seen": 8577072,
      "step": 15470
    },
    {
      "epoch": 271.49557522123894,
      "grad_norm": 8.5625,
      "learning_rate": 3.370080384659369e-05,
      "loss": 1.0787,
      "num_input_tokens_seen": 8579440,
      "step": 15475
    },
    {
      "epoch": 271.5840707964602,
      "grad_norm": 10.875,
      "learning_rate": 3.36915994620688e-05,
      "loss": 0.9044,
      "num_input_tokens_seen": 8581904,
      "step": 15480
    },
    {
      "epoch": 271.6725663716814,
      "grad_norm": 8.0625,
      "learning_rate": 3.3682393737190035e-05,
      "loss": 0.9286,
      "num_input_tokens_seen": 8584656,
      "step": 15485
    },
    {
      "epoch": 271.76106194690266,
      "grad_norm": 13.75,
      "learning_rate": 3.3673186673377054e-05,
      "loss": 1.2246,
      "num_input_tokens_seen": 8587440,
      "step": 15490
    },
    {
      "epoch": 271.8495575221239,
      "grad_norm": 10.8125,
      "learning_rate": 3.366397827204969e-05,
      "loss": 1.0678,
      "num_input_tokens_seen": 8590064,
      "step": 15495
    },
    {
      "epoch": 271.9380530973451,
      "grad_norm": 4.40625,
      "learning_rate": 3.3654768534628e-05,
      "loss": 0.89,
      "num_input_tokens_seen": 8592960,
      "step": 15500
    },
    {
      "epoch": 272.01769911504425,
      "grad_norm": 10.0625,
      "learning_rate": 3.3645557462532245e-05,
      "loss": 0.97,
      "num_input_tokens_seen": 8595168,
      "step": 15505
    },
    {
      "epoch": 272.1061946902655,
      "grad_norm": 6.625,
      "learning_rate": 3.363634505718288e-05,
      "loss": 1.0509,
      "num_input_tokens_seen": 8598400,
      "step": 15510
    },
    {
      "epoch": 272.1946902654867,
      "grad_norm": 11.0625,
      "learning_rate": 3.362713132000057e-05,
      "loss": 1.0383,
      "num_input_tokens_seen": 8600800,
      "step": 15515
    },
    {
      "epoch": 272.283185840708,
      "grad_norm": 6.84375,
      "learning_rate": 3.36179162524062e-05,
      "loss": 0.9113,
      "num_input_tokens_seen": 8603856,
      "step": 15520
    },
    {
      "epoch": 272.37168141592923,
      "grad_norm": 7.8125,
      "learning_rate": 3.3608699855820846e-05,
      "loss": 0.8757,
      "num_input_tokens_seen": 8606416,
      "step": 15525
    },
    {
      "epoch": 272.46017699115043,
      "grad_norm": 7.96875,
      "learning_rate": 3.359948213166578e-05,
      "loss": 1.228,
      "num_input_tokens_seen": 8609920,
      "step": 15530
    },
    {
      "epoch": 272.5486725663717,
      "grad_norm": 7.875,
      "learning_rate": 3.359026308136252e-05,
      "loss": 0.8548,
      "num_input_tokens_seen": 8612592,
      "step": 15535
    },
    {
      "epoch": 272.6371681415929,
      "grad_norm": 9.3125,
      "learning_rate": 3.358104270633272e-05,
      "loss": 1.2415,
      "num_input_tokens_seen": 8615584,
      "step": 15540
    },
    {
      "epoch": 272.72566371681415,
      "grad_norm": 12.3125,
      "learning_rate": 3.357182100799831e-05,
      "loss": 1.2554,
      "num_input_tokens_seen": 8618000,
      "step": 15545
    },
    {
      "epoch": 272.8141592920354,
      "grad_norm": 8.75,
      "learning_rate": 3.3562597987781384e-05,
      "loss": 0.9412,
      "num_input_tokens_seen": 8620800,
      "step": 15550
    },
    {
      "epoch": 272.9026548672566,
      "grad_norm": 6.65625,
      "learning_rate": 3.355337364710424e-05,
      "loss": 1.0395,
      "num_input_tokens_seen": 8623520,
      "step": 15555
    },
    {
      "epoch": 272.9911504424779,
      "grad_norm": 8.5,
      "learning_rate": 3.354414798738939e-05,
      "loss": 1.0864,
      "num_input_tokens_seen": 8626192,
      "step": 15560
    },
    {
      "epoch": 273.070796460177,
      "grad_norm": 11.8125,
      "learning_rate": 3.353492101005955e-05,
      "loss": 1.1554,
      "num_input_tokens_seen": 8628112,
      "step": 15565
    },
    {
      "epoch": 273.1592920353982,
      "grad_norm": 6.53125,
      "learning_rate": 3.352569271653763e-05,
      "loss": 0.9232,
      "num_input_tokens_seen": 8630560,
      "step": 15570
    },
    {
      "epoch": 273.24778761061947,
      "grad_norm": 9.9375,
      "learning_rate": 3.351646310824675e-05,
      "loss": 1.0754,
      "num_input_tokens_seen": 8633376,
      "step": 15575
    },
    {
      "epoch": 273.3362831858407,
      "grad_norm": 9.5625,
      "learning_rate": 3.350723218661023e-05,
      "loss": 1.0919,
      "num_input_tokens_seen": 8636592,
      "step": 15580
    },
    {
      "epoch": 273.42477876106193,
      "grad_norm": 9.6875,
      "learning_rate": 3.349799995305162e-05,
      "loss": 0.971,
      "num_input_tokens_seen": 8639328,
      "step": 15585
    },
    {
      "epoch": 273.5132743362832,
      "grad_norm": 8.0,
      "learning_rate": 3.348876640899461e-05,
      "loss": 1.0971,
      "num_input_tokens_seen": 8642048,
      "step": 15590
    },
    {
      "epoch": 273.60176991150445,
      "grad_norm": 10.4375,
      "learning_rate": 3.3479531555863144e-05,
      "loss": 1.0318,
      "num_input_tokens_seen": 8645424,
      "step": 15595
    },
    {
      "epoch": 273.69026548672565,
      "grad_norm": 11.125,
      "learning_rate": 3.3470295395081344e-05,
      "loss": 0.9525,
      "num_input_tokens_seen": 8648256,
      "step": 15600
    },
    {
      "epoch": 273.69026548672565,
      "eval_loss": 1.0306756496429443,
      "eval_runtime": 0.9336,
      "eval_samples_per_second": 26.779,
      "eval_steps_per_second": 13.925,
      "num_input_tokens_seen": 8648256,
      "step": 15600
    },
    {
      "epoch": 273.7787610619469,
      "grad_norm": 9.1875,
      "learning_rate": 3.3461057928073556e-05,
      "loss": 1.0864,
      "num_input_tokens_seen": 8650896,
      "step": 15605
    },
    {
      "epoch": 273.86725663716817,
      "grad_norm": 7.21875,
      "learning_rate": 3.345181915626431e-05,
      "loss": 0.8419,
      "num_input_tokens_seen": 8653872,
      "step": 15610
    },
    {
      "epoch": 273.95575221238937,
      "grad_norm": 7.53125,
      "learning_rate": 3.344257908107834e-05,
      "loss": 1.256,
      "num_input_tokens_seen": 8656544,
      "step": 15615
    },
    {
      "epoch": 274.0353982300885,
      "grad_norm": 6.875,
      "learning_rate": 3.343333770394058e-05,
      "loss": 1.212,
      "num_input_tokens_seen": 8658960,
      "step": 15620
    },
    {
      "epoch": 274.12389380530976,
      "grad_norm": 11.0625,
      "learning_rate": 3.342409502627616e-05,
      "loss": 1.3062,
      "num_input_tokens_seen": 8661728,
      "step": 15625
    },
    {
      "epoch": 274.21238938053096,
      "grad_norm": 5.59375,
      "learning_rate": 3.341485104951043e-05,
      "loss": 1.038,
      "num_input_tokens_seen": 8664304,
      "step": 15630
    },
    {
      "epoch": 274.3008849557522,
      "grad_norm": 7.84375,
      "learning_rate": 3.340560577506892e-05,
      "loss": 1.0537,
      "num_input_tokens_seen": 8667248,
      "step": 15635
    },
    {
      "epoch": 274.3893805309734,
      "grad_norm": 6.15625,
      "learning_rate": 3.339635920437735e-05,
      "loss": 1.0205,
      "num_input_tokens_seen": 8670176,
      "step": 15640
    },
    {
      "epoch": 274.4778761061947,
      "grad_norm": 11.25,
      "learning_rate": 3.338711133886169e-05,
      "loss": 1.1069,
      "num_input_tokens_seen": 8673200,
      "step": 15645
    },
    {
      "epoch": 274.56637168141594,
      "grad_norm": 10.1875,
      "learning_rate": 3.3377862179948064e-05,
      "loss": 1.105,
      "num_input_tokens_seen": 8675808,
      "step": 15650
    },
    {
      "epoch": 274.65486725663715,
      "grad_norm": 8.9375,
      "learning_rate": 3.336861172906281e-05,
      "loss": 0.9658,
      "num_input_tokens_seen": 8679008,
      "step": 15655
    },
    {
      "epoch": 274.7433628318584,
      "grad_norm": 6.65625,
      "learning_rate": 3.335935998763245e-05,
      "loss": 0.9692,
      "num_input_tokens_seen": 8681856,
      "step": 15660
    },
    {
      "epoch": 274.83185840707966,
      "grad_norm": 7.90625,
      "learning_rate": 3.3350106957083744e-05,
      "loss": 0.9062,
      "num_input_tokens_seen": 8684528,
      "step": 15665
    },
    {
      "epoch": 274.92035398230087,
      "grad_norm": 5.3125,
      "learning_rate": 3.33408526388436e-05,
      "loss": 0.9582,
      "num_input_tokens_seen": 8687136,
      "step": 15670
    },
    {
      "epoch": 275.0,
      "grad_norm": 7.875,
      "learning_rate": 3.3331597034339166e-05,
      "loss": 1.2125,
      "num_input_tokens_seen": 8689400,
      "step": 15675
    },
    {
      "epoch": 275.08849557522126,
      "grad_norm": 7.03125,
      "learning_rate": 3.3322340144997764e-05,
      "loss": 0.9836,
      "num_input_tokens_seen": 8691784,
      "step": 15680
    },
    {
      "epoch": 275.17699115044246,
      "grad_norm": 6.75,
      "learning_rate": 3.331308197224693e-05,
      "loss": 1.1755,
      "num_input_tokens_seen": 8694616,
      "step": 15685
    },
    {
      "epoch": 275.2654867256637,
      "grad_norm": 7.875,
      "learning_rate": 3.330382251751438e-05,
      "loss": 1.1425,
      "num_input_tokens_seen": 8697288,
      "step": 15690
    },
    {
      "epoch": 275.353982300885,
      "grad_norm": 9.75,
      "learning_rate": 3.3294561782228054e-05,
      "loss": 1.1025,
      "num_input_tokens_seen": 8700488,
      "step": 15695
    },
    {
      "epoch": 275.4424778761062,
      "grad_norm": 8.6875,
      "learning_rate": 3.328529976781607e-05,
      "loss": 1.0238,
      "num_input_tokens_seen": 8702824,
      "step": 15700
    },
    {
      "epoch": 275.53097345132744,
      "grad_norm": 6.65625,
      "learning_rate": 3.327603647570673e-05,
      "loss": 0.9523,
      "num_input_tokens_seen": 8705608,
      "step": 15705
    },
    {
      "epoch": 275.6194690265487,
      "grad_norm": 8.375,
      "learning_rate": 3.326677190732857e-05,
      "loss": 1.0347,
      "num_input_tokens_seen": 8708504,
      "step": 15710
    },
    {
      "epoch": 275.7079646017699,
      "grad_norm": 9.0,
      "learning_rate": 3.325750606411029e-05,
      "loss": 0.9199,
      "num_input_tokens_seen": 8711480,
      "step": 15715
    },
    {
      "epoch": 275.79646017699116,
      "grad_norm": 10.375,
      "learning_rate": 3.3248238947480804e-05,
      "loss": 1.0089,
      "num_input_tokens_seen": 8714408,
      "step": 15720
    },
    {
      "epoch": 275.88495575221236,
      "grad_norm": 8.8125,
      "learning_rate": 3.323897055886922e-05,
      "loss": 1.1479,
      "num_input_tokens_seen": 8717112,
      "step": 15725
    },
    {
      "epoch": 275.9734513274336,
      "grad_norm": 6.40625,
      "learning_rate": 3.322970089970484e-05,
      "loss": 1.0878,
      "num_input_tokens_seen": 8719912,
      "step": 15730
    },
    {
      "epoch": 276.05309734513276,
      "grad_norm": 7.4375,
      "learning_rate": 3.3220429971417165e-05,
      "loss": 1.1826,
      "num_input_tokens_seen": 8721888,
      "step": 15735
    },
    {
      "epoch": 276.14159292035396,
      "grad_norm": 7.90625,
      "learning_rate": 3.321115777543588e-05,
      "loss": 1.0303,
      "num_input_tokens_seen": 8724368,
      "step": 15740
    },
    {
      "epoch": 276.2300884955752,
      "grad_norm": 6.59375,
      "learning_rate": 3.320188431319088e-05,
      "loss": 1.0987,
      "num_input_tokens_seen": 8727056,
      "step": 15745
    },
    {
      "epoch": 276.3185840707965,
      "grad_norm": 7.78125,
      "learning_rate": 3.319260958611224e-05,
      "loss": 1.1164,
      "num_input_tokens_seen": 8729440,
      "step": 15750
    },
    {
      "epoch": 276.4070796460177,
      "grad_norm": 6.96875,
      "learning_rate": 3.3183333595630256e-05,
      "loss": 0.9424,
      "num_input_tokens_seen": 8732672,
      "step": 15755
    },
    {
      "epoch": 276.49557522123894,
      "grad_norm": 7.4375,
      "learning_rate": 3.317405634317538e-05,
      "loss": 1.08,
      "num_input_tokens_seen": 8736048,
      "step": 15760
    },
    {
      "epoch": 276.5840707964602,
      "grad_norm": 8.4375,
      "learning_rate": 3.3164777830178315e-05,
      "loss": 1.1242,
      "num_input_tokens_seen": 8739120,
      "step": 15765
    },
    {
      "epoch": 276.6725663716814,
      "grad_norm": 7.59375,
      "learning_rate": 3.315549805806989e-05,
      "loss": 0.967,
      "num_input_tokens_seen": 8742032,
      "step": 15770
    },
    {
      "epoch": 276.76106194690266,
      "grad_norm": 8.125,
      "learning_rate": 3.314621702828118e-05,
      "loss": 1.1101,
      "num_input_tokens_seen": 8744720,
      "step": 15775
    },
    {
      "epoch": 276.8495575221239,
      "grad_norm": 8.25,
      "learning_rate": 3.313693474224342e-05,
      "loss": 1.0191,
      "num_input_tokens_seen": 8747520,
      "step": 15780
    },
    {
      "epoch": 276.9380530973451,
      "grad_norm": 11.9375,
      "learning_rate": 3.312765120138809e-05,
      "loss": 1.1856,
      "num_input_tokens_seen": 8750496,
      "step": 15785
    },
    {
      "epoch": 277.01769911504425,
      "grad_norm": 9.0,
      "learning_rate": 3.311836640714679e-05,
      "loss": 0.9718,
      "num_input_tokens_seen": 8753192,
      "step": 15790
    },
    {
      "epoch": 277.1061946902655,
      "grad_norm": 8.25,
      "learning_rate": 3.310908036095137e-05,
      "loss": 0.9629,
      "num_input_tokens_seen": 8756168,
      "step": 15795
    },
    {
      "epoch": 277.1946902654867,
      "grad_norm": 9.1875,
      "learning_rate": 3.309979306423386e-05,
      "loss": 1.1752,
      "num_input_tokens_seen": 8758760,
      "step": 15800
    },
    {
      "epoch": 277.1946902654867,
      "eval_loss": 1.0300042629241943,
      "eval_runtime": 0.9356,
      "eval_samples_per_second": 26.719,
      "eval_steps_per_second": 13.894,
      "num_input_tokens_seen": 8758760,
      "step": 15800
    },
    {
      "epoch": 277.283185840708,
      "grad_norm": 11.6875,
      "learning_rate": 3.309050451842647e-05,
      "loss": 1.1687,
      "num_input_tokens_seen": 8760984,
      "step": 15805
    },
    {
      "epoch": 277.37168141592923,
      "grad_norm": 7.21875,
      "learning_rate": 3.3081214724961604e-05,
      "loss": 0.9345,
      "num_input_tokens_seen": 8763672,
      "step": 15810
    },
    {
      "epoch": 277.46017699115043,
      "grad_norm": 11.5,
      "learning_rate": 3.307192368527188e-05,
      "loss": 1.3161,
      "num_input_tokens_seen": 8766680,
      "step": 15815
    },
    {
      "epoch": 277.5486725663717,
      "grad_norm": 8.9375,
      "learning_rate": 3.306263140079008e-05,
      "loss": 1.0934,
      "num_input_tokens_seen": 8769656,
      "step": 15820
    },
    {
      "epoch": 277.6371681415929,
      "grad_norm": 8.5,
      "learning_rate": 3.30533378729492e-05,
      "loss": 1.1138,
      "num_input_tokens_seen": 8772216,
      "step": 15825
    },
    {
      "epoch": 277.72566371681415,
      "grad_norm": 6.71875,
      "learning_rate": 3.304404310318242e-05,
      "loss": 0.9439,
      "num_input_tokens_seen": 8775480,
      "step": 15830
    },
    {
      "epoch": 277.8141592920354,
      "grad_norm": 9.3125,
      "learning_rate": 3.3034747092923105e-05,
      "loss": 1.1676,
      "num_input_tokens_seen": 8778600,
      "step": 15835
    },
    {
      "epoch": 277.9026548672566,
      "grad_norm": 6.90625,
      "learning_rate": 3.3025449843604806e-05,
      "loss": 0.7977,
      "num_input_tokens_seen": 8781224,
      "step": 15840
    },
    {
      "epoch": 277.9911504424779,
      "grad_norm": 7.0,
      "learning_rate": 3.30161513566613e-05,
      "loss": 0.9382,
      "num_input_tokens_seen": 8783528,
      "step": 15845
    },
    {
      "epoch": 278.070796460177,
      "grad_norm": 9.0,
      "learning_rate": 3.3006851633526506e-05,
      "loss": 1.0163,
      "num_input_tokens_seen": 8785896,
      "step": 15850
    },
    {
      "epoch": 278.1592920353982,
      "grad_norm": 7.59375,
      "learning_rate": 3.2997550675634584e-05,
      "loss": 1.1124,
      "num_input_tokens_seen": 8788488,
      "step": 15855
    },
    {
      "epoch": 278.24778761061947,
      "grad_norm": 7.375,
      "learning_rate": 3.2988248484419825e-05,
      "loss": 1.0931,
      "num_input_tokens_seen": 8791016,
      "step": 15860
    },
    {
      "epoch": 278.3362831858407,
      "grad_norm": 13.3125,
      "learning_rate": 3.2978945061316776e-05,
      "loss": 0.9691,
      "num_input_tokens_seen": 8793832,
      "step": 15865
    },
    {
      "epoch": 278.42477876106193,
      "grad_norm": 8.0,
      "learning_rate": 3.296964040776013e-05,
      "loss": 1.0751,
      "num_input_tokens_seen": 8796600,
      "step": 15870
    },
    {
      "epoch": 278.5132743362832,
      "grad_norm": 8.6875,
      "learning_rate": 3.296033452518478e-05,
      "loss": 1.1322,
      "num_input_tokens_seen": 8799160,
      "step": 15875
    },
    {
      "epoch": 278.60176991150445,
      "grad_norm": 6.125,
      "learning_rate": 3.2951027415025806e-05,
      "loss": 1.0315,
      "num_input_tokens_seen": 8802456,
      "step": 15880
    },
    {
      "epoch": 278.69026548672565,
      "grad_norm": 6.9375,
      "learning_rate": 3.294171907871849e-05,
      "loss": 1.0636,
      "num_input_tokens_seen": 8805592,
      "step": 15885
    },
    {
      "epoch": 278.7787610619469,
      "grad_norm": 9.3125,
      "learning_rate": 3.293240951769828e-05,
      "loss": 1.1642,
      "num_input_tokens_seen": 8808328,
      "step": 15890
    },
    {
      "epoch": 278.86725663716817,
      "grad_norm": 8.5625,
      "learning_rate": 3.2923098733400846e-05,
      "loss": 0.811,
      "num_input_tokens_seen": 8811016,
      "step": 15895
    },
    {
      "epoch": 278.95575221238937,
      "grad_norm": 8.4375,
      "learning_rate": 3.291378672726202e-05,
      "loss": 1.084,
      "num_input_tokens_seen": 8813928,
      "step": 15900
    },
    {
      "epoch": 279.0353982300885,
      "grad_norm": 7.75,
      "learning_rate": 3.2904473500717824e-05,
      "loss": 0.9349,
      "num_input_tokens_seen": 8816104,
      "step": 15905
    },
    {
      "epoch": 279.12389380530976,
      "grad_norm": 6.4375,
      "learning_rate": 3.289515905520449e-05,
      "loss": 1.0729,
      "num_input_tokens_seen": 8818744,
      "step": 15910
    },
    {
      "epoch": 279.21238938053096,
      "grad_norm": 7.4375,
      "learning_rate": 3.288584339215841e-05,
      "loss": 0.9028,
      "num_input_tokens_seen": 8821304,
      "step": 15915
    },
    {
      "epoch": 279.3008849557522,
      "grad_norm": 9.5625,
      "learning_rate": 3.287652651301617e-05,
      "loss": 0.9049,
      "num_input_tokens_seen": 8824312,
      "step": 15920
    },
    {
      "epoch": 279.3893805309734,
      "grad_norm": 8.375,
      "learning_rate": 3.286720841921457e-05,
      "loss": 0.9368,
      "num_input_tokens_seen": 8827112,
      "step": 15925
    },
    {
      "epoch": 279.4778761061947,
      "grad_norm": 9.3125,
      "learning_rate": 3.285788911219056e-05,
      "loss": 1.2904,
      "num_input_tokens_seen": 8829976,
      "step": 15930
    },
    {
      "epoch": 279.56637168141594,
      "grad_norm": 7.84375,
      "learning_rate": 3.284856859338131e-05,
      "loss": 1.0144,
      "num_input_tokens_seen": 8832952,
      "step": 15935
    },
    {
      "epoch": 279.65486725663715,
      "grad_norm": 8.0625,
      "learning_rate": 3.283924686422414e-05,
      "loss": 1.1852,
      "num_input_tokens_seen": 8835640,
      "step": 15940
    },
    {
      "epoch": 279.7433628318584,
      "grad_norm": 7.5,
      "learning_rate": 3.282992392615659e-05,
      "loss": 0.9078,
      "num_input_tokens_seen": 8838376,
      "step": 15945
    },
    {
      "epoch": 279.83185840707966,
      "grad_norm": 8.875,
      "learning_rate": 3.282059978061638e-05,
      "loss": 1.0848,
      "num_input_tokens_seen": 8841496,
      "step": 15950
    },
    {
      "epoch": 279.92035398230087,
      "grad_norm": 9.375,
      "learning_rate": 3.28112744290414e-05,
      "loss": 1.2046,
      "num_input_tokens_seen": 8844264,
      "step": 15955
    },
    {
      "epoch": 280.0,
      "grad_norm": 9.3125,
      "learning_rate": 3.280194787286974e-05,
      "loss": 1.0715,
      "num_input_tokens_seen": 8846600,
      "step": 15960
    },
    {
      "epoch": 280.08849557522126,
      "grad_norm": 6.5,
      "learning_rate": 3.2792620113539674e-05,
      "loss": 0.9228,
      "num_input_tokens_seen": 8849192,
      "step": 15965
    },
    {
      "epoch": 280.17699115044246,
      "grad_norm": 13.5,
      "learning_rate": 3.278329115248966e-05,
      "loss": 1.2632,
      "num_input_tokens_seen": 8852264,
      "step": 15970
    },
    {
      "epoch": 280.2654867256637,
      "grad_norm": 9.125,
      "learning_rate": 3.277396099115834e-05,
      "loss": 1.2848,
      "num_input_tokens_seen": 8854744,
      "step": 15975
    },
    {
      "epoch": 280.353982300885,
      "grad_norm": 7.65625,
      "learning_rate": 3.276462963098454e-05,
      "loss": 1.0297,
      "num_input_tokens_seen": 8857448,
      "step": 15980
    },
    {
      "epoch": 280.4424778761062,
      "grad_norm": 9.5625,
      "learning_rate": 3.275529707340728e-05,
      "loss": 1.2087,
      "num_input_tokens_seen": 8860424,
      "step": 15985
    },
    {
      "epoch": 280.53097345132744,
      "grad_norm": 8.1875,
      "learning_rate": 3.274596331986574e-05,
      "loss": 1.0286,
      "num_input_tokens_seen": 8863560,
      "step": 15990
    },
    {
      "epoch": 280.6194690265487,
      "grad_norm": 11.0625,
      "learning_rate": 3.273662837179932e-05,
      "loss": 1.1951,
      "num_input_tokens_seen": 8865976,
      "step": 15995
    },
    {
      "epoch": 280.7079646017699,
      "grad_norm": 7.40625,
      "learning_rate": 3.272729223064758e-05,
      "loss": 0.9125,
      "num_input_tokens_seen": 8868600,
      "step": 16000
    },
    {
      "epoch": 280.7079646017699,
      "eval_loss": 1.0274407863616943,
      "eval_runtime": 0.9383,
      "eval_samples_per_second": 26.644,
      "eval_steps_per_second": 13.855,
      "num_input_tokens_seen": 8868600,
      "step": 16000
    },
    {
      "epoch": 280.79646017699116,
      "grad_norm": 7.09375,
      "learning_rate": 3.2717954897850264e-05,
      "loss": 0.9608,
      "num_input_tokens_seen": 8871256,
      "step": 16005
    },
    {
      "epoch": 280.88495575221236,
      "grad_norm": 6.84375,
      "learning_rate": 3.270861637484733e-05,
      "loss": 0.8898,
      "num_input_tokens_seen": 8874680,
      "step": 16010
    },
    {
      "epoch": 280.9734513274336,
      "grad_norm": 8.875,
      "learning_rate": 3.2699276663078867e-05,
      "loss": 0.9148,
      "num_input_tokens_seen": 8877816,
      "step": 16015
    },
    {
      "epoch": 281.05309734513276,
      "grad_norm": 7.375,
      "learning_rate": 3.268993576398519e-05,
      "loss": 1.107,
      "num_input_tokens_seen": 8880176,
      "step": 16020
    },
    {
      "epoch": 281.14159292035396,
      "grad_norm": 8.875,
      "learning_rate": 3.268059367900678e-05,
      "loss": 0.8289,
      "num_input_tokens_seen": 8882880,
      "step": 16025
    },
    {
      "epoch": 281.2300884955752,
      "grad_norm": 7.34375,
      "learning_rate": 3.26712504095843e-05,
      "loss": 1.0745,
      "num_input_tokens_seen": 8885616,
      "step": 16030
    },
    {
      "epoch": 281.3185840707965,
      "grad_norm": 7.53125,
      "learning_rate": 3.2661905957158615e-05,
      "loss": 1.0082,
      "num_input_tokens_seen": 8888192,
      "step": 16035
    },
    {
      "epoch": 281.4070796460177,
      "grad_norm": 8.25,
      "learning_rate": 3.2652560323170734e-05,
      "loss": 1.1328,
      "num_input_tokens_seen": 8891664,
      "step": 16040
    },
    {
      "epoch": 281.49557522123894,
      "grad_norm": 9.5625,
      "learning_rate": 3.264321350906189e-05,
      "loss": 1.1006,
      "num_input_tokens_seen": 8894336,
      "step": 16045
    },
    {
      "epoch": 281.5840707964602,
      "grad_norm": 6.75,
      "learning_rate": 3.263386551627346e-05,
      "loss": 1.1683,
      "num_input_tokens_seen": 8896752,
      "step": 16050
    },
    {
      "epoch": 281.6725663716814,
      "grad_norm": 6.9375,
      "learning_rate": 3.2624516346247055e-05,
      "loss": 0.9368,
      "num_input_tokens_seen": 8899680,
      "step": 16055
    },
    {
      "epoch": 281.76106194690266,
      "grad_norm": 8.875,
      "learning_rate": 3.2615166000424404e-05,
      "loss": 1.0705,
      "num_input_tokens_seen": 8902336,
      "step": 16060
    },
    {
      "epoch": 281.8495575221239,
      "grad_norm": 9.3125,
      "learning_rate": 3.260581448024745e-05,
      "loss": 0.9518,
      "num_input_tokens_seen": 8905184,
      "step": 16065
    },
    {
      "epoch": 281.9380530973451,
      "grad_norm": 11.0625,
      "learning_rate": 3.2596461787158335e-05,
      "loss": 1.185,
      "num_input_tokens_seen": 8908240,
      "step": 16070
    },
    {
      "epoch": 282.01769911504425,
      "grad_norm": 8.6875,
      "learning_rate": 3.258710792259934e-05,
      "loss": 1.1221,
      "num_input_tokens_seen": 8910864,
      "step": 16075
    },
    {
      "epoch": 282.1061946902655,
      "grad_norm": 9.25,
      "learning_rate": 3.257775288801296e-05,
      "loss": 0.9365,
      "num_input_tokens_seen": 8913360,
      "step": 16080
    },
    {
      "epoch": 282.1946902654867,
      "grad_norm": 8.375,
      "learning_rate": 3.256839668484186e-05,
      "loss": 1.0035,
      "num_input_tokens_seen": 8916352,
      "step": 16085
    },
    {
      "epoch": 282.283185840708,
      "grad_norm": 12.75,
      "learning_rate": 3.255903931452888e-05,
      "loss": 1.335,
      "num_input_tokens_seen": 8918896,
      "step": 16090
    },
    {
      "epoch": 282.37168141592923,
      "grad_norm": 7.28125,
      "learning_rate": 3.2549680778517045e-05,
      "loss": 1.0082,
      "num_input_tokens_seen": 8921584,
      "step": 16095
    },
    {
      "epoch": 282.46017699115043,
      "grad_norm": 7.21875,
      "learning_rate": 3.2540321078249556e-05,
      "loss": 1.0713,
      "num_input_tokens_seen": 8924336,
      "step": 16100
    },
    {
      "epoch": 282.5486725663717,
      "grad_norm": 8.25,
      "learning_rate": 3.2530960215169795e-05,
      "loss": 1.1199,
      "num_input_tokens_seen": 8926800,
      "step": 16105
    },
    {
      "epoch": 282.6371681415929,
      "grad_norm": 10.0625,
      "learning_rate": 3.2521598190721345e-05,
      "loss": 0.9382,
      "num_input_tokens_seen": 8929888,
      "step": 16110
    },
    {
      "epoch": 282.72566371681415,
      "grad_norm": 7.9375,
      "learning_rate": 3.251223500634792e-05,
      "loss": 1.1549,
      "num_input_tokens_seen": 8932896,
      "step": 16115
    },
    {
      "epoch": 282.8141592920354,
      "grad_norm": 7.375,
      "learning_rate": 3.2502870663493445e-05,
      "loss": 1.0145,
      "num_input_tokens_seen": 8935728,
      "step": 16120
    },
    {
      "epoch": 282.9026548672566,
      "grad_norm": 7.0625,
      "learning_rate": 3.249350516360203e-05,
      "loss": 1.0167,
      "num_input_tokens_seen": 8938624,
      "step": 16125
    },
    {
      "epoch": 282.9911504424779,
      "grad_norm": 8.75,
      "learning_rate": 3.248413850811797e-05,
      "loss": 1.0157,
      "num_input_tokens_seen": 8941968,
      "step": 16130
    },
    {
      "epoch": 283.070796460177,
      "grad_norm": 7.25,
      "learning_rate": 3.2474770698485677e-05,
      "loss": 0.8019,
      "num_input_tokens_seen": 8944344,
      "step": 16135
    },
    {
      "epoch": 283.1592920353982,
      "grad_norm": 9.9375,
      "learning_rate": 3.246540173614983e-05,
      "loss": 1.199,
      "num_input_tokens_seen": 8947192,
      "step": 16140
    },
    {
      "epoch": 283.24778761061947,
      "grad_norm": 8.1875,
      "learning_rate": 3.2456031622555197e-05,
      "loss": 1.0084,
      "num_input_tokens_seen": 8949976,
      "step": 16145
    },
    {
      "epoch": 283.3362831858407,
      "grad_norm": 11.375,
      "learning_rate": 3.2446660359146794e-05,
      "loss": 0.9527,
      "num_input_tokens_seen": 8952296,
      "step": 16150
    },
    {
      "epoch": 283.42477876106193,
      "grad_norm": 8.5,
      "learning_rate": 3.2437287947369786e-05,
      "loss": 0.9869,
      "num_input_tokens_seen": 8954904,
      "step": 16155
    },
    {
      "epoch": 283.5132743362832,
      "grad_norm": 7.46875,
      "learning_rate": 3.2427914388669525e-05,
      "loss": 1.1029,
      "num_input_tokens_seen": 8957304,
      "step": 16160
    },
    {
      "epoch": 283.60176991150445,
      "grad_norm": 9.3125,
      "learning_rate": 3.241853968449151e-05,
      "loss": 1.0548,
      "num_input_tokens_seen": 8960328,
      "step": 16165
    },
    {
      "epoch": 283.69026548672565,
      "grad_norm": 12.75,
      "learning_rate": 3.240916383628144e-05,
      "loss": 1.0663,
      "num_input_tokens_seen": 8963576,
      "step": 16170
    },
    {
      "epoch": 283.7787610619469,
      "grad_norm": 8.3125,
      "learning_rate": 3.239978684548521e-05,
      "loss": 1.0641,
      "num_input_tokens_seen": 8966472,
      "step": 16175
    },
    {
      "epoch": 283.86725663716817,
      "grad_norm": 10.0625,
      "learning_rate": 3.239040871354885e-05,
      "loss": 1.0207,
      "num_input_tokens_seen": 8969640,
      "step": 16180
    },
    {
      "epoch": 283.95575221238937,
      "grad_norm": 8.125,
      "learning_rate": 3.2381029441918596e-05,
      "loss": 1.1495,
      "num_input_tokens_seen": 8972408,
      "step": 16185
    },
    {
      "epoch": 284.0353982300885,
      "grad_norm": 6.875,
      "learning_rate": 3.2371649032040845e-05,
      "loss": 1.2248,
      "num_input_tokens_seen": 8974808,
      "step": 16190
    },
    {
      "epoch": 284.12389380530976,
      "grad_norm": 8.75,
      "learning_rate": 3.2362267485362174e-05,
      "loss": 1.0855,
      "num_input_tokens_seen": 8977912,
      "step": 16195
    },
    {
      "epoch": 284.21238938053096,
      "grad_norm": 9.0625,
      "learning_rate": 3.235288480332934e-05,
      "loss": 1.0887,
      "num_input_tokens_seen": 8981000,
      "step": 16200
    },
    {
      "epoch": 284.21238938053096,
      "eval_loss": 1.0315310955047607,
      "eval_runtime": 0.9335,
      "eval_samples_per_second": 26.78,
      "eval_steps_per_second": 13.926,
      "num_input_tokens_seen": 8981000,
      "step": 16200
    },
    {
      "epoch": 284.3008849557522,
      "grad_norm": 8.4375,
      "learning_rate": 3.234350098738927e-05,
      "loss": 0.9963,
      "num_input_tokens_seen": 8983960,
      "step": 16205
    },
    {
      "epoch": 284.3893805309734,
      "grad_norm": 8.125,
      "learning_rate": 3.233411603898906e-05,
      "loss": 1.2321,
      "num_input_tokens_seen": 8986744,
      "step": 16210
    },
    {
      "epoch": 284.4778761061947,
      "grad_norm": 5.5,
      "learning_rate": 3.232472995957599e-05,
      "loss": 0.8735,
      "num_input_tokens_seen": 8989256,
      "step": 16215
    },
    {
      "epoch": 284.56637168141594,
      "grad_norm": 8.0625,
      "learning_rate": 3.231534275059751e-05,
      "loss": 1.0322,
      "num_input_tokens_seen": 8991976,
      "step": 16220
    },
    {
      "epoch": 284.65486725663715,
      "grad_norm": 11.4375,
      "learning_rate": 3.230595441350125e-05,
      "loss": 1.1017,
      "num_input_tokens_seen": 8994552,
      "step": 16225
    },
    {
      "epoch": 284.7433628318584,
      "grad_norm": 6.9375,
      "learning_rate": 3.2296564949735e-05,
      "loss": 1.0151,
      "num_input_tokens_seen": 8997256,
      "step": 16230
    },
    {
      "epoch": 284.83185840707966,
      "grad_norm": 8.0625,
      "learning_rate": 3.228717436074675e-05,
      "loss": 1.0506,
      "num_input_tokens_seen": 9000184,
      "step": 16235
    },
    {
      "epoch": 284.92035398230087,
      "grad_norm": 7.625,
      "learning_rate": 3.227778264798463e-05,
      "loss": 0.9812,
      "num_input_tokens_seen": 9002856,
      "step": 16240
    },
    {
      "epoch": 285.0,
      "grad_norm": 10.3125,
      "learning_rate": 3.226838981289698e-05,
      "loss": 0.9774,
      "num_input_tokens_seen": 9005544,
      "step": 16245
    },
    {
      "epoch": 285.08849557522126,
      "grad_norm": 9.0,
      "learning_rate": 3.225899585693227e-05,
      "loss": 1.2658,
      "num_input_tokens_seen": 9008536,
      "step": 16250
    },
    {
      "epoch": 285.17699115044246,
      "grad_norm": 8.5625,
      "learning_rate": 3.224960078153918e-05,
      "loss": 1.0677,
      "num_input_tokens_seen": 9011592,
      "step": 16255
    },
    {
      "epoch": 285.2654867256637,
      "grad_norm": 6.78125,
      "learning_rate": 3.224020458816655e-05,
      "loss": 0.8369,
      "num_input_tokens_seen": 9014152,
      "step": 16260
    },
    {
      "epoch": 285.353982300885,
      "grad_norm": 9.0625,
      "learning_rate": 3.223080727826337e-05,
      "loss": 1.07,
      "num_input_tokens_seen": 9017192,
      "step": 16265
    },
    {
      "epoch": 285.4424778761062,
      "grad_norm": 9.25,
      "learning_rate": 3.222140885327885e-05,
      "loss": 1.1509,
      "num_input_tokens_seen": 9019800,
      "step": 16270
    },
    {
      "epoch": 285.53097345132744,
      "grad_norm": 11.625,
      "learning_rate": 3.221200931466234e-05,
      "loss": 0.9461,
      "num_input_tokens_seen": 9022872,
      "step": 16275
    },
    {
      "epoch": 285.6194690265487,
      "grad_norm": 8.6875,
      "learning_rate": 3.220260866386336e-05,
      "loss": 0.9125,
      "num_input_tokens_seen": 9025832,
      "step": 16280
    },
    {
      "epoch": 285.7079646017699,
      "grad_norm": 6.09375,
      "learning_rate": 3.21932069023316e-05,
      "loss": 1.1163,
      "num_input_tokens_seen": 9028280,
      "step": 16285
    },
    {
      "epoch": 285.79646017699116,
      "grad_norm": 9.25,
      "learning_rate": 3.218380403151695e-05,
      "loss": 1.0238,
      "num_input_tokens_seen": 9030728,
      "step": 16290
    },
    {
      "epoch": 285.88495575221236,
      "grad_norm": 6.84375,
      "learning_rate": 3.217440005286943e-05,
      "loss": 1.2177,
      "num_input_tokens_seen": 9033528,
      "step": 16295
    },
    {
      "epoch": 285.9734513274336,
      "grad_norm": 9.5,
      "learning_rate": 3.216499496783928e-05,
      "loss": 1.017,
      "num_input_tokens_seen": 9036376,
      "step": 16300
    },
    {
      "epoch": 286.05309734513276,
      "grad_norm": 8.3125,
      "learning_rate": 3.2155588777876856e-05,
      "loss": 1.1657,
      "num_input_tokens_seen": 9038928,
      "step": 16305
    },
    {
      "epoch": 286.14159292035396,
      "grad_norm": 10.5,
      "learning_rate": 3.214618148443273e-05,
      "loss": 0.9554,
      "num_input_tokens_seen": 9041568,
      "step": 16310
    },
    {
      "epoch": 286.2300884955752,
      "grad_norm": 6.8125,
      "learning_rate": 3.2136773088957595e-05,
      "loss": 1.1939,
      "num_input_tokens_seen": 9044384,
      "step": 16315
    },
    {
      "epoch": 286.3185840707965,
      "grad_norm": 12.0625,
      "learning_rate": 3.2127363592902374e-05,
      "loss": 1.0847,
      "num_input_tokens_seen": 9047120,
      "step": 16320
    },
    {
      "epoch": 286.4070796460177,
      "grad_norm": 9.125,
      "learning_rate": 3.211795299771812e-05,
      "loss": 1.0262,
      "num_input_tokens_seen": 9050032,
      "step": 16325
    },
    {
      "epoch": 286.49557522123894,
      "grad_norm": 7.78125,
      "learning_rate": 3.210854130485605e-05,
      "loss": 0.8861,
      "num_input_tokens_seen": 9052640,
      "step": 16330
    },
    {
      "epoch": 286.5840707964602,
      "grad_norm": 8.3125,
      "learning_rate": 3.209912851576759e-05,
      "loss": 1.0723,
      "num_input_tokens_seen": 9055600,
      "step": 16335
    },
    {
      "epoch": 286.6725663716814,
      "grad_norm": 9.5625,
      "learning_rate": 3.208971463190431e-05,
      "loss": 1.1532,
      "num_input_tokens_seen": 9058000,
      "step": 16340
    },
    {
      "epoch": 286.76106194690266,
      "grad_norm": 9.75,
      "learning_rate": 3.208029965471793e-05,
      "loss": 0.9808,
      "num_input_tokens_seen": 9060960,
      "step": 16345
    },
    {
      "epoch": 286.8495575221239,
      "grad_norm": 7.5625,
      "learning_rate": 3.2070883585660364e-05,
      "loss": 0.9688,
      "num_input_tokens_seen": 9064208,
      "step": 16350
    },
    {
      "epoch": 286.9380530973451,
      "grad_norm": 9.125,
      "learning_rate": 3.20614664261837e-05,
      "loss": 1.1474,
      "num_input_tokens_seen": 9066704,
      "step": 16355
    },
    {
      "epoch": 287.01769911504425,
      "grad_norm": 8.375,
      "learning_rate": 3.205204817774016e-05,
      "loss": 1.1428,
      "num_input_tokens_seen": 9068864,
      "step": 16360
    },
    {
      "epoch": 287.1061946902655,
      "grad_norm": 9.5625,
      "learning_rate": 3.204262884178218e-05,
      "loss": 1.092,
      "num_input_tokens_seen": 9071792,
      "step": 16365
    },
    {
      "epoch": 287.1946902654867,
      "grad_norm": 6.84375,
      "learning_rate": 3.2033208419762314e-05,
      "loss": 1.0317,
      "num_input_tokens_seen": 9074416,
      "step": 16370
    },
    {
      "epoch": 287.283185840708,
      "grad_norm": 9.0,
      "learning_rate": 3.2023786913133344e-05,
      "loss": 0.9597,
      "num_input_tokens_seen": 9077008,
      "step": 16375
    },
    {
      "epoch": 287.37168141592923,
      "grad_norm": 12.0,
      "learning_rate": 3.201436432334816e-05,
      "loss": 1.1306,
      "num_input_tokens_seen": 9079904,
      "step": 16380
    },
    {
      "epoch": 287.46017699115043,
      "grad_norm": 6.75,
      "learning_rate": 3.2004940651859844e-05,
      "loss": 1.1224,
      "num_input_tokens_seen": 9082848,
      "step": 16385
    },
    {
      "epoch": 287.5486725663717,
      "grad_norm": 12.75,
      "learning_rate": 3.1995515900121655e-05,
      "loss": 1.1686,
      "num_input_tokens_seen": 9085888,
      "step": 16390
    },
    {
      "epoch": 287.6371681415929,
      "grad_norm": 9.4375,
      "learning_rate": 3.1986090069587e-05,
      "loss": 0.988,
      "num_input_tokens_seen": 9088640,
      "step": 16395
    },
    {
      "epoch": 287.72566371681415,
      "grad_norm": 8.5625,
      "learning_rate": 3.1976663161709466e-05,
      "loss": 1.1229,
      "num_input_tokens_seen": 9091424,
      "step": 16400
    },
    {
      "epoch": 287.72566371681415,
      "eval_loss": 1.0302485227584839,
      "eval_runtime": 0.9352,
      "eval_samples_per_second": 26.732,
      "eval_steps_per_second": 13.901,
      "num_input_tokens_seen": 9091424,
      "step": 16400
    },
    {
      "epoch": 287.8141592920354,
      "grad_norm": 9.5,
      "learning_rate": 3.196723517794279e-05,
      "loss": 1.1235,
      "num_input_tokens_seen": 9094592,
      "step": 16405
    },
    {
      "epoch": 287.9026548672566,
      "grad_norm": 7.90625,
      "learning_rate": 3.19578061197409e-05,
      "loss": 0.985,
      "num_input_tokens_seen": 9097472,
      "step": 16410
    },
    {
      "epoch": 287.9911504424779,
      "grad_norm": 7.375,
      "learning_rate": 3.194837598855787e-05,
      "loss": 0.9747,
      "num_input_tokens_seen": 9099936,
      "step": 16415
    },
    {
      "epoch": 288.070796460177,
      "grad_norm": 8.0625,
      "learning_rate": 3.193894478584794e-05,
      "loss": 1.2698,
      "num_input_tokens_seen": 9102288,
      "step": 16420
    },
    {
      "epoch": 288.1592920353982,
      "grad_norm": 7.46875,
      "learning_rate": 3.192951251306553e-05,
      "loss": 1.0454,
      "num_input_tokens_seen": 9105472,
      "step": 16425
    },
    {
      "epoch": 288.24778761061947,
      "grad_norm": 8.875,
      "learning_rate": 3.192007917166521e-05,
      "loss": 0.9291,
      "num_input_tokens_seen": 9108096,
      "step": 16430
    },
    {
      "epoch": 288.3362831858407,
      "grad_norm": 8.6875,
      "learning_rate": 3.191064476310171e-05,
      "loss": 1.0306,
      "num_input_tokens_seen": 9110752,
      "step": 16435
    },
    {
      "epoch": 288.42477876106193,
      "grad_norm": 7.03125,
      "learning_rate": 3.1901209288829944e-05,
      "loss": 1.0292,
      "num_input_tokens_seen": 9113632,
      "step": 16440
    },
    {
      "epoch": 288.5132743362832,
      "grad_norm": 9.5,
      "learning_rate": 3.1891772750304985e-05,
      "loss": 0.9448,
      "num_input_tokens_seen": 9116784,
      "step": 16445
    },
    {
      "epoch": 288.60176991150445,
      "grad_norm": 8.9375,
      "learning_rate": 3.188233514898206e-05,
      "loss": 1.1529,
      "num_input_tokens_seen": 9119440,
      "step": 16450
    },
    {
      "epoch": 288.69026548672565,
      "grad_norm": 6.53125,
      "learning_rate": 3.187289648631657e-05,
      "loss": 0.9815,
      "num_input_tokens_seen": 9122544,
      "step": 16455
    },
    {
      "epoch": 288.7787610619469,
      "grad_norm": 8.9375,
      "learning_rate": 3.186345676376406e-05,
      "loss": 1.1374,
      "num_input_tokens_seen": 9125040,
      "step": 16460
    },
    {
      "epoch": 288.86725663716817,
      "grad_norm": 8.3125,
      "learning_rate": 3.1854015982780275e-05,
      "loss": 1.0509,
      "num_input_tokens_seen": 9127200,
      "step": 16465
    },
    {
      "epoch": 288.95575221238937,
      "grad_norm": 7.75,
      "learning_rate": 3.1844574144821084e-05,
      "loss": 0.9645,
      "num_input_tokens_seen": 9129760,
      "step": 16470
    },
    {
      "epoch": 289.0353982300885,
      "grad_norm": 10.0625,
      "learning_rate": 3.1835131251342554e-05,
      "loss": 1.1034,
      "num_input_tokens_seen": 9132360,
      "step": 16475
    },
    {
      "epoch": 289.12389380530976,
      "grad_norm": 6.8125,
      "learning_rate": 3.182568730380089e-05,
      "loss": 0.8654,
      "num_input_tokens_seen": 9135240,
      "step": 16480
    },
    {
      "epoch": 289.21238938053096,
      "grad_norm": 9.1875,
      "learning_rate": 3.181624230365245e-05,
      "loss": 1.0435,
      "num_input_tokens_seen": 9137848,
      "step": 16485
    },
    {
      "epoch": 289.3008849557522,
      "grad_norm": 7.96875,
      "learning_rate": 3.180679625235381e-05,
      "loss": 1.0674,
      "num_input_tokens_seen": 9140872,
      "step": 16490
    },
    {
      "epoch": 289.3893805309734,
      "grad_norm": 9.1875,
      "learning_rate": 3.1797349151361646e-05,
      "loss": 1.0937,
      "num_input_tokens_seen": 9143848,
      "step": 16495
    },
    {
      "epoch": 289.4778761061947,
      "grad_norm": 17.75,
      "learning_rate": 3.178790100213281e-05,
      "loss": 1.1755,
      "num_input_tokens_seen": 9146536,
      "step": 16500
    },
    {
      "epoch": 289.56637168141594,
      "grad_norm": 8.875,
      "learning_rate": 3.1778451806124346e-05,
      "loss": 1.1035,
      "num_input_tokens_seen": 9149240,
      "step": 16505
    },
    {
      "epoch": 289.65486725663715,
      "grad_norm": 8.75,
      "learning_rate": 3.176900156479342e-05,
      "loss": 1.1155,
      "num_input_tokens_seen": 9151608,
      "step": 16510
    },
    {
      "epoch": 289.7433628318584,
      "grad_norm": 11.4375,
      "learning_rate": 3.17595502795974e-05,
      "loss": 1.1972,
      "num_input_tokens_seen": 9154408,
      "step": 16515
    },
    {
      "epoch": 289.83185840707966,
      "grad_norm": 8.0625,
      "learning_rate": 3.175009795199377e-05,
      "loss": 0.9546,
      "num_input_tokens_seen": 9157368,
      "step": 16520
    },
    {
      "epoch": 289.92035398230087,
      "grad_norm": 8.9375,
      "learning_rate": 3.1740644583440224e-05,
      "loss": 0.8475,
      "num_input_tokens_seen": 9160712,
      "step": 16525
    },
    {
      "epoch": 290.0,
      "grad_norm": 14.4375,
      "learning_rate": 3.173119017539457e-05,
      "loss": 1.1871,
      "num_input_tokens_seen": 9162896,
      "step": 16530
    },
    {
      "epoch": 290.08849557522126,
      "grad_norm": 7.59375,
      "learning_rate": 3.172173472931479e-05,
      "loss": 0.9476,
      "num_input_tokens_seen": 9165936,
      "step": 16535
    },
    {
      "epoch": 290.17699115044246,
      "grad_norm": 7.71875,
      "learning_rate": 3.1712278246659055e-05,
      "loss": 0.9857,
      "num_input_tokens_seen": 9168768,
      "step": 16540
    },
    {
      "epoch": 290.2654867256637,
      "grad_norm": 5.5,
      "learning_rate": 3.170282072888566e-05,
      "loss": 0.9983,
      "num_input_tokens_seen": 9171776,
      "step": 16545
    },
    {
      "epoch": 290.353982300885,
      "grad_norm": 9.375,
      "learning_rate": 3.169336217745307e-05,
      "loss": 1.2425,
      "num_input_tokens_seen": 9174528,
      "step": 16550
    },
    {
      "epoch": 290.4424778761062,
      "grad_norm": 9.6875,
      "learning_rate": 3.1683902593819924e-05,
      "loss": 1.1753,
      "num_input_tokens_seen": 9177216,
      "step": 16555
    },
    {
      "epoch": 290.53097345132744,
      "grad_norm": 7.9375,
      "learning_rate": 3.1674441979445e-05,
      "loss": 1.0264,
      "num_input_tokens_seen": 9180208,
      "step": 16560
    },
    {
      "epoch": 290.6194690265487,
      "grad_norm": 8.3125,
      "learning_rate": 3.166498033578725e-05,
      "loss": 1.2543,
      "num_input_tokens_seen": 9182624,
      "step": 16565
    },
    {
      "epoch": 290.7079646017699,
      "grad_norm": 8.375,
      "learning_rate": 3.165551766430578e-05,
      "loss": 1.0422,
      "num_input_tokens_seen": 9185664,
      "step": 16570
    },
    {
      "epoch": 290.79646017699116,
      "grad_norm": 9.875,
      "learning_rate": 3.164605396645984e-05,
      "loss": 0.9292,
      "num_input_tokens_seen": 9188752,
      "step": 16575
    },
    {
      "epoch": 290.88495575221236,
      "grad_norm": 12.1875,
      "learning_rate": 3.163658924370886e-05,
      "loss": 0.9419,
      "num_input_tokens_seen": 9191536,
      "step": 16580
    },
    {
      "epoch": 290.9734513274336,
      "grad_norm": 6.3125,
      "learning_rate": 3.1627123497512415e-05,
      "loss": 1.097,
      "num_input_tokens_seen": 9193920,
      "step": 16585
    },
    {
      "epoch": 291.05309734513276,
      "grad_norm": 10.25,
      "learning_rate": 3.1617656729330245e-05,
      "loss": 1.035,
      "num_input_tokens_seen": 9196224,
      "step": 16590
    },
    {
      "epoch": 291.14159292035396,
      "grad_norm": 7.375,
      "learning_rate": 3.1608188940622255e-05,
      "loss": 1.2187,
      "num_input_tokens_seen": 9199184,
      "step": 16595
    },
    {
      "epoch": 291.2300884955752,
      "grad_norm": 7.46875,
      "learning_rate": 3.159872013284847e-05,
      "loss": 0.9959,
      "num_input_tokens_seen": 9202432,
      "step": 16600
    },
    {
      "epoch": 291.2300884955752,
      "eval_loss": 1.0354750156402588,
      "eval_runtime": 0.9346,
      "eval_samples_per_second": 26.749,
      "eval_steps_per_second": 13.91,
      "num_input_tokens_seen": 9202432,
      "step": 16600
    },
    {
      "epoch": 291.3185840707965,
      "grad_norm": 6.96875,
      "learning_rate": 3.1589250307469134e-05,
      "loss": 0.9126,
      "num_input_tokens_seen": 9205136,
      "step": 16605
    },
    {
      "epoch": 291.4070796460177,
      "grad_norm": 6.6875,
      "learning_rate": 3.1579779465944586e-05,
      "loss": 1.0909,
      "num_input_tokens_seen": 9207344,
      "step": 16610
    },
    {
      "epoch": 291.49557522123894,
      "grad_norm": 7.9375,
      "learning_rate": 3.1570307609735363e-05,
      "loss": 1.0403,
      "num_input_tokens_seen": 9210192,
      "step": 16615
    },
    {
      "epoch": 291.5840707964602,
      "grad_norm": 7.28125,
      "learning_rate": 3.156083474030213e-05,
      "loss": 1.0111,
      "num_input_tokens_seen": 9212784,
      "step": 16620
    },
    {
      "epoch": 291.6725663716814,
      "grad_norm": 9.1875,
      "learning_rate": 3.155136085910573e-05,
      "loss": 1.1066,
      "num_input_tokens_seen": 9215536,
      "step": 16625
    },
    {
      "epoch": 291.76106194690266,
      "grad_norm": 7.40625,
      "learning_rate": 3.154188596760717e-05,
      "loss": 1.018,
      "num_input_tokens_seen": 9218448,
      "step": 16630
    },
    {
      "epoch": 291.8495575221239,
      "grad_norm": 10.9375,
      "learning_rate": 3.153241006726757e-05,
      "loss": 0.9593,
      "num_input_tokens_seen": 9221520,
      "step": 16635
    },
    {
      "epoch": 291.9380530973451,
      "grad_norm": 6.96875,
      "learning_rate": 3.152293315954825e-05,
      "loss": 1.1437,
      "num_input_tokens_seen": 9224672,
      "step": 16640
    },
    {
      "epoch": 292.01769911504425,
      "grad_norm": 7.1875,
      "learning_rate": 3.1513455245910666e-05,
      "loss": 1.0192,
      "num_input_tokens_seen": 9226712,
      "step": 16645
    },
    {
      "epoch": 292.1061946902655,
      "grad_norm": 10.125,
      "learning_rate": 3.150397632781643e-05,
      "loss": 0.9257,
      "num_input_tokens_seen": 9229528,
      "step": 16650
    },
    {
      "epoch": 292.1946902654867,
      "grad_norm": 7.8125,
      "learning_rate": 3.149449640672731e-05,
      "loss": 1.1901,
      "num_input_tokens_seen": 9232168,
      "step": 16655
    },
    {
      "epoch": 292.283185840708,
      "grad_norm": 9.5,
      "learning_rate": 3.148501548410523e-05,
      "loss": 1.1607,
      "num_input_tokens_seen": 9234808,
      "step": 16660
    },
    {
      "epoch": 292.37168141592923,
      "grad_norm": 9.5625,
      "learning_rate": 3.1475533561412256e-05,
      "loss": 1.2104,
      "num_input_tokens_seen": 9237656,
      "step": 16665
    },
    {
      "epoch": 292.46017699115043,
      "grad_norm": 8.5625,
      "learning_rate": 3.146605064011065e-05,
      "loss": 1.2303,
      "num_input_tokens_seen": 9240312,
      "step": 16670
    },
    {
      "epoch": 292.5486725663717,
      "grad_norm": 9.4375,
      "learning_rate": 3.145656672166277e-05,
      "loss": 0.9538,
      "num_input_tokens_seen": 9242760,
      "step": 16675
    },
    {
      "epoch": 292.6371681415929,
      "grad_norm": 11.25,
      "learning_rate": 3.144708180753116e-05,
      "loss": 0.821,
      "num_input_tokens_seen": 9245384,
      "step": 16680
    },
    {
      "epoch": 292.72566371681415,
      "grad_norm": 7.71875,
      "learning_rate": 3.143759589917851e-05,
      "loss": 1.1449,
      "num_input_tokens_seen": 9248264,
      "step": 16685
    },
    {
      "epoch": 292.8141592920354,
      "grad_norm": 7.53125,
      "learning_rate": 3.142810899806768e-05,
      "loss": 0.9532,
      "num_input_tokens_seen": 9251528,
      "step": 16690
    },
    {
      "epoch": 292.9026548672566,
      "grad_norm": 6.71875,
      "learning_rate": 3.141862110566166e-05,
      "loss": 1.1274,
      "num_input_tokens_seen": 9254392,
      "step": 16695
    },
    {
      "epoch": 292.9911504424779,
      "grad_norm": 7.1875,
      "learning_rate": 3.1409132223423606e-05,
      "loss": 0.9513,
      "num_input_tokens_seen": 9257304,
      "step": 16700
    },
    {
      "epoch": 293.070796460177,
      "grad_norm": 7.21875,
      "learning_rate": 3.139964235281682e-05,
      "loss": 1.159,
      "num_input_tokens_seen": 9259488,
      "step": 16705
    },
    {
      "epoch": 293.1592920353982,
      "grad_norm": 8.1875,
      "learning_rate": 3.139015149530476e-05,
      "loss": 1.0729,
      "num_input_tokens_seen": 9262672,
      "step": 16710
    },
    {
      "epoch": 293.24778761061947,
      "grad_norm": 10.0,
      "learning_rate": 3.1380659652351034e-05,
      "loss": 1.0913,
      "num_input_tokens_seen": 9265264,
      "step": 16715
    },
    {
      "epoch": 293.3362831858407,
      "grad_norm": 7.375,
      "learning_rate": 3.137116682541941e-05,
      "loss": 0.827,
      "num_input_tokens_seen": 9268192,
      "step": 16720
    },
    {
      "epoch": 293.42477876106193,
      "grad_norm": 8.4375,
      "learning_rate": 3.136167301597379e-05,
      "loss": 0.8314,
      "num_input_tokens_seen": 9271136,
      "step": 16725
    },
    {
      "epoch": 293.5132743362832,
      "grad_norm": 15.8125,
      "learning_rate": 3.1352178225478254e-05,
      "loss": 1.201,
      "num_input_tokens_seen": 9273936,
      "step": 16730
    },
    {
      "epoch": 293.60176991150445,
      "grad_norm": 7.125,
      "learning_rate": 3.1342682455396996e-05,
      "loss": 1.06,
      "num_input_tokens_seen": 9276768,
      "step": 16735
    },
    {
      "epoch": 293.69026548672565,
      "grad_norm": 11.25,
      "learning_rate": 3.133318570719441e-05,
      "loss": 1.2197,
      "num_input_tokens_seen": 9279488,
      "step": 16740
    },
    {
      "epoch": 293.7787610619469,
      "grad_norm": 8.0,
      "learning_rate": 3.132368798233499e-05,
      "loss": 1.0079,
      "num_input_tokens_seen": 9282416,
      "step": 16745
    },
    {
      "epoch": 293.86725663716817,
      "grad_norm": 7.5,
      "learning_rate": 3.131418928228342e-05,
      "loss": 1.124,
      "num_input_tokens_seen": 9285488,
      "step": 16750
    },
    {
      "epoch": 293.95575221238937,
      "grad_norm": 5.78125,
      "learning_rate": 3.1304689608504514e-05,
      "loss": 0.9629,
      "num_input_tokens_seen": 9288096,
      "step": 16755
    },
    {
      "epoch": 294.0353982300885,
      "grad_norm": 7.125,
      "learning_rate": 3.129518896246324e-05,
      "loss": 1.192,
      "num_input_tokens_seen": 9290312,
      "step": 16760
    },
    {
      "epoch": 294.12389380530976,
      "grad_norm": 7.3125,
      "learning_rate": 3.128568734562472e-05,
      "loss": 0.9869,
      "num_input_tokens_seen": 9293672,
      "step": 16765
    },
    {
      "epoch": 294.21238938053096,
      "grad_norm": 10.1875,
      "learning_rate": 3.127618475945421e-05,
      "loss": 1.0776,
      "num_input_tokens_seen": 9296376,
      "step": 16770
    },
    {
      "epoch": 294.3008849557522,
      "grad_norm": 11.125,
      "learning_rate": 3.126668120541715e-05,
      "loss": 1.2105,
      "num_input_tokens_seen": 9299240,
      "step": 16775
    },
    {
      "epoch": 294.3893805309734,
      "grad_norm": 7.6875,
      "learning_rate": 3.1257176684979096e-05,
      "loss": 1.0554,
      "num_input_tokens_seen": 9301832,
      "step": 16780
    },
    {
      "epoch": 294.4778761061947,
      "grad_norm": 8.875,
      "learning_rate": 3.124767119960576e-05,
      "loss": 1.1172,
      "num_input_tokens_seen": 9304472,
      "step": 16785
    },
    {
      "epoch": 294.56637168141594,
      "grad_norm": 8.5625,
      "learning_rate": 3.123816475076301e-05,
      "loss": 1.0724,
      "num_input_tokens_seen": 9307144,
      "step": 16790
    },
    {
      "epoch": 294.65486725663715,
      "grad_norm": 5.8125,
      "learning_rate": 3.122865733991687e-05,
      "loss": 1.088,
      "num_input_tokens_seen": 9310040,
      "step": 16795
    },
    {
      "epoch": 294.7433628318584,
      "grad_norm": 8.0,
      "learning_rate": 3.1219148968533486e-05,
      "loss": 1.1444,
      "num_input_tokens_seen": 9312888,
      "step": 16800
    },
    {
      "epoch": 294.7433628318584,
      "eval_loss": 1.0290122032165527,
      "eval_runtime": 0.9356,
      "eval_samples_per_second": 26.722,
      "eval_steps_per_second": 13.895,
      "num_input_tokens_seen": 9312888,
      "step": 16800
    },
    {
      "epoch": 294.83185840707966,
      "grad_norm": 8.1875,
      "learning_rate": 3.120963963807918e-05,
      "loss": 1.0711,
      "num_input_tokens_seen": 9315624,
      "step": 16805
    },
    {
      "epoch": 294.92035398230087,
      "grad_norm": 9.6875,
      "learning_rate": 3.12001293500204e-05,
      "loss": 0.986,
      "num_input_tokens_seen": 9318344,
      "step": 16810
    },
    {
      "epoch": 295.0,
      "grad_norm": 8.75,
      "learning_rate": 3.1190618105823765e-05,
      "loss": 0.8504,
      "num_input_tokens_seen": 9320832,
      "step": 16815
    },
    {
      "epoch": 295.08849557522126,
      "grad_norm": 11.875,
      "learning_rate": 3.118110590695603e-05,
      "loss": 1.0173,
      "num_input_tokens_seen": 9323328,
      "step": 16820
    },
    {
      "epoch": 295.17699115044246,
      "grad_norm": 8.6875,
      "learning_rate": 3.117159275488407e-05,
      "loss": 0.8516,
      "num_input_tokens_seen": 9326256,
      "step": 16825
    },
    {
      "epoch": 295.2654867256637,
      "grad_norm": 12.9375,
      "learning_rate": 3.1162078651074956e-05,
      "loss": 1.1232,
      "num_input_tokens_seen": 9328976,
      "step": 16830
    },
    {
      "epoch": 295.353982300885,
      "grad_norm": 7.0625,
      "learning_rate": 3.1152563596995885e-05,
      "loss": 0.941,
      "num_input_tokens_seen": 9331776,
      "step": 16835
    },
    {
      "epoch": 295.4424778761062,
      "grad_norm": 14.8125,
      "learning_rate": 3.1143047594114186e-05,
      "loss": 1.0459,
      "num_input_tokens_seen": 9334336,
      "step": 16840
    },
    {
      "epoch": 295.53097345132744,
      "grad_norm": 9.75,
      "learning_rate": 3.113353064389734e-05,
      "loss": 1.3993,
      "num_input_tokens_seen": 9336928,
      "step": 16845
    },
    {
      "epoch": 295.6194690265487,
      "grad_norm": 8.75,
      "learning_rate": 3.1124012747812993e-05,
      "loss": 0.9069,
      "num_input_tokens_seen": 9339632,
      "step": 16850
    },
    {
      "epoch": 295.7079646017699,
      "grad_norm": 10.0,
      "learning_rate": 3.1114493907328936e-05,
      "loss": 1.2101,
      "num_input_tokens_seen": 9342544,
      "step": 16855
    },
    {
      "epoch": 295.79646017699116,
      "grad_norm": 8.1875,
      "learning_rate": 3.110497412391306e-05,
      "loss": 1.0849,
      "num_input_tokens_seen": 9345472,
      "step": 16860
    },
    {
      "epoch": 295.88495575221236,
      "grad_norm": 11.5625,
      "learning_rate": 3.1095453399033466e-05,
      "loss": 0.8912,
      "num_input_tokens_seen": 9348720,
      "step": 16865
    },
    {
      "epoch": 295.9734513274336,
      "grad_norm": 8.625,
      "learning_rate": 3.108593173415835e-05,
      "loss": 1.0197,
      "num_input_tokens_seen": 9351520,
      "step": 16870
    },
    {
      "epoch": 296.05309734513276,
      "grad_norm": 7.875,
      "learning_rate": 3.107640913075609e-05,
      "loss": 1.1543,
      "num_input_tokens_seen": 9353912,
      "step": 16875
    },
    {
      "epoch": 296.14159292035396,
      "grad_norm": 7.53125,
      "learning_rate": 3.106688559029517e-05,
      "loss": 1.1111,
      "num_input_tokens_seen": 9356568,
      "step": 16880
    },
    {
      "epoch": 296.2300884955752,
      "grad_norm": 10.5625,
      "learning_rate": 3.105736111424425e-05,
      "loss": 1.0514,
      "num_input_tokens_seen": 9358952,
      "step": 16885
    },
    {
      "epoch": 296.3185840707965,
      "grad_norm": 8.9375,
      "learning_rate": 3.1047835704072136e-05,
      "loss": 0.9686,
      "num_input_tokens_seen": 9361336,
      "step": 16890
    },
    {
      "epoch": 296.4070796460177,
      "grad_norm": 9.3125,
      "learning_rate": 3.103830936124775e-05,
      "loss": 0.8991,
      "num_input_tokens_seen": 9364360,
      "step": 16895
    },
    {
      "epoch": 296.49557522123894,
      "grad_norm": 7.25,
      "learning_rate": 3.102878208724018e-05,
      "loss": 1.172,
      "num_input_tokens_seen": 9367496,
      "step": 16900
    },
    {
      "epoch": 296.5840707964602,
      "grad_norm": 9.9375,
      "learning_rate": 3.101925388351865e-05,
      "loss": 0.9979,
      "num_input_tokens_seen": 9369864,
      "step": 16905
    },
    {
      "epoch": 296.6725663716814,
      "grad_norm": 8.5,
      "learning_rate": 3.1009724751552515e-05,
      "loss": 0.9825,
      "num_input_tokens_seen": 9372552,
      "step": 16910
    },
    {
      "epoch": 296.76106194690266,
      "grad_norm": 7.5625,
      "learning_rate": 3.100019469281131e-05,
      "loss": 0.9846,
      "num_input_tokens_seen": 9375416,
      "step": 16915
    },
    {
      "epoch": 296.8495575221239,
      "grad_norm": 11.8125,
      "learning_rate": 3.0990663708764685e-05,
      "loss": 1.0432,
      "num_input_tokens_seen": 9378600,
      "step": 16920
    },
    {
      "epoch": 296.9380530973451,
      "grad_norm": 9.4375,
      "learning_rate": 3.098113180088243e-05,
      "loss": 1.1746,
      "num_input_tokens_seen": 9381816,
      "step": 16925
    },
    {
      "epoch": 297.01769911504425,
      "grad_norm": 10.5,
      "learning_rate": 3.097159897063448e-05,
      "loss": 1.5458,
      "num_input_tokens_seen": 9384272,
      "step": 16930
    },
    {
      "epoch": 297.1061946902655,
      "grad_norm": 9.8125,
      "learning_rate": 3.096206521949094e-05,
      "loss": 1.139,
      "num_input_tokens_seen": 9387376,
      "step": 16935
    },
    {
      "epoch": 297.1946902654867,
      "grad_norm": 8.1875,
      "learning_rate": 3.0952530548922006e-05,
      "loss": 0.9567,
      "num_input_tokens_seen": 9390032,
      "step": 16940
    },
    {
      "epoch": 297.283185840708,
      "grad_norm": 11.0625,
      "learning_rate": 3.0942994960398064e-05,
      "loss": 0.9288,
      "num_input_tokens_seen": 9393104,
      "step": 16945
    },
    {
      "epoch": 297.37168141592923,
      "grad_norm": 7.5,
      "learning_rate": 3.093345845538961e-05,
      "loss": 1.1417,
      "num_input_tokens_seen": 9395648,
      "step": 16950
    },
    {
      "epoch": 297.46017699115043,
      "grad_norm": 6.0625,
      "learning_rate": 3.09239210353673e-05,
      "loss": 1.0289,
      "num_input_tokens_seen": 9398368,
      "step": 16955
    },
    {
      "epoch": 297.5486725663717,
      "grad_norm": 8.1875,
      "learning_rate": 3.0914382701801926e-05,
      "loss": 0.9084,
      "num_input_tokens_seen": 9401056,
      "step": 16960
    },
    {
      "epoch": 297.6371681415929,
      "grad_norm": 8.0,
      "learning_rate": 3.090484345616441e-05,
      "loss": 1.1697,
      "num_input_tokens_seen": 9403744,
      "step": 16965
    },
    {
      "epoch": 297.72566371681415,
      "grad_norm": 7.375,
      "learning_rate": 3.0895303299925825e-05,
      "loss": 1.0103,
      "num_input_tokens_seen": 9406576,
      "step": 16970
    },
    {
      "epoch": 297.8141592920354,
      "grad_norm": 8.75,
      "learning_rate": 3.0885762234557393e-05,
      "loss": 1.1492,
      "num_input_tokens_seen": 9409040,
      "step": 16975
    },
    {
      "epoch": 297.9026548672566,
      "grad_norm": 10.5,
      "learning_rate": 3.087622026153045e-05,
      "loss": 1.0936,
      "num_input_tokens_seen": 9411792,
      "step": 16980
    },
    {
      "epoch": 297.9911504424779,
      "grad_norm": 7.6875,
      "learning_rate": 3.086667738231651e-05,
      "loss": 0.963,
      "num_input_tokens_seen": 9415008,
      "step": 16985
    },
    {
      "epoch": 298.070796460177,
      "grad_norm": 9.9375,
      "learning_rate": 3.085713359838718e-05,
      "loss": 1.2074,
      "num_input_tokens_seen": 9417672,
      "step": 16990
    },
    {
      "epoch": 298.1592920353982,
      "grad_norm": 6.90625,
      "learning_rate": 3.084758891121425e-05,
      "loss": 0.9617,
      "num_input_tokens_seen": 9420504,
      "step": 16995
    },
    {
      "epoch": 298.24778761061947,
      "grad_norm": 9.75,
      "learning_rate": 3.083804332226963e-05,
      "loss": 1.1921,
      "num_input_tokens_seen": 9423320,
      "step": 17000
    },
    {
      "epoch": 298.24778761061947,
      "eval_loss": 1.0299712419509888,
      "eval_runtime": 0.933,
      "eval_samples_per_second": 26.796,
      "eval_steps_per_second": 13.934,
      "num_input_tokens_seen": 9423320,
      "step": 17000
    },
    {
      "epoch": 298.3362831858407,
      "grad_norm": 8.8125,
      "learning_rate": 3.082849683302536e-05,
      "loss": 0.8426,
      "num_input_tokens_seen": 9426168,
      "step": 17005
    },
    {
      "epoch": 298.42477876106193,
      "grad_norm": 10.4375,
      "learning_rate": 3.081894944495363e-05,
      "loss": 1.1646,
      "num_input_tokens_seen": 9428552,
      "step": 17010
    },
    {
      "epoch": 298.5132743362832,
      "grad_norm": 8.5625,
      "learning_rate": 3.080940115952677e-05,
      "loss": 1.1065,
      "num_input_tokens_seen": 9431384,
      "step": 17015
    },
    {
      "epoch": 298.60176991150445,
      "grad_norm": 8.4375,
      "learning_rate": 3.0799851978217245e-05,
      "loss": 0.9857,
      "num_input_tokens_seen": 9433960,
      "step": 17020
    },
    {
      "epoch": 298.69026548672565,
      "grad_norm": 8.6875,
      "learning_rate": 3.0790301902497666e-05,
      "loss": 0.9011,
      "num_input_tokens_seen": 9437224,
      "step": 17025
    },
    {
      "epoch": 298.7787610619469,
      "grad_norm": 8.5,
      "learning_rate": 3.078075093384076e-05,
      "loss": 1.0153,
      "num_input_tokens_seen": 9440344,
      "step": 17030
    },
    {
      "epoch": 298.86725663716817,
      "grad_norm": 9.8125,
      "learning_rate": 3.077119907371942e-05,
      "loss": 1.1367,
      "num_input_tokens_seen": 9443096,
      "step": 17035
    },
    {
      "epoch": 298.95575221238937,
      "grad_norm": 6.53125,
      "learning_rate": 3.076164632360666e-05,
      "loss": 1.0892,
      "num_input_tokens_seen": 9445624,
      "step": 17040
    },
    {
      "epoch": 299.0353982300885,
      "grad_norm": 10.125,
      "learning_rate": 3.075209268497563e-05,
      "loss": 1.0695,
      "num_input_tokens_seen": 9447824,
      "step": 17045
    },
    {
      "epoch": 299.12389380530976,
      "grad_norm": 9.0,
      "learning_rate": 3.074253815929961e-05,
      "loss": 1.0528,
      "num_input_tokens_seen": 9450640,
      "step": 17050
    },
    {
      "epoch": 299.21238938053096,
      "grad_norm": 10.1875,
      "learning_rate": 3.0732982748052054e-05,
      "loss": 1.0411,
      "num_input_tokens_seen": 9453680,
      "step": 17055
    },
    {
      "epoch": 299.3008849557522,
      "grad_norm": 9.625,
      "learning_rate": 3.072342645270651e-05,
      "loss": 1.1859,
      "num_input_tokens_seen": 9456288,
      "step": 17060
    },
    {
      "epoch": 299.3893805309734,
      "grad_norm": 8.25,
      "learning_rate": 3.071386927473668e-05,
      "loss": 1.0065,
      "num_input_tokens_seen": 9459072,
      "step": 17065
    },
    {
      "epoch": 299.4778761061947,
      "grad_norm": 11.375,
      "learning_rate": 3.0704311215616404e-05,
      "loss": 1.0994,
      "num_input_tokens_seen": 9461952,
      "step": 17070
    },
    {
      "epoch": 299.56637168141594,
      "grad_norm": 8.375,
      "learning_rate": 3.0694752276819656e-05,
      "loss": 1.0611,
      "num_input_tokens_seen": 9464640,
      "step": 17075
    },
    {
      "epoch": 299.65486725663715,
      "grad_norm": 8.25,
      "learning_rate": 3.068519245982054e-05,
      "loss": 1.0226,
      "num_input_tokens_seen": 9467440,
      "step": 17080
    },
    {
      "epoch": 299.7433628318584,
      "grad_norm": 10.9375,
      "learning_rate": 3.0675631766093304e-05,
      "loss": 1.0261,
      "num_input_tokens_seen": 9470064,
      "step": 17085
    },
    {
      "epoch": 299.83185840707966,
      "grad_norm": 7.59375,
      "learning_rate": 3.066607019711232e-05,
      "loss": 0.9015,
      "num_input_tokens_seen": 9473360,
      "step": 17090
    },
    {
      "epoch": 299.92035398230087,
      "grad_norm": 8.0,
      "learning_rate": 3.065650775435211e-05,
      "loss": 1.2262,
      "num_input_tokens_seen": 9476192,
      "step": 17095
    },
    {
      "epoch": 300.0,
      "grad_norm": 8.8125,
      "learning_rate": 3.0646944439287326e-05,
      "loss": 1.0049,
      "num_input_tokens_seen": 9478304,
      "step": 17100
    },
    {
      "epoch": 300.08849557522126,
      "grad_norm": 7.65625,
      "learning_rate": 3.0637380253392736e-05,
      "loss": 1.0436,
      "num_input_tokens_seen": 9480976,
      "step": 17105
    },
    {
      "epoch": 300.17699115044246,
      "grad_norm": 9.0625,
      "learning_rate": 3.062781519814327e-05,
      "loss": 1.0448,
      "num_input_tokens_seen": 9483616,
      "step": 17110
    },
    {
      "epoch": 300.2654867256637,
      "grad_norm": 7.0625,
      "learning_rate": 3.0618249275013985e-05,
      "loss": 1.0026,
      "num_input_tokens_seen": 9486640,
      "step": 17115
    },
    {
      "epoch": 300.353982300885,
      "grad_norm": 6.78125,
      "learning_rate": 3.060868248548005e-05,
      "loss": 1.0654,
      "num_input_tokens_seen": 9489696,
      "step": 17120
    },
    {
      "epoch": 300.4424778761062,
      "grad_norm": 10.1875,
      "learning_rate": 3.0599114831016796e-05,
      "loss": 1.2261,
      "num_input_tokens_seen": 9492544,
      "step": 17125
    },
    {
      "epoch": 300.53097345132744,
      "grad_norm": 9.25,
      "learning_rate": 3.0589546313099666e-05,
      "loss": 1.1312,
      "num_input_tokens_seen": 9495040,
      "step": 17130
    },
    {
      "epoch": 300.6194690265487,
      "grad_norm": 7.09375,
      "learning_rate": 3.0579976933204255e-05,
      "loss": 1.0126,
      "num_input_tokens_seen": 9498368,
      "step": 17135
    },
    {
      "epoch": 300.7079646017699,
      "grad_norm": 8.625,
      "learning_rate": 3.0570406692806284e-05,
      "loss": 1.0639,
      "num_input_tokens_seen": 9501088,
      "step": 17140
    },
    {
      "epoch": 300.79646017699116,
      "grad_norm": 9.375,
      "learning_rate": 3.05608355933816e-05,
      "loss": 1.141,
      "num_input_tokens_seen": 9503712,
      "step": 17145
    },
    {
      "epoch": 300.88495575221236,
      "grad_norm": 6.53125,
      "learning_rate": 3.055126363640618e-05,
      "loss": 0.8737,
      "num_input_tokens_seen": 9506720,
      "step": 17150
    },
    {
      "epoch": 300.9734513274336,
      "grad_norm": 8.0625,
      "learning_rate": 3.0541690823356146e-05,
      "loss": 1.004,
      "num_input_tokens_seen": 9509600,
      "step": 17155
    },
    {
      "epoch": 301.05309734513276,
      "grad_norm": 7.78125,
      "learning_rate": 3.053211715570775e-05,
      "loss": 0.861,
      "num_input_tokens_seen": 9511688,
      "step": 17160
    },
    {
      "epoch": 301.14159292035396,
      "grad_norm": 7.78125,
      "learning_rate": 3.052254263493736e-05,
      "loss": 1.0101,
      "num_input_tokens_seen": 9514280,
      "step": 17165
    },
    {
      "epoch": 301.2300884955752,
      "grad_norm": 8.1875,
      "learning_rate": 3.0512967262521498e-05,
      "loss": 1.0004,
      "num_input_tokens_seen": 9517048,
      "step": 17170
    },
    {
      "epoch": 301.3185840707965,
      "grad_norm": 8.6875,
      "learning_rate": 3.0503391039936803e-05,
      "loss": 0.9438,
      "num_input_tokens_seen": 9520136,
      "step": 17175
    },
    {
      "epoch": 301.4070796460177,
      "grad_norm": 8.125,
      "learning_rate": 3.0493813968660056e-05,
      "loss": 0.9935,
      "num_input_tokens_seen": 9523128,
      "step": 17180
    },
    {
      "epoch": 301.49557522123894,
      "grad_norm": 9.0,
      "learning_rate": 3.0484236050168153e-05,
      "loss": 1.0684,
      "num_input_tokens_seen": 9526008,
      "step": 17185
    },
    {
      "epoch": 301.5840707964602,
      "grad_norm": 9.625,
      "learning_rate": 3.0474657285938123e-05,
      "loss": 1.3268,
      "num_input_tokens_seen": 9528504,
      "step": 17190
    },
    {
      "epoch": 301.6725663716814,
      "grad_norm": 6.125,
      "learning_rate": 3.046507767744715e-05,
      "loss": 0.9716,
      "num_input_tokens_seen": 9531432,
      "step": 17195
    },
    {
      "epoch": 301.76106194690266,
      "grad_norm": 9.6875,
      "learning_rate": 3.045549722617252e-05,
      "loss": 1.1624,
      "num_input_tokens_seen": 9533896,
      "step": 17200
    },
    {
      "epoch": 301.76106194690266,
      "eval_loss": 1.0265752077102661,
      "eval_runtime": 0.9375,
      "eval_samples_per_second": 26.665,
      "eval_steps_per_second": 13.866,
      "num_input_tokens_seen": 9533896,
      "step": 17200
    },
    {
      "epoch": 301.8495575221239,
      "grad_norm": 10.4375,
      "learning_rate": 3.0445915933591658e-05,
      "loss": 1.1091,
      "num_input_tokens_seen": 9536616,
      "step": 17205
    },
    {
      "epoch": 301.9380530973451,
      "grad_norm": 7.65625,
      "learning_rate": 3.0436333801182114e-05,
      "loss": 1.0137,
      "num_input_tokens_seen": 9540312,
      "step": 17210
    },
    {
      "epoch": 302.01769911504425,
      "grad_norm": 7.75,
      "learning_rate": 3.0426750830421596e-05,
      "loss": 1.0982,
      "num_input_tokens_seen": 9542792,
      "step": 17215
    },
    {
      "epoch": 302.1061946902655,
      "grad_norm": 6.9375,
      "learning_rate": 3.0417167022787897e-05,
      "loss": 1.1492,
      "num_input_tokens_seen": 9545224,
      "step": 17220
    },
    {
      "epoch": 302.1946902654867,
      "grad_norm": 10.625,
      "learning_rate": 3.0407582379758966e-05,
      "loss": 1.1866,
      "num_input_tokens_seen": 9547832,
      "step": 17225
    },
    {
      "epoch": 302.283185840708,
      "grad_norm": 13.1875,
      "learning_rate": 3.039799690281287e-05,
      "loss": 1.0298,
      "num_input_tokens_seen": 9550648,
      "step": 17230
    },
    {
      "epoch": 302.37168141592923,
      "grad_norm": 7.6875,
      "learning_rate": 3.0388410593427823e-05,
      "loss": 1.0899,
      "num_input_tokens_seen": 9553528,
      "step": 17235
    },
    {
      "epoch": 302.46017699115043,
      "grad_norm": 6.8125,
      "learning_rate": 3.0378823453082146e-05,
      "loss": 1.0427,
      "num_input_tokens_seen": 9556184,
      "step": 17240
    },
    {
      "epoch": 302.5486725663717,
      "grad_norm": 9.6875,
      "learning_rate": 3.03692354832543e-05,
      "loss": 1.2204,
      "num_input_tokens_seen": 9559192,
      "step": 17245
    },
    {
      "epoch": 302.6371681415929,
      "grad_norm": 5.875,
      "learning_rate": 3.0359646685422865e-05,
      "loss": 1.0397,
      "num_input_tokens_seen": 9561848,
      "step": 17250
    },
    {
      "epoch": 302.72566371681415,
      "grad_norm": 8.125,
      "learning_rate": 3.035005706106656e-05,
      "loss": 0.9908,
      "num_input_tokens_seen": 9564344,
      "step": 17255
    },
    {
      "epoch": 302.8141592920354,
      "grad_norm": 7.3125,
      "learning_rate": 3.034046661166422e-05,
      "loss": 0.9939,
      "num_input_tokens_seen": 9566872,
      "step": 17260
    },
    {
      "epoch": 302.9026548672566,
      "grad_norm": 8.25,
      "learning_rate": 3.033087533869482e-05,
      "loss": 0.9341,
      "num_input_tokens_seen": 9569960,
      "step": 17265
    },
    {
      "epoch": 302.9911504424779,
      "grad_norm": 8.9375,
      "learning_rate": 3.0321283243637444e-05,
      "loss": 0.9542,
      "num_input_tokens_seen": 9573176,
      "step": 17270
    },
    {
      "epoch": 303.070796460177,
      "grad_norm": 6.875,
      "learning_rate": 3.0311690327971326e-05,
      "loss": 0.8825,
      "num_input_tokens_seen": 9575320,
      "step": 17275
    },
    {
      "epoch": 303.1592920353982,
      "grad_norm": 9.5625,
      "learning_rate": 3.030209659317581e-05,
      "loss": 1.1345,
      "num_input_tokens_seen": 9578472,
      "step": 17280
    },
    {
      "epoch": 303.24778761061947,
      "grad_norm": 8.375,
      "learning_rate": 3.0292502040730362e-05,
      "loss": 1.0075,
      "num_input_tokens_seen": 9581336,
      "step": 17285
    },
    {
      "epoch": 303.3362831858407,
      "grad_norm": 8.9375,
      "learning_rate": 3.0282906672114597e-05,
      "loss": 0.9123,
      "num_input_tokens_seen": 9584296,
      "step": 17290
    },
    {
      "epoch": 303.42477876106193,
      "grad_norm": 8.5,
      "learning_rate": 3.027331048880823e-05,
      "loss": 0.9618,
      "num_input_tokens_seen": 9587240,
      "step": 17295
    },
    {
      "epoch": 303.5132743362832,
      "grad_norm": 10.375,
      "learning_rate": 3.0263713492291123e-05,
      "loss": 1.2594,
      "num_input_tokens_seen": 9589944,
      "step": 17300
    },
    {
      "epoch": 303.60176991150445,
      "grad_norm": 10.0625,
      "learning_rate": 3.0254115684043242e-05,
      "loss": 0.9913,
      "num_input_tokens_seen": 9592792,
      "step": 17305
    },
    {
      "epoch": 303.69026548672565,
      "grad_norm": 9.25,
      "learning_rate": 3.024451706554469e-05,
      "loss": 0.8994,
      "num_input_tokens_seen": 9595976,
      "step": 17310
    },
    {
      "epoch": 303.7787610619469,
      "grad_norm": 8.0625,
      "learning_rate": 3.0234917638275705e-05,
      "loss": 1.2282,
      "num_input_tokens_seen": 9598520,
      "step": 17315
    },
    {
      "epoch": 303.86725663716817,
      "grad_norm": 9.8125,
      "learning_rate": 3.0225317403716635e-05,
      "loss": 1.08,
      "num_input_tokens_seen": 9601112,
      "step": 17320
    },
    {
      "epoch": 303.95575221238937,
      "grad_norm": 12.9375,
      "learning_rate": 3.0215716363347956e-05,
      "loss": 1.2637,
      "num_input_tokens_seen": 9603608,
      "step": 17325
    },
    {
      "epoch": 304.0353982300885,
      "grad_norm": 8.875,
      "learning_rate": 3.0206114518650275e-05,
      "loss": 1.0555,
      "num_input_tokens_seen": 9606152,
      "step": 17330
    },
    {
      "epoch": 304.12389380530976,
      "grad_norm": 9.625,
      "learning_rate": 3.0196511871104304e-05,
      "loss": 1.0557,
      "num_input_tokens_seen": 9609144,
      "step": 17335
    },
    {
      "epoch": 304.21238938053096,
      "grad_norm": 8.3125,
      "learning_rate": 3.01869084221909e-05,
      "loss": 1.1774,
      "num_input_tokens_seen": 9611816,
      "step": 17340
    },
    {
      "epoch": 304.3008849557522,
      "grad_norm": 6.8125,
      "learning_rate": 3.0177304173391037e-05,
      "loss": 0.8646,
      "num_input_tokens_seen": 9614296,
      "step": 17345
    },
    {
      "epoch": 304.3893805309734,
      "grad_norm": 7.9375,
      "learning_rate": 3.01676991261858e-05,
      "loss": 1.1192,
      "num_input_tokens_seen": 9616792,
      "step": 17350
    },
    {
      "epoch": 304.4778761061947,
      "grad_norm": 7.46875,
      "learning_rate": 3.015809328205642e-05,
      "loss": 1.2137,
      "num_input_tokens_seen": 9619528,
      "step": 17355
    },
    {
      "epoch": 304.56637168141594,
      "grad_norm": 8.125,
      "learning_rate": 3.0148486642484248e-05,
      "loss": 1.0533,
      "num_input_tokens_seen": 9622232,
      "step": 17360
    },
    {
      "epoch": 304.65486725663715,
      "grad_norm": 7.5625,
      "learning_rate": 3.0138879208950722e-05,
      "loss": 1.0525,
      "num_input_tokens_seen": 9624840,
      "step": 17365
    },
    {
      "epoch": 304.7433628318584,
      "grad_norm": 10.375,
      "learning_rate": 3.012927098293744e-05,
      "loss": 1.2229,
      "num_input_tokens_seen": 9627960,
      "step": 17370
    },
    {
      "epoch": 304.83185840707966,
      "grad_norm": 7.25,
      "learning_rate": 3.0119661965926123e-05,
      "loss": 0.9107,
      "num_input_tokens_seen": 9631272,
      "step": 17375
    },
    {
      "epoch": 304.92035398230087,
      "grad_norm": 11.9375,
      "learning_rate": 3.0110052159398587e-05,
      "loss": 0.8143,
      "num_input_tokens_seen": 9634040,
      "step": 17380
    },
    {
      "epoch": 305.0,
      "grad_norm": 7.90625,
      "learning_rate": 3.0100441564836802e-05,
      "loss": 0.9329,
      "num_input_tokens_seen": 9636376,
      "step": 17385
    },
    {
      "epoch": 305.08849557522126,
      "grad_norm": 7.46875,
      "learning_rate": 3.0090830183722817e-05,
      "loss": 1.021,
      "num_input_tokens_seen": 9639528,
      "step": 17390
    },
    {
      "epoch": 305.17699115044246,
      "grad_norm": 6.90625,
      "learning_rate": 3.0081218017538852e-05,
      "loss": 1.0345,
      "num_input_tokens_seen": 9642328,
      "step": 17395
    },
    {
      "epoch": 305.2654867256637,
      "grad_norm": 9.8125,
      "learning_rate": 3.0071605067767212e-05,
      "loss": 1.1738,
      "num_input_tokens_seen": 9644952,
      "step": 17400
    },
    {
      "epoch": 305.2654867256637,
      "eval_loss": 1.0265377759933472,
      "eval_runtime": 0.9335,
      "eval_samples_per_second": 26.781,
      "eval_steps_per_second": 13.926,
      "num_input_tokens_seen": 9644952,
      "step": 17400
    },
    {
      "epoch": 305.353982300885,
      "grad_norm": 8.8125,
      "learning_rate": 3.006199133589034e-05,
      "loss": 1.014,
      "num_input_tokens_seen": 9647640,
      "step": 17405
    },
    {
      "epoch": 305.4424778761062,
      "grad_norm": 9.5625,
      "learning_rate": 3.005237682339079e-05,
      "loss": 1.045,
      "num_input_tokens_seen": 9650504,
      "step": 17410
    },
    {
      "epoch": 305.53097345132744,
      "grad_norm": 8.5625,
      "learning_rate": 3.0042761531751228e-05,
      "loss": 1.1287,
      "num_input_tokens_seen": 9653256,
      "step": 17415
    },
    {
      "epoch": 305.6194690265487,
      "grad_norm": 7.5,
      "learning_rate": 3.0033145462454482e-05,
      "loss": 0.9232,
      "num_input_tokens_seen": 9655880,
      "step": 17420
    },
    {
      "epoch": 305.7079646017699,
      "grad_norm": 11.25,
      "learning_rate": 3.002352861698345e-05,
      "loss": 1.1022,
      "num_input_tokens_seen": 9659096,
      "step": 17425
    },
    {
      "epoch": 305.79646017699116,
      "grad_norm": 8.6875,
      "learning_rate": 3.0013910996821178e-05,
      "loss": 1.0602,
      "num_input_tokens_seen": 9661896,
      "step": 17430
    },
    {
      "epoch": 305.88495575221236,
      "grad_norm": 8.4375,
      "learning_rate": 3.0004292603450817e-05,
      "loss": 1.0304,
      "num_input_tokens_seen": 9664552,
      "step": 17435
    },
    {
      "epoch": 305.9734513274336,
      "grad_norm": 11.3125,
      "learning_rate": 2.9994673438355653e-05,
      "loss": 1.0614,
      "num_input_tokens_seen": 9667352,
      "step": 17440
    },
    {
      "epoch": 306.05309734513276,
      "grad_norm": 8.375,
      "learning_rate": 2.9985053503019078e-05,
      "loss": 1.1912,
      "num_input_tokens_seen": 9669344,
      "step": 17445
    },
    {
      "epoch": 306.14159292035396,
      "grad_norm": 9.5625,
      "learning_rate": 2.99754327989246e-05,
      "loss": 1.0969,
      "num_input_tokens_seen": 9672208,
      "step": 17450
    },
    {
      "epoch": 306.2300884955752,
      "grad_norm": 8.0625,
      "learning_rate": 2.9965811327555864e-05,
      "loss": 1.2343,
      "num_input_tokens_seen": 9674784,
      "step": 17455
    },
    {
      "epoch": 306.3185840707965,
      "grad_norm": 8.875,
      "learning_rate": 2.995618909039662e-05,
      "loss": 0.9909,
      "num_input_tokens_seen": 9677680,
      "step": 17460
    },
    {
      "epoch": 306.4070796460177,
      "grad_norm": 8.4375,
      "learning_rate": 2.9946566088930727e-05,
      "loss": 0.9664,
      "num_input_tokens_seen": 9680192,
      "step": 17465
    },
    {
      "epoch": 306.49557522123894,
      "grad_norm": 9.3125,
      "learning_rate": 2.9936942324642192e-05,
      "loss": 0.9828,
      "num_input_tokens_seen": 9682752,
      "step": 17470
    },
    {
      "epoch": 306.5840707964602,
      "grad_norm": 9.0625,
      "learning_rate": 2.9927317799015097e-05,
      "loss": 1.0939,
      "num_input_tokens_seen": 9686464,
      "step": 17475
    },
    {
      "epoch": 306.6725663716814,
      "grad_norm": 8.75,
      "learning_rate": 2.9917692513533685e-05,
      "loss": 1.0908,
      "num_input_tokens_seen": 9688832,
      "step": 17480
    },
    {
      "epoch": 306.76106194690266,
      "grad_norm": 8.125,
      "learning_rate": 2.990806646968229e-05,
      "loss": 1.2137,
      "num_input_tokens_seen": 9691968,
      "step": 17485
    },
    {
      "epoch": 306.8495575221239,
      "grad_norm": 11.0625,
      "learning_rate": 2.989843966894536e-05,
      "loss": 0.8778,
      "num_input_tokens_seen": 9694656,
      "step": 17490
    },
    {
      "epoch": 306.9380530973451,
      "grad_norm": 9.0,
      "learning_rate": 2.9888812112807472e-05,
      "loss": 0.9872,
      "num_input_tokens_seen": 9697664,
      "step": 17495
    },
    {
      "epoch": 307.01769911504425,
      "grad_norm": 10.9375,
      "learning_rate": 2.987918380275333e-05,
      "loss": 1.1367,
      "num_input_tokens_seen": 9699792,
      "step": 17500
    },
    {
      "epoch": 307.1061946902655,
      "grad_norm": 6.375,
      "learning_rate": 2.9869554740267724e-05,
      "loss": 0.952,
      "num_input_tokens_seen": 9702592,
      "step": 17505
    },
    {
      "epoch": 307.1946902654867,
      "grad_norm": 10.4375,
      "learning_rate": 2.9859924926835585e-05,
      "loss": 1.1709,
      "num_input_tokens_seen": 9705040,
      "step": 17510
    },
    {
      "epoch": 307.283185840708,
      "grad_norm": 7.96875,
      "learning_rate": 2.9850294363941944e-05,
      "loss": 1.0162,
      "num_input_tokens_seen": 9707776,
      "step": 17515
    },
    {
      "epoch": 307.37168141592923,
      "grad_norm": 8.3125,
      "learning_rate": 2.9840663053071967e-05,
      "loss": 0.9457,
      "num_input_tokens_seen": 9710784,
      "step": 17520
    },
    {
      "epoch": 307.46017699115043,
      "grad_norm": 9.0625,
      "learning_rate": 2.983103099571091e-05,
      "loss": 1.1107,
      "num_input_tokens_seen": 9713072,
      "step": 17525
    },
    {
      "epoch": 307.5486725663717,
      "grad_norm": 8.0,
      "learning_rate": 2.9821398193344164e-05,
      "loss": 0.9502,
      "num_input_tokens_seen": 9716112,
      "step": 17530
    },
    {
      "epoch": 307.6371681415929,
      "grad_norm": 8.75,
      "learning_rate": 2.9811764647457226e-05,
      "loss": 1.1929,
      "num_input_tokens_seen": 9718704,
      "step": 17535
    },
    {
      "epoch": 307.72566371681415,
      "grad_norm": 8.6875,
      "learning_rate": 2.9802130359535714e-05,
      "loss": 1.0216,
      "num_input_tokens_seen": 9721680,
      "step": 17540
    },
    {
      "epoch": 307.8141592920354,
      "grad_norm": 6.125,
      "learning_rate": 2.979249533106535e-05,
      "loss": 0.9361,
      "num_input_tokens_seen": 9724304,
      "step": 17545
    },
    {
      "epoch": 307.9026548672566,
      "grad_norm": 8.8125,
      "learning_rate": 2.9782859563531986e-05,
      "loss": 1.1466,
      "num_input_tokens_seen": 9727296,
      "step": 17550
    },
    {
      "epoch": 307.9911504424779,
      "grad_norm": 7.875,
      "learning_rate": 2.977322305842156e-05,
      "loss": 1.0913,
      "num_input_tokens_seen": 9730144,
      "step": 17555
    },
    {
      "epoch": 308.070796460177,
      "grad_norm": 7.90625,
      "learning_rate": 2.9763585817220162e-05,
      "loss": 1.0863,
      "num_input_tokens_seen": 9732832,
      "step": 17560
    },
    {
      "epoch": 308.1592920353982,
      "grad_norm": 10.6875,
      "learning_rate": 2.975394784141397e-05,
      "loss": 1.0964,
      "num_input_tokens_seen": 9735440,
      "step": 17565
    },
    {
      "epoch": 308.24778761061947,
      "grad_norm": 6.3125,
      "learning_rate": 2.974430913248928e-05,
      "loss": 1.0384,
      "num_input_tokens_seen": 9738592,
      "step": 17570
    },
    {
      "epoch": 308.3362831858407,
      "grad_norm": 8.6875,
      "learning_rate": 2.9734669691932497e-05,
      "loss": 0.9231,
      "num_input_tokens_seen": 9741008,
      "step": 17575
    },
    {
      "epoch": 308.42477876106193,
      "grad_norm": 7.75,
      "learning_rate": 2.9725029521230147e-05,
      "loss": 1.1013,
      "num_input_tokens_seen": 9743632,
      "step": 17580
    },
    {
      "epoch": 308.5132743362832,
      "grad_norm": 10.1875,
      "learning_rate": 2.9715388621868873e-05,
      "loss": 1.22,
      "num_input_tokens_seen": 9746624,
      "step": 17585
    },
    {
      "epoch": 308.60176991150445,
      "grad_norm": 9.9375,
      "learning_rate": 2.970574699533541e-05,
      "loss": 1.1676,
      "num_input_tokens_seen": 9749056,
      "step": 17590
    },
    {
      "epoch": 308.69026548672565,
      "grad_norm": 12.4375,
      "learning_rate": 2.969610464311662e-05,
      "loss": 1.1442,
      "num_input_tokens_seen": 9752208,
      "step": 17595
    },
    {
      "epoch": 308.7787610619469,
      "grad_norm": 8.875,
      "learning_rate": 2.9686461566699487e-05,
      "loss": 0.8876,
      "num_input_tokens_seen": 9754832,
      "step": 17600
    },
    {
      "epoch": 308.7787610619469,
      "eval_loss": 1.0249994993209839,
      "eval_runtime": 0.9396,
      "eval_samples_per_second": 26.606,
      "eval_steps_per_second": 13.835,
      "num_input_tokens_seen": 9754832,
      "step": 17600
    },
    {
      "epoch": 308.86725663716817,
      "grad_norm": 10.5625,
      "learning_rate": 2.9676817767571086e-05,
      "loss": 0.9828,
      "num_input_tokens_seen": 9757472,
      "step": 17605
    },
    {
      "epoch": 308.95575221238937,
      "grad_norm": 8.125,
      "learning_rate": 2.966717324721861e-05,
      "loss": 0.9035,
      "num_input_tokens_seen": 9760704,
      "step": 17610
    },
    {
      "epoch": 309.0353982300885,
      "grad_norm": 7.875,
      "learning_rate": 2.9657528007129366e-05,
      "loss": 1.0418,
      "num_input_tokens_seen": 9763032,
      "step": 17615
    },
    {
      "epoch": 309.12389380530976,
      "grad_norm": 8.125,
      "learning_rate": 2.9647882048790777e-05,
      "loss": 1.0378,
      "num_input_tokens_seen": 9765592,
      "step": 17620
    },
    {
      "epoch": 309.21238938053096,
      "grad_norm": 10.8125,
      "learning_rate": 2.963823537369037e-05,
      "loss": 1.1115,
      "num_input_tokens_seen": 9768664,
      "step": 17625
    },
    {
      "epoch": 309.3008849557522,
      "grad_norm": 7.78125,
      "learning_rate": 2.9628587983315775e-05,
      "loss": 0.9924,
      "num_input_tokens_seen": 9771832,
      "step": 17630
    },
    {
      "epoch": 309.3893805309734,
      "grad_norm": 12.0625,
      "learning_rate": 2.9618939879154746e-05,
      "loss": 0.9843,
      "num_input_tokens_seen": 9774504,
      "step": 17635
    },
    {
      "epoch": 309.4778761061947,
      "grad_norm": 7.1875,
      "learning_rate": 2.9609291062695143e-05,
      "loss": 1.0238,
      "num_input_tokens_seen": 9777160,
      "step": 17640
    },
    {
      "epoch": 309.56637168141594,
      "grad_norm": 11.5,
      "learning_rate": 2.9599641535424938e-05,
      "loss": 1.0965,
      "num_input_tokens_seen": 9779896,
      "step": 17645
    },
    {
      "epoch": 309.65486725663715,
      "grad_norm": 9.125,
      "learning_rate": 2.9589991298832202e-05,
      "loss": 1.077,
      "num_input_tokens_seen": 9782248,
      "step": 17650
    },
    {
      "epoch": 309.7433628318584,
      "grad_norm": 9.5,
      "learning_rate": 2.958034035440513e-05,
      "loss": 1.0816,
      "num_input_tokens_seen": 9785432,
      "step": 17655
    },
    {
      "epoch": 309.83185840707966,
      "grad_norm": 9.6875,
      "learning_rate": 2.957068870363201e-05,
      "loss": 0.9344,
      "num_input_tokens_seen": 9788312,
      "step": 17660
    },
    {
      "epoch": 309.92035398230087,
      "grad_norm": 8.0,
      "learning_rate": 2.956103634800126e-05,
      "loss": 1.0973,
      "num_input_tokens_seen": 9791048,
      "step": 17665
    },
    {
      "epoch": 310.0,
      "grad_norm": 39.25,
      "learning_rate": 2.9551383289001384e-05,
      "loss": 1.1944,
      "num_input_tokens_seen": 9793536,
      "step": 17670
    },
    {
      "epoch": 310.08849557522126,
      "grad_norm": 12.625,
      "learning_rate": 2.9541729528121005e-05,
      "loss": 0.9966,
      "num_input_tokens_seen": 9795968,
      "step": 17675
    },
    {
      "epoch": 310.17699115044246,
      "grad_norm": 9.25,
      "learning_rate": 2.9532075066848856e-05,
      "loss": 1.1498,
      "num_input_tokens_seen": 9798784,
      "step": 17680
    },
    {
      "epoch": 310.2654867256637,
      "grad_norm": 8.25,
      "learning_rate": 2.9522419906673786e-05,
      "loss": 0.9774,
      "num_input_tokens_seen": 9801616,
      "step": 17685
    },
    {
      "epoch": 310.353982300885,
      "grad_norm": 8.625,
      "learning_rate": 2.951276404908474e-05,
      "loss": 0.9368,
      "num_input_tokens_seen": 9804896,
      "step": 17690
    },
    {
      "epoch": 310.4424778761062,
      "grad_norm": 14.0625,
      "learning_rate": 2.9503107495570752e-05,
      "loss": 1.0681,
      "num_input_tokens_seen": 9807808,
      "step": 17695
    },
    {
      "epoch": 310.53097345132744,
      "grad_norm": 10.125,
      "learning_rate": 2.9493450247621003e-05,
      "loss": 1.1606,
      "num_input_tokens_seen": 9810736,
      "step": 17700
    },
    {
      "epoch": 310.6194690265487,
      "grad_norm": 12.0625,
      "learning_rate": 2.948379230672476e-05,
      "loss": 1.0398,
      "num_input_tokens_seen": 9813248,
      "step": 17705
    },
    {
      "epoch": 310.7079646017699,
      "grad_norm": 7.46875,
      "learning_rate": 2.9474133674371396e-05,
      "loss": 1.0859,
      "num_input_tokens_seen": 9815904,
      "step": 17710
    },
    {
      "epoch": 310.79646017699116,
      "grad_norm": 6.96875,
      "learning_rate": 2.9464474352050387e-05,
      "loss": 1.1094,
      "num_input_tokens_seen": 9818640,
      "step": 17715
    },
    {
      "epoch": 310.88495575221236,
      "grad_norm": 8.0,
      "learning_rate": 2.9454814341251336e-05,
      "loss": 0.9691,
      "num_input_tokens_seen": 9821600,
      "step": 17720
    },
    {
      "epoch": 310.9734513274336,
      "grad_norm": 8.5625,
      "learning_rate": 2.9445153643463942e-05,
      "loss": 1.0831,
      "num_input_tokens_seen": 9824144,
      "step": 17725
    },
    {
      "epoch": 311.05309734513276,
      "grad_norm": 11.25,
      "learning_rate": 2.943549226017798e-05,
      "loss": 1.0571,
      "num_input_tokens_seen": 9826296,
      "step": 17730
    },
    {
      "epoch": 311.14159292035396,
      "grad_norm": 10.4375,
      "learning_rate": 2.942583019288337e-05,
      "loss": 1.1939,
      "num_input_tokens_seen": 9828920,
      "step": 17735
    },
    {
      "epoch": 311.2300884955752,
      "grad_norm": 8.5625,
      "learning_rate": 2.9416167443070132e-05,
      "loss": 0.9092,
      "num_input_tokens_seen": 9831864,
      "step": 17740
    },
    {
      "epoch": 311.3185840707965,
      "grad_norm": 8.625,
      "learning_rate": 2.9406504012228375e-05,
      "loss": 1.2318,
      "num_input_tokens_seen": 9834936,
      "step": 17745
    },
    {
      "epoch": 311.4070796460177,
      "grad_norm": 10.1875,
      "learning_rate": 2.939683990184832e-05,
      "loss": 1.1843,
      "num_input_tokens_seen": 9837736,
      "step": 17750
    },
    {
      "epoch": 311.49557522123894,
      "grad_norm": 11.125,
      "learning_rate": 2.93871751134203e-05,
      "loss": 0.9522,
      "num_input_tokens_seen": 9840744,
      "step": 17755
    },
    {
      "epoch": 311.5840707964602,
      "grad_norm": 8.25,
      "learning_rate": 2.9377509648434752e-05,
      "loss": 0.9942,
      "num_input_tokens_seen": 9843704,
      "step": 17760
    },
    {
      "epoch": 311.6725663716814,
      "grad_norm": 6.8125,
      "learning_rate": 2.9367843508382203e-05,
      "loss": 0.917,
      "num_input_tokens_seen": 9846536,
      "step": 17765
    },
    {
      "epoch": 311.76106194690266,
      "grad_norm": 7.8125,
      "learning_rate": 2.9358176694753293e-05,
      "loss": 0.9897,
      "num_input_tokens_seen": 9849016,
      "step": 17770
    },
    {
      "epoch": 311.8495575221239,
      "grad_norm": 7.40625,
      "learning_rate": 2.9348509209038766e-05,
      "loss": 1.025,
      "num_input_tokens_seen": 9852296,
      "step": 17775
    },
    {
      "epoch": 311.9380530973451,
      "grad_norm": 7.6875,
      "learning_rate": 2.933884105272947e-05,
      "loss": 1.1357,
      "num_input_tokens_seen": 9855064,
      "step": 17780
    },
    {
      "epoch": 312.01769911504425,
      "grad_norm": 8.0625,
      "learning_rate": 2.9329172227316366e-05,
      "loss": 1.0578,
      "num_input_tokens_seen": 9857296,
      "step": 17785
    },
    {
      "epoch": 312.1061946902655,
      "grad_norm": 7.65625,
      "learning_rate": 2.93195027342905e-05,
      "loss": 0.953,
      "num_input_tokens_seen": 9859696,
      "step": 17790
    },
    {
      "epoch": 312.1946902654867,
      "grad_norm": 9.0625,
      "learning_rate": 2.9309832575143024e-05,
      "loss": 0.9329,
      "num_input_tokens_seen": 9863136,
      "step": 17795
    },
    {
      "epoch": 312.283185840708,
      "grad_norm": 8.8125,
      "learning_rate": 2.930016175136521e-05,
      "loss": 0.9924,
      "num_input_tokens_seen": 9866256,
      "step": 17800
    },
    {
      "epoch": 312.283185840708,
      "eval_loss": 1.0314615964889526,
      "eval_runtime": 0.9332,
      "eval_samples_per_second": 26.79,
      "eval_steps_per_second": 13.931,
      "num_input_tokens_seen": 9866256,
      "step": 17800
    },
    {
      "epoch": 312.37168141592923,
      "grad_norm": 8.625,
      "learning_rate": 2.9290490264448412e-05,
      "loss": 1.3123,
      "num_input_tokens_seen": 9868704,
      "step": 17805
    },
    {
      "epoch": 312.46017699115043,
      "grad_norm": 8.6875,
      "learning_rate": 2.9280818115884094e-05,
      "loss": 1.0327,
      "num_input_tokens_seen": 9871168,
      "step": 17810
    },
    {
      "epoch": 312.5486725663717,
      "grad_norm": 13.875,
      "learning_rate": 2.9271145307163828e-05,
      "loss": 1.1079,
      "num_input_tokens_seen": 9874080,
      "step": 17815
    },
    {
      "epoch": 312.6371681415929,
      "grad_norm": 8.375,
      "learning_rate": 2.9261471839779287e-05,
      "loss": 1.0425,
      "num_input_tokens_seen": 9877104,
      "step": 17820
    },
    {
      "epoch": 312.72566371681415,
      "grad_norm": 7.96875,
      "learning_rate": 2.925179771522223e-05,
      "loss": 0.9889,
      "num_input_tokens_seen": 9879824,
      "step": 17825
    },
    {
      "epoch": 312.8141592920354,
      "grad_norm": 7.53125,
      "learning_rate": 2.9242122934984535e-05,
      "loss": 1.2172,
      "num_input_tokens_seen": 9882304,
      "step": 17830
    },
    {
      "epoch": 312.9026548672566,
      "grad_norm": 11.125,
      "learning_rate": 2.9232447500558176e-05,
      "loss": 1.0964,
      "num_input_tokens_seen": 9885488,
      "step": 17835
    },
    {
      "epoch": 312.9911504424779,
      "grad_norm": 6.71875,
      "learning_rate": 2.9222771413435225e-05,
      "loss": 0.9609,
      "num_input_tokens_seen": 9887936,
      "step": 17840
    },
    {
      "epoch": 313.070796460177,
      "grad_norm": 8.4375,
      "learning_rate": 2.9213094675107848e-05,
      "loss": 0.9013,
      "num_input_tokens_seen": 9890208,
      "step": 17845
    },
    {
      "epoch": 313.1592920353982,
      "grad_norm": 9.75,
      "learning_rate": 2.9203417287068335e-05,
      "loss": 0.9186,
      "num_input_tokens_seen": 9893232,
      "step": 17850
    },
    {
      "epoch": 313.24778761061947,
      "grad_norm": 8.875,
      "learning_rate": 2.9193739250809042e-05,
      "loss": 1.0892,
      "num_input_tokens_seen": 9896016,
      "step": 17855
    },
    {
      "epoch": 313.3362831858407,
      "grad_norm": 8.0625,
      "learning_rate": 2.9184060567822463e-05,
      "loss": 1.0635,
      "num_input_tokens_seen": 9898416,
      "step": 17860
    },
    {
      "epoch": 313.42477876106193,
      "grad_norm": 9.0625,
      "learning_rate": 2.9174381239601166e-05,
      "loss": 1.1437,
      "num_input_tokens_seen": 9900800,
      "step": 17865
    },
    {
      "epoch": 313.5132743362832,
      "grad_norm": 8.3125,
      "learning_rate": 2.916470126763783e-05,
      "loss": 1.115,
      "num_input_tokens_seen": 9903984,
      "step": 17870
    },
    {
      "epoch": 313.60176991150445,
      "grad_norm": 6.03125,
      "learning_rate": 2.9155020653425203e-05,
      "loss": 0.9656,
      "num_input_tokens_seen": 9906672,
      "step": 17875
    },
    {
      "epoch": 313.69026548672565,
      "grad_norm": 7.71875,
      "learning_rate": 2.9145339398456184e-05,
      "loss": 0.8937,
      "num_input_tokens_seen": 9908928,
      "step": 17880
    },
    {
      "epoch": 313.7787610619469,
      "grad_norm": 8.0,
      "learning_rate": 2.913565750422374e-05,
      "loss": 1.1263,
      "num_input_tokens_seen": 9911696,
      "step": 17885
    },
    {
      "epoch": 313.86725663716817,
      "grad_norm": 8.0625,
      "learning_rate": 2.9125974972220938e-05,
      "loss": 1.0792,
      "num_input_tokens_seen": 9914784,
      "step": 17890
    },
    {
      "epoch": 313.95575221238937,
      "grad_norm": 6.40625,
      "learning_rate": 2.9116291803940932e-05,
      "loss": 1.1287,
      "num_input_tokens_seen": 9917872,
      "step": 17895
    },
    {
      "epoch": 314.0353982300885,
      "grad_norm": 7.46875,
      "learning_rate": 2.910660800087701e-05,
      "loss": 1.1243,
      "num_input_tokens_seen": 9920584,
      "step": 17900
    },
    {
      "epoch": 314.12389380530976,
      "grad_norm": 10.0,
      "learning_rate": 2.909692356452254e-05,
      "loss": 1.1947,
      "num_input_tokens_seen": 9923288,
      "step": 17905
    },
    {
      "epoch": 314.21238938053096,
      "grad_norm": 7.65625,
      "learning_rate": 2.9087238496370962e-05,
      "loss": 1.0003,
      "num_input_tokens_seen": 9926296,
      "step": 17910
    },
    {
      "epoch": 314.3008849557522,
      "grad_norm": 7.5625,
      "learning_rate": 2.907755279791583e-05,
      "loss": 1.0906,
      "num_input_tokens_seen": 9928872,
      "step": 17915
    },
    {
      "epoch": 314.3893805309734,
      "grad_norm": 8.0625,
      "learning_rate": 2.906786647065083e-05,
      "loss": 0.9266,
      "num_input_tokens_seen": 9931272,
      "step": 17920
    },
    {
      "epoch": 314.4778761061947,
      "grad_norm": 7.71875,
      "learning_rate": 2.9058179516069695e-05,
      "loss": 1.0137,
      "num_input_tokens_seen": 9934760,
      "step": 17925
    },
    {
      "epoch": 314.56637168141594,
      "grad_norm": 9.875,
      "learning_rate": 2.9048491935666282e-05,
      "loss": 1.043,
      "num_input_tokens_seen": 9937832,
      "step": 17930
    },
    {
      "epoch": 314.65486725663715,
      "grad_norm": 7.59375,
      "learning_rate": 2.9038803730934534e-05,
      "loss": 0.9804,
      "num_input_tokens_seen": 9940568,
      "step": 17935
    },
    {
      "epoch": 314.7433628318584,
      "grad_norm": 9.6875,
      "learning_rate": 2.9029114903368503e-05,
      "loss": 1.1324,
      "num_input_tokens_seen": 9942840,
      "step": 17940
    },
    {
      "epoch": 314.83185840707966,
      "grad_norm": 8.25,
      "learning_rate": 2.9019425454462318e-05,
      "loss": 1.1248,
      "num_input_tokens_seen": 9945464,
      "step": 17945
    },
    {
      "epoch": 314.92035398230087,
      "grad_norm": 6.96875,
      "learning_rate": 2.9009735385710212e-05,
      "loss": 1.0001,
      "num_input_tokens_seen": 9948024,
      "step": 17950
    },
    {
      "epoch": 315.0,
      "grad_norm": 17.875,
      "learning_rate": 2.900004469860652e-05,
      "loss": 1.2495,
      "num_input_tokens_seen": 9950520,
      "step": 17955
    },
    {
      "epoch": 315.08849557522126,
      "grad_norm": 10.0,
      "learning_rate": 2.8990353394645668e-05,
      "loss": 1.0382,
      "num_input_tokens_seen": 9953448,
      "step": 17960
    },
    {
      "epoch": 315.17699115044246,
      "grad_norm": 8.6875,
      "learning_rate": 2.8980661475322186e-05,
      "loss": 1.0104,
      "num_input_tokens_seen": 9956568,
      "step": 17965
    },
    {
      "epoch": 315.2654867256637,
      "grad_norm": 7.71875,
      "learning_rate": 2.897096894213067e-05,
      "loss": 0.9964,
      "num_input_tokens_seen": 9959416,
      "step": 17970
    },
    {
      "epoch": 315.353982300885,
      "grad_norm": 11.125,
      "learning_rate": 2.8961275796565845e-05,
      "loss": 1.1783,
      "num_input_tokens_seen": 9962456,
      "step": 17975
    },
    {
      "epoch": 315.4424778761062,
      "grad_norm": 7.78125,
      "learning_rate": 2.8951582040122517e-05,
      "loss": 1.0132,
      "num_input_tokens_seen": 9965128,
      "step": 17980
    },
    {
      "epoch": 315.53097345132744,
      "grad_norm": 7.375,
      "learning_rate": 2.894188767429557e-05,
      "loss": 1.1861,
      "num_input_tokens_seen": 9967496,
      "step": 17985
    },
    {
      "epoch": 315.6194690265487,
      "grad_norm": 8.75,
      "learning_rate": 2.8932192700580014e-05,
      "loss": 0.968,
      "num_input_tokens_seen": 9970328,
      "step": 17990
    },
    {
      "epoch": 315.7079646017699,
      "grad_norm": 11.6875,
      "learning_rate": 2.8922497120470916e-05,
      "loss": 1.1832,
      "num_input_tokens_seen": 9972952,
      "step": 17995
    },
    {
      "epoch": 315.79646017699116,
      "grad_norm": 6.5625,
      "learning_rate": 2.891280093546348e-05,
      "loss": 0.8758,
      "num_input_tokens_seen": 9975768,
      "step": 18000
    },
    {
      "epoch": 315.79646017699116,
      "eval_loss": 1.0255134105682373,
      "eval_runtime": 0.934,
      "eval_samples_per_second": 26.767,
      "eval_steps_per_second": 13.919,
      "num_input_tokens_seen": 9975768,
      "step": 18000
    },
    {
      "epoch": 315.88495575221236,
      "grad_norm": 9.625,
      "learning_rate": 2.890310414705297e-05,
      "loss": 1.1042,
      "num_input_tokens_seen": 9978424,
      "step": 18005
    },
    {
      "epoch": 315.9734513274336,
      "grad_norm": 8.75,
      "learning_rate": 2.8893406756734742e-05,
      "loss": 1.0088,
      "num_input_tokens_seen": 9981496,
      "step": 18010
    },
    {
      "epoch": 316.05309734513276,
      "grad_norm": 8.0625,
      "learning_rate": 2.888370876600427e-05,
      "loss": 1.0035,
      "num_input_tokens_seen": 9983584,
      "step": 18015
    },
    {
      "epoch": 316.14159292035396,
      "grad_norm": 8.3125,
      "learning_rate": 2.8874010176357104e-05,
      "loss": 1.1174,
      "num_input_tokens_seen": 9986096,
      "step": 18020
    },
    {
      "epoch": 316.2300884955752,
      "grad_norm": 8.875,
      "learning_rate": 2.886431098928888e-05,
      "loss": 1.178,
      "num_input_tokens_seen": 9988576,
      "step": 18025
    },
    {
      "epoch": 316.3185840707965,
      "grad_norm": 8.9375,
      "learning_rate": 2.885461120629534e-05,
      "loss": 1.1467,
      "num_input_tokens_seen": 9991216,
      "step": 18030
    },
    {
      "epoch": 316.4070796460177,
      "grad_norm": 7.28125,
      "learning_rate": 2.8844910828872317e-05,
      "loss": 1.0631,
      "num_input_tokens_seen": 9994064,
      "step": 18035
    },
    {
      "epoch": 316.49557522123894,
      "grad_norm": 7.28125,
      "learning_rate": 2.8835209858515715e-05,
      "loss": 0.8351,
      "num_input_tokens_seen": 9996896,
      "step": 18040
    },
    {
      "epoch": 316.5840707964602,
      "grad_norm": 12.3125,
      "learning_rate": 2.8825508296721566e-05,
      "loss": 1.0411,
      "num_input_tokens_seen": 9999648,
      "step": 18045
    },
    {
      "epoch": 316.6725663716814,
      "grad_norm": 10.75,
      "learning_rate": 2.881580614498596e-05,
      "loss": 1.0093,
      "num_input_tokens_seen": 10002928,
      "step": 18050
    },
    {
      "epoch": 316.76106194690266,
      "grad_norm": 7.78125,
      "learning_rate": 2.8806103404805103e-05,
      "loss": 1.0937,
      "num_input_tokens_seen": 10005840,
      "step": 18055
    },
    {
      "epoch": 316.8495575221239,
      "grad_norm": 9.3125,
      "learning_rate": 2.8796400077675257e-05,
      "loss": 1.2026,
      "num_input_tokens_seen": 10009216,
      "step": 18060
    },
    {
      "epoch": 316.9380530973451,
      "grad_norm": 7.25,
      "learning_rate": 2.8786696165092812e-05,
      "loss": 0.9192,
      "num_input_tokens_seen": 10011680,
      "step": 18065
    },
    {
      "epoch": 317.01769911504425,
      "grad_norm": 8.4375,
      "learning_rate": 2.8776991668554236e-05,
      "loss": 1.1036,
      "num_input_tokens_seen": 10014320,
      "step": 18070
    },
    {
      "epoch": 317.1061946902655,
      "grad_norm": 7.46875,
      "learning_rate": 2.876728658955608e-05,
      "loss": 0.9459,
      "num_input_tokens_seen": 10016720,
      "step": 18075
    },
    {
      "epoch": 317.1946902654867,
      "grad_norm": 8.1875,
      "learning_rate": 2.8757580929594986e-05,
      "loss": 1.043,
      "num_input_tokens_seen": 10019296,
      "step": 18080
    },
    {
      "epoch": 317.283185840708,
      "grad_norm": 11.125,
      "learning_rate": 2.87478746901677e-05,
      "loss": 1.1507,
      "num_input_tokens_seen": 10022416,
      "step": 18085
    },
    {
      "epoch": 317.37168141592923,
      "grad_norm": 8.1875,
      "learning_rate": 2.873816787277103e-05,
      "loss": 1.0529,
      "num_input_tokens_seen": 10025136,
      "step": 18090
    },
    {
      "epoch": 317.46017699115043,
      "grad_norm": 7.5625,
      "learning_rate": 2.8728460478901903e-05,
      "loss": 1.2229,
      "num_input_tokens_seen": 10028144,
      "step": 18095
    },
    {
      "epoch": 317.5486725663717,
      "grad_norm": 8.75,
      "learning_rate": 2.8718752510057307e-05,
      "loss": 1.0497,
      "num_input_tokens_seen": 10030560,
      "step": 18100
    },
    {
      "epoch": 317.6371681415929,
      "grad_norm": 8.125,
      "learning_rate": 2.870904396773435e-05,
      "loss": 1.0653,
      "num_input_tokens_seen": 10033744,
      "step": 18105
    },
    {
      "epoch": 317.72566371681415,
      "grad_norm": 8.625,
      "learning_rate": 2.86993348534302e-05,
      "loss": 0.9717,
      "num_input_tokens_seen": 10036176,
      "step": 18110
    },
    {
      "epoch": 317.8141592920354,
      "grad_norm": 10.0,
      "learning_rate": 2.868962516864212e-05,
      "loss": 0.9717,
      "num_input_tokens_seen": 10039296,
      "step": 18115
    },
    {
      "epoch": 317.9026548672566,
      "grad_norm": 11.6875,
      "learning_rate": 2.8679914914867477e-05,
      "loss": 1.1541,
      "num_input_tokens_seen": 10042000,
      "step": 18120
    },
    {
      "epoch": 317.9911504424779,
      "grad_norm": 9.0625,
      "learning_rate": 2.8670204093603713e-05,
      "loss": 0.9938,
      "num_input_tokens_seen": 10044944,
      "step": 18125
    },
    {
      "epoch": 318.070796460177,
      "grad_norm": 9.125,
      "learning_rate": 2.8660492706348357e-05,
      "loss": 0.9865,
      "num_input_tokens_seen": 10047104,
      "step": 18130
    },
    {
      "epoch": 318.1592920353982,
      "grad_norm": 8.875,
      "learning_rate": 2.8650780754599022e-05,
      "loss": 0.882,
      "num_input_tokens_seen": 10049888,
      "step": 18135
    },
    {
      "epoch": 318.24778761061947,
      "grad_norm": 8.3125,
      "learning_rate": 2.8641068239853407e-05,
      "loss": 0.9583,
      "num_input_tokens_seen": 10052496,
      "step": 18140
    },
    {
      "epoch": 318.3362831858407,
      "grad_norm": 7.90625,
      "learning_rate": 2.863135516360932e-05,
      "loss": 1.2115,
      "num_input_tokens_seen": 10055392,
      "step": 18145
    },
    {
      "epoch": 318.42477876106193,
      "grad_norm": 7.46875,
      "learning_rate": 2.8621641527364633e-05,
      "loss": 1.1437,
      "num_input_tokens_seen": 10058784,
      "step": 18150
    },
    {
      "epoch": 318.5132743362832,
      "grad_norm": 11.6875,
      "learning_rate": 2.8611927332617313e-05,
      "loss": 1.2838,
      "num_input_tokens_seen": 10061376,
      "step": 18155
    },
    {
      "epoch": 318.60176991150445,
      "grad_norm": 7.21875,
      "learning_rate": 2.8602212580865405e-05,
      "loss": 1.1172,
      "num_input_tokens_seen": 10063968,
      "step": 18160
    },
    {
      "epoch": 318.69026548672565,
      "grad_norm": 6.84375,
      "learning_rate": 2.859249727360705e-05,
      "loss": 0.9445,
      "num_input_tokens_seen": 10066592,
      "step": 18165
    },
    {
      "epoch": 318.7787610619469,
      "grad_norm": 10.75,
      "learning_rate": 2.8582781412340465e-05,
      "loss": 1.0248,
      "num_input_tokens_seen": 10069648,
      "step": 18170
    },
    {
      "epoch": 318.86725663716817,
      "grad_norm": 7.90625,
      "learning_rate": 2.857306499856397e-05,
      "loss": 1.0122,
      "num_input_tokens_seen": 10072160,
      "step": 18175
    },
    {
      "epoch": 318.95575221238937,
      "grad_norm": 7.8125,
      "learning_rate": 2.856334803377594e-05,
      "loss": 1.0615,
      "num_input_tokens_seen": 10075024,
      "step": 18180
    },
    {
      "epoch": 319.0353982300885,
      "grad_norm": 8.3125,
      "learning_rate": 2.8553630519474867e-05,
      "loss": 1.1757,
      "num_input_tokens_seen": 10077672,
      "step": 18185
    },
    {
      "epoch": 319.12389380530976,
      "grad_norm": 10.0625,
      "learning_rate": 2.8543912457159317e-05,
      "loss": 1.1073,
      "num_input_tokens_seen": 10080472,
      "step": 18190
    },
    {
      "epoch": 319.21238938053096,
      "grad_norm": 9.5625,
      "learning_rate": 2.853419384832792e-05,
      "loss": 1.1513,
      "num_input_tokens_seen": 10083048,
      "step": 18195
    },
    {
      "epoch": 319.3008849557522,
      "grad_norm": 8.25,
      "learning_rate": 2.8524474694479423e-05,
      "loss": 0.9882,
      "num_input_tokens_seen": 10086392,
      "step": 18200
    },
    {
      "epoch": 319.3008849557522,
      "eval_loss": 1.0310791730880737,
      "eval_runtime": 0.9334,
      "eval_samples_per_second": 26.784,
      "eval_steps_per_second": 13.928,
      "num_input_tokens_seen": 10086392,
      "step": 18200
    },
    {
      "epoch": 319.3893805309734,
      "grad_norm": 9.5625,
      "learning_rate": 2.851475499711264e-05,
      "loss": 0.98,
      "num_input_tokens_seen": 10089128,
      "step": 18205
    },
    {
      "epoch": 319.4778761061947,
      "grad_norm": 9.375,
      "learning_rate": 2.8505034757726468e-05,
      "loss": 1.2909,
      "num_input_tokens_seen": 10091640,
      "step": 18210
    },
    {
      "epoch": 319.56637168141594,
      "grad_norm": 10.375,
      "learning_rate": 2.8495313977819886e-05,
      "loss": 1.0069,
      "num_input_tokens_seen": 10094136,
      "step": 18215
    },
    {
      "epoch": 319.65486725663715,
      "grad_norm": 6.625,
      "learning_rate": 2.8485592658891956e-05,
      "loss": 0.9636,
      "num_input_tokens_seen": 10096712,
      "step": 18220
    },
    {
      "epoch": 319.7433628318584,
      "grad_norm": 11.5625,
      "learning_rate": 2.8475870802441844e-05,
      "loss": 1.0223,
      "num_input_tokens_seen": 10099784,
      "step": 18225
    },
    {
      "epoch": 319.83185840707966,
      "grad_norm": 7.375,
      "learning_rate": 2.8466148409968774e-05,
      "loss": 1.0139,
      "num_input_tokens_seen": 10102600,
      "step": 18230
    },
    {
      "epoch": 319.92035398230087,
      "grad_norm": 8.6875,
      "learning_rate": 2.8456425482972067e-05,
      "loss": 1.0474,
      "num_input_tokens_seen": 10106056,
      "step": 18235
    },
    {
      "epoch": 320.0,
      "grad_norm": 15.1875,
      "learning_rate": 2.84467020229511e-05,
      "loss": 0.9193,
      "num_input_tokens_seen": 10108392,
      "step": 18240
    },
    {
      "epoch": 320.08849557522126,
      "grad_norm": 8.3125,
      "learning_rate": 2.8436978031405375e-05,
      "loss": 1.0832,
      "num_input_tokens_seen": 10111464,
      "step": 18245
    },
    {
      "epoch": 320.17699115044246,
      "grad_norm": 8.3125,
      "learning_rate": 2.842725350983445e-05,
      "loss": 1.0932,
      "num_input_tokens_seen": 10113816,
      "step": 18250
    },
    {
      "epoch": 320.2654867256637,
      "grad_norm": 6.28125,
      "learning_rate": 2.8417528459737957e-05,
      "loss": 0.9717,
      "num_input_tokens_seen": 10116904,
      "step": 18255
    },
    {
      "epoch": 320.353982300885,
      "grad_norm": 8.375,
      "learning_rate": 2.8407802882615624e-05,
      "loss": 0.9911,
      "num_input_tokens_seen": 10119656,
      "step": 18260
    },
    {
      "epoch": 320.4424778761062,
      "grad_norm": 9.4375,
      "learning_rate": 2.8398076779967277e-05,
      "loss": 1.2896,
      "num_input_tokens_seen": 10122104,
      "step": 18265
    },
    {
      "epoch": 320.53097345132744,
      "grad_norm": 6.6875,
      "learning_rate": 2.8388350153292774e-05,
      "loss": 1.0553,
      "num_input_tokens_seen": 10124888,
      "step": 18270
    },
    {
      "epoch": 320.6194690265487,
      "grad_norm": 9.0625,
      "learning_rate": 2.8378623004092103e-05,
      "loss": 1.196,
      "num_input_tokens_seen": 10127368,
      "step": 18275
    },
    {
      "epoch": 320.7079646017699,
      "grad_norm": 8.75,
      "learning_rate": 2.8368895333865302e-05,
      "loss": 1.0137,
      "num_input_tokens_seen": 10130280,
      "step": 18280
    },
    {
      "epoch": 320.79646017699116,
      "grad_norm": 7.28125,
      "learning_rate": 2.835916714411251e-05,
      "loss": 0.9254,
      "num_input_tokens_seen": 10132968,
      "step": 18285
    },
    {
      "epoch": 320.88495575221236,
      "grad_norm": 10.375,
      "learning_rate": 2.8349438436333926e-05,
      "loss": 0.9974,
      "num_input_tokens_seen": 10135880,
      "step": 18290
    },
    {
      "epoch": 320.9734513274336,
      "grad_norm": 8.6875,
      "learning_rate": 2.833970921202984e-05,
      "loss": 0.9384,
      "num_input_tokens_seen": 10138680,
      "step": 18295
    },
    {
      "epoch": 321.05309734513276,
      "grad_norm": 8.3125,
      "learning_rate": 2.8329979472700628e-05,
      "loss": 1.145,
      "num_input_tokens_seen": 10141232,
      "step": 18300
    },
    {
      "epoch": 321.14159292035396,
      "grad_norm": 8.625,
      "learning_rate": 2.832024921984674e-05,
      "loss": 0.9392,
      "num_input_tokens_seen": 10144160,
      "step": 18305
    },
    {
      "epoch": 321.2300884955752,
      "grad_norm": 8.6875,
      "learning_rate": 2.8310518454968693e-05,
      "loss": 0.974,
      "num_input_tokens_seen": 10147200,
      "step": 18310
    },
    {
      "epoch": 321.3185840707965,
      "grad_norm": 9.375,
      "learning_rate": 2.8300787179567095e-05,
      "loss": 0.9252,
      "num_input_tokens_seen": 10150112,
      "step": 18315
    },
    {
      "epoch": 321.4070796460177,
      "grad_norm": 7.21875,
      "learning_rate": 2.8291055395142636e-05,
      "loss": 1.1663,
      "num_input_tokens_seen": 10153024,
      "step": 18320
    },
    {
      "epoch": 321.49557522123894,
      "grad_norm": 9.0,
      "learning_rate": 2.8281323103196073e-05,
      "loss": 0.9042,
      "num_input_tokens_seen": 10155472,
      "step": 18325
    },
    {
      "epoch": 321.5840707964602,
      "grad_norm": 8.8125,
      "learning_rate": 2.8271590305228256e-05,
      "loss": 1.2689,
      "num_input_tokens_seen": 10158096,
      "step": 18330
    },
    {
      "epoch": 321.6725663716814,
      "grad_norm": 7.125,
      "learning_rate": 2.82618570027401e-05,
      "loss": 0.9588,
      "num_input_tokens_seen": 10161152,
      "step": 18335
    },
    {
      "epoch": 321.76106194690266,
      "grad_norm": 9.0,
      "learning_rate": 2.8252123197232604e-05,
      "loss": 1.1391,
      "num_input_tokens_seen": 10164096,
      "step": 18340
    },
    {
      "epoch": 321.8495575221239,
      "grad_norm": 7.25,
      "learning_rate": 2.8242388890206843e-05,
      "loss": 1.0193,
      "num_input_tokens_seen": 10167296,
      "step": 18345
    },
    {
      "epoch": 321.9380530973451,
      "grad_norm": 10.0,
      "learning_rate": 2.8232654083163967e-05,
      "loss": 1.208,
      "num_input_tokens_seen": 10169904,
      "step": 18350
    },
    {
      "epoch": 322.01769911504425,
      "grad_norm": 9.5625,
      "learning_rate": 2.822291877760521e-05,
      "loss": 1.1741,
      "num_input_tokens_seen": 10171752,
      "step": 18355
    },
    {
      "epoch": 322.1061946902655,
      "grad_norm": 7.59375,
      "learning_rate": 2.8213182975031864e-05,
      "loss": 1.1053,
      "num_input_tokens_seen": 10174680,
      "step": 18360
    },
    {
      "epoch": 322.1946902654867,
      "grad_norm": 8.375,
      "learning_rate": 2.8203446676945337e-05,
      "loss": 1.1157,
      "num_input_tokens_seen": 10177448,
      "step": 18365
    },
    {
      "epoch": 322.283185840708,
      "grad_norm": 13.4375,
      "learning_rate": 2.8193709884847075e-05,
      "loss": 0.9103,
      "num_input_tokens_seen": 10180408,
      "step": 18370
    },
    {
      "epoch": 322.37168141592923,
      "grad_norm": 8.5,
      "learning_rate": 2.8183972600238605e-05,
      "loss": 1.1746,
      "num_input_tokens_seen": 10183480,
      "step": 18375
    },
    {
      "epoch": 322.46017699115043,
      "grad_norm": 10.9375,
      "learning_rate": 2.817423482462156e-05,
      "loss": 1.2227,
      "num_input_tokens_seen": 10186600,
      "step": 18380
    },
    {
      "epoch": 322.5486725663717,
      "grad_norm": 10.0625,
      "learning_rate": 2.8164496559497605e-05,
      "loss": 1.0123,
      "num_input_tokens_seen": 10189384,
      "step": 18385
    },
    {
      "epoch": 322.6371681415929,
      "grad_norm": 11.0,
      "learning_rate": 2.815475780636852e-05,
      "loss": 1.1357,
      "num_input_tokens_seen": 10191736,
      "step": 18390
    },
    {
      "epoch": 322.72566371681415,
      "grad_norm": 5.875,
      "learning_rate": 2.814501856673613e-05,
      "loss": 0.9374,
      "num_input_tokens_seen": 10194232,
      "step": 18395
    },
    {
      "epoch": 322.8141592920354,
      "grad_norm": 10.3125,
      "learning_rate": 2.8135278842102353e-05,
      "loss": 1.0167,
      "num_input_tokens_seen": 10197432,
      "step": 18400
    },
    {
      "epoch": 322.8141592920354,
      "eval_loss": 1.033027172088623,
      "eval_runtime": 0.9382,
      "eval_samples_per_second": 26.647,
      "eval_steps_per_second": 13.857,
      "num_input_tokens_seen": 10197432,
      "step": 18400
    },
    {
      "epoch": 322.9026548672566,
      "grad_norm": 6.40625,
      "learning_rate": 2.8125538633969183e-05,
      "loss": 0.9095,
      "num_input_tokens_seen": 10199896,
      "step": 18405
    },
    {
      "epoch": 322.9911504424779,
      "grad_norm": 6.5625,
      "learning_rate": 2.8115797943838677e-05,
      "loss": 0.957,
      "num_input_tokens_seen": 10202616,
      "step": 18410
    },
    {
      "epoch": 323.070796460177,
      "grad_norm": 11.375,
      "learning_rate": 2.810605677321298e-05,
      "loss": 1.0465,
      "num_input_tokens_seen": 10204904,
      "step": 18415
    },
    {
      "epoch": 323.1592920353982,
      "grad_norm": 6.0,
      "learning_rate": 2.809631512359428e-05,
      "loss": 0.8963,
      "num_input_tokens_seen": 10207944,
      "step": 18420
    },
    {
      "epoch": 323.24778761061947,
      "grad_norm": 9.5,
      "learning_rate": 2.8086572996484884e-05,
      "loss": 0.9648,
      "num_input_tokens_seen": 10210664,
      "step": 18425
    },
    {
      "epoch": 323.3362831858407,
      "grad_norm": 7.90625,
      "learning_rate": 2.8076830393387143e-05,
      "loss": 0.9527,
      "num_input_tokens_seen": 10213832,
      "step": 18430
    },
    {
      "epoch": 323.42477876106193,
      "grad_norm": 8.4375,
      "learning_rate": 2.8067087315803497e-05,
      "loss": 0.9823,
      "num_input_tokens_seen": 10216376,
      "step": 18435
    },
    {
      "epoch": 323.5132743362832,
      "grad_norm": 9.375,
      "learning_rate": 2.8057343765236433e-05,
      "loss": 1.0432,
      "num_input_tokens_seen": 10219480,
      "step": 18440
    },
    {
      "epoch": 323.60176991150445,
      "grad_norm": 7.90625,
      "learning_rate": 2.804759974318854e-05,
      "loss": 1.1854,
      "num_input_tokens_seen": 10222312,
      "step": 18445
    },
    {
      "epoch": 323.69026548672565,
      "grad_norm": 11.0625,
      "learning_rate": 2.8037855251162482e-05,
      "loss": 1.2225,
      "num_input_tokens_seen": 10225016,
      "step": 18450
    },
    {
      "epoch": 323.7787610619469,
      "grad_norm": 8.125,
      "learning_rate": 2.802811029066096e-05,
      "loss": 1.2132,
      "num_input_tokens_seen": 10227896,
      "step": 18455
    },
    {
      "epoch": 323.86725663716817,
      "grad_norm": 7.5,
      "learning_rate": 2.8018364863186764e-05,
      "loss": 0.8975,
      "num_input_tokens_seen": 10230600,
      "step": 18460
    },
    {
      "epoch": 323.95575221238937,
      "grad_norm": 7.375,
      "learning_rate": 2.800861897024279e-05,
      "loss": 1.2089,
      "num_input_tokens_seen": 10233176,
      "step": 18465
    },
    {
      "epoch": 324.0353982300885,
      "grad_norm": 10.1875,
      "learning_rate": 2.799887261333196e-05,
      "loss": 1.2354,
      "num_input_tokens_seen": 10235416,
      "step": 18470
    },
    {
      "epoch": 324.12389380530976,
      "grad_norm": 10.3125,
      "learning_rate": 2.798912579395728e-05,
      "loss": 1.152,
      "num_input_tokens_seen": 10238120,
      "step": 18475
    },
    {
      "epoch": 324.21238938053096,
      "grad_norm": 9.0,
      "learning_rate": 2.797937851362185e-05,
      "loss": 0.8656,
      "num_input_tokens_seen": 10240920,
      "step": 18480
    },
    {
      "epoch": 324.3008849557522,
      "grad_norm": 8.5625,
      "learning_rate": 2.7969630773828802e-05,
      "loss": 0.8742,
      "num_input_tokens_seen": 10243512,
      "step": 18485
    },
    {
      "epoch": 324.3893805309734,
      "grad_norm": 6.8125,
      "learning_rate": 2.7959882576081382e-05,
      "loss": 0.9314,
      "num_input_tokens_seen": 10246584,
      "step": 18490
    },
    {
      "epoch": 324.4778761061947,
      "grad_norm": 7.46875,
      "learning_rate": 2.795013392188286e-05,
      "loss": 1.0255,
      "num_input_tokens_seen": 10249368,
      "step": 18495
    },
    {
      "epoch": 324.56637168141594,
      "grad_norm": 10.5,
      "learning_rate": 2.7940384812736614e-05,
      "loss": 1.1422,
      "num_input_tokens_seen": 10252184,
      "step": 18500
    },
    {
      "epoch": 324.65486725663715,
      "grad_norm": 6.78125,
      "learning_rate": 2.7930635250146087e-05,
      "loss": 1.1634,
      "num_input_tokens_seen": 10254808,
      "step": 18505
    },
    {
      "epoch": 324.7433628318584,
      "grad_norm": 9.0625,
      "learning_rate": 2.792088523561477e-05,
      "loss": 1.1867,
      "num_input_tokens_seen": 10257448,
      "step": 18510
    },
    {
      "epoch": 324.83185840707966,
      "grad_norm": 7.9375,
      "learning_rate": 2.7911134770646246e-05,
      "loss": 1.0938,
      "num_input_tokens_seen": 10259864,
      "step": 18515
    },
    {
      "epoch": 324.92035398230087,
      "grad_norm": 7.84375,
      "learning_rate": 2.7901383856744157e-05,
      "loss": 0.9879,
      "num_input_tokens_seen": 10262968,
      "step": 18520
    },
    {
      "epoch": 325.0,
      "grad_norm": 13.25,
      "learning_rate": 2.7891632495412217e-05,
      "loss": 1.0475,
      "num_input_tokens_seen": 10265552,
      "step": 18525
    },
    {
      "epoch": 325.08849557522126,
      "grad_norm": 11.3125,
      "learning_rate": 2.7881880688154205e-05,
      "loss": 1.0475,
      "num_input_tokens_seen": 10268560,
      "step": 18530
    },
    {
      "epoch": 325.17699115044246,
      "grad_norm": 8.1875,
      "learning_rate": 2.7872128436473977e-05,
      "loss": 1.0405,
      "num_input_tokens_seen": 10271120,
      "step": 18535
    },
    {
      "epoch": 325.2654867256637,
      "grad_norm": 12.0,
      "learning_rate": 2.7862375741875448e-05,
      "loss": 1.3835,
      "num_input_tokens_seen": 10273568,
      "step": 18540
    },
    {
      "epoch": 325.353982300885,
      "grad_norm": 7.375,
      "learning_rate": 2.785262260586261e-05,
      "loss": 1.1001,
      "num_input_tokens_seen": 10276672,
      "step": 18545
    },
    {
      "epoch": 325.4424778761062,
      "grad_norm": 8.0625,
      "learning_rate": 2.7842869029939517e-05,
      "loss": 0.9832,
      "num_input_tokens_seen": 10279344,
      "step": 18550
    },
    {
      "epoch": 325.53097345132744,
      "grad_norm": 8.5,
      "learning_rate": 2.7833115015610296e-05,
      "loss": 1.0457,
      "num_input_tokens_seen": 10282240,
      "step": 18555
    },
    {
      "epoch": 325.6194690265487,
      "grad_norm": 8.4375,
      "learning_rate": 2.7823360564379136e-05,
      "loss": 0.9437,
      "num_input_tokens_seen": 10285232,
      "step": 18560
    },
    {
      "epoch": 325.7079646017699,
      "grad_norm": 7.84375,
      "learning_rate": 2.7813605677750297e-05,
      "loss": 0.9637,
      "num_input_tokens_seen": 10288384,
      "step": 18565
    },
    {
      "epoch": 325.79646017699116,
      "grad_norm": 10.125,
      "learning_rate": 2.7803850357228102e-05,
      "loss": 1.0073,
      "num_input_tokens_seen": 10291392,
      "step": 18570
    },
    {
      "epoch": 325.88495575221236,
      "grad_norm": 7.875,
      "learning_rate": 2.779409460431695e-05,
      "loss": 1.065,
      "num_input_tokens_seen": 10294160,
      "step": 18575
    },
    {
      "epoch": 325.9734513274336,
      "grad_norm": 8.375,
      "learning_rate": 2.778433842052129e-05,
      "loss": 1.0541,
      "num_input_tokens_seen": 10296560,
      "step": 18580
    },
    {
      "epoch": 326.05309734513276,
      "grad_norm": 7.6875,
      "learning_rate": 2.7774581807345664e-05,
      "loss": 1.1126,
      "num_input_tokens_seen": 10298712,
      "step": 18585
    },
    {
      "epoch": 326.14159292035396,
      "grad_norm": 8.0,
      "learning_rate": 2.776482476629465e-05,
      "loss": 1.2431,
      "num_input_tokens_seen": 10301528,
      "step": 18590
    },
    {
      "epoch": 326.2300884955752,
      "grad_norm": 8.1875,
      "learning_rate": 2.7755067298872924e-05,
      "loss": 1.1212,
      "num_input_tokens_seen": 10303976,
      "step": 18595
    },
    {
      "epoch": 326.3185840707965,
      "grad_norm": 8.25,
      "learning_rate": 2.774530940658518e-05,
      "loss": 0.849,
      "num_input_tokens_seen": 10307224,
      "step": 18600
    },
    {
      "epoch": 326.3185840707965,
      "eval_loss": 1.0195845365524292,
      "eval_runtime": 0.9341,
      "eval_samples_per_second": 26.764,
      "eval_steps_per_second": 13.917,
      "num_input_tokens_seen": 10307224,
      "step": 18600
    },
    {
      "epoch": 326.4070796460177,
      "grad_norm": 6.28125,
      "learning_rate": 2.7735551090936236e-05,
      "loss": 1.0216,
      "num_input_tokens_seen": 10310136,
      "step": 18605
    },
    {
      "epoch": 326.49557522123894,
      "grad_norm": 6.875,
      "learning_rate": 2.7725792353430934e-05,
      "loss": 1.0499,
      "num_input_tokens_seen": 10312968,
      "step": 18610
    },
    {
      "epoch": 326.5840707964602,
      "grad_norm": 10.125,
      "learning_rate": 2.77160331955742e-05,
      "loss": 1.0173,
      "num_input_tokens_seen": 10315864,
      "step": 18615
    },
    {
      "epoch": 326.6725663716814,
      "grad_norm": 9.5,
      "learning_rate": 2.7706273618871008e-05,
      "loss": 1.2084,
      "num_input_tokens_seen": 10318488,
      "step": 18620
    },
    {
      "epoch": 326.76106194690266,
      "grad_norm": 8.5,
      "learning_rate": 2.769651362482642e-05,
      "loss": 0.9971,
      "num_input_tokens_seen": 10321480,
      "step": 18625
    },
    {
      "epoch": 326.8495575221239,
      "grad_norm": 7.21875,
      "learning_rate": 2.768675321494555e-05,
      "loss": 0.8995,
      "num_input_tokens_seen": 10324344,
      "step": 18630
    },
    {
      "epoch": 326.9380530973451,
      "grad_norm": 8.875,
      "learning_rate": 2.7676992390733565e-05,
      "loss": 1.1134,
      "num_input_tokens_seen": 10326808,
      "step": 18635
    },
    {
      "epoch": 327.01769911504425,
      "grad_norm": 7.34375,
      "learning_rate": 2.766723115369571e-05,
      "loss": 1.0464,
      "num_input_tokens_seen": 10329528,
      "step": 18640
    },
    {
      "epoch": 327.1061946902655,
      "grad_norm": 9.1875,
      "learning_rate": 2.765746950533729e-05,
      "loss": 1.1925,
      "num_input_tokens_seen": 10332232,
      "step": 18645
    },
    {
      "epoch": 327.1946902654867,
      "grad_norm": 10.5,
      "learning_rate": 2.7647707447163684e-05,
      "loss": 1.1434,
      "num_input_tokens_seen": 10335592,
      "step": 18650
    },
    {
      "epoch": 327.283185840708,
      "grad_norm": 8.4375,
      "learning_rate": 2.7637944980680315e-05,
      "loss": 0.9891,
      "num_input_tokens_seen": 10338552,
      "step": 18655
    },
    {
      "epoch": 327.37168141592923,
      "grad_norm": 11.0,
      "learning_rate": 2.762818210739268e-05,
      "loss": 0.9961,
      "num_input_tokens_seen": 10341032,
      "step": 18660
    },
    {
      "epoch": 327.46017699115043,
      "grad_norm": 8.875,
      "learning_rate": 2.7618418828806332e-05,
      "loss": 1.0415,
      "num_input_tokens_seen": 10343832,
      "step": 18665
    },
    {
      "epoch": 327.5486725663717,
      "grad_norm": 6.96875,
      "learning_rate": 2.76086551464269e-05,
      "loss": 1.0667,
      "num_input_tokens_seen": 10346600,
      "step": 18670
    },
    {
      "epoch": 327.6371681415929,
      "grad_norm": 8.625,
      "learning_rate": 2.759889106176006e-05,
      "loss": 0.9939,
      "num_input_tokens_seen": 10349160,
      "step": 18675
    },
    {
      "epoch": 327.72566371681415,
      "grad_norm": 9.875,
      "learning_rate": 2.758912657631156e-05,
      "loss": 1.0757,
      "num_input_tokens_seen": 10352104,
      "step": 18680
    },
    {
      "epoch": 327.8141592920354,
      "grad_norm": 7.90625,
      "learning_rate": 2.7579361691587198e-05,
      "loss": 1.2531,
      "num_input_tokens_seen": 10354568,
      "step": 18685
    },
    {
      "epoch": 327.9026548672566,
      "grad_norm": 6.28125,
      "learning_rate": 2.756959640909285e-05,
      "loss": 0.9867,
      "num_input_tokens_seen": 10357480,
      "step": 18690
    },
    {
      "epoch": 327.9911504424779,
      "grad_norm": 11.625,
      "learning_rate": 2.7559830730334452e-05,
      "loss": 0.9157,
      "num_input_tokens_seen": 10360776,
      "step": 18695
    },
    {
      "epoch": 328.070796460177,
      "grad_norm": 8.875,
      "learning_rate": 2.7550064656817988e-05,
      "loss": 1.0437,
      "num_input_tokens_seen": 10362728,
      "step": 18700
    },
    {
      "epoch": 328.1592920353982,
      "grad_norm": 8.5625,
      "learning_rate": 2.7540298190049503e-05,
      "loss": 1.1556,
      "num_input_tokens_seen": 10365304,
      "step": 18705
    },
    {
      "epoch": 328.24778761061947,
      "grad_norm": 7.3125,
      "learning_rate": 2.7530531331535107e-05,
      "loss": 0.8751,
      "num_input_tokens_seen": 10368200,
      "step": 18710
    },
    {
      "epoch": 328.3362831858407,
      "grad_norm": 7.9375,
      "learning_rate": 2.752076408278099e-05,
      "loss": 1.0407,
      "num_input_tokens_seen": 10371688,
      "step": 18715
    },
    {
      "epoch": 328.42477876106193,
      "grad_norm": 8.625,
      "learning_rate": 2.751099644529337e-05,
      "loss": 1.199,
      "num_input_tokens_seen": 10374488,
      "step": 18720
    },
    {
      "epoch": 328.5132743362832,
      "grad_norm": 7.9375,
      "learning_rate": 2.7501228420578533e-05,
      "loss": 1.024,
      "num_input_tokens_seen": 10377192,
      "step": 18725
    },
    {
      "epoch": 328.60176991150445,
      "grad_norm": 6.46875,
      "learning_rate": 2.7491460010142857e-05,
      "loss": 1.0363,
      "num_input_tokens_seen": 10379912,
      "step": 18730
    },
    {
      "epoch": 328.69026548672565,
      "grad_norm": 7.3125,
      "learning_rate": 2.7481691215492727e-05,
      "loss": 1.032,
      "num_input_tokens_seen": 10382648,
      "step": 18735
    },
    {
      "epoch": 328.7787610619469,
      "grad_norm": 9.5625,
      "learning_rate": 2.747192203813463e-05,
      "loss": 0.9909,
      "num_input_tokens_seen": 10385656,
      "step": 18740
    },
    {
      "epoch": 328.86725663716817,
      "grad_norm": 7.96875,
      "learning_rate": 2.7462152479575087e-05,
      "loss": 1.0846,
      "num_input_tokens_seen": 10388344,
      "step": 18745
    },
    {
      "epoch": 328.95575221238937,
      "grad_norm": 8.9375,
      "learning_rate": 2.7452382541320697e-05,
      "loss": 1.093,
      "num_input_tokens_seen": 10391240,
      "step": 18750
    },
    {
      "epoch": 329.0353982300885,
      "grad_norm": 8.0625,
      "learning_rate": 2.7442612224878096e-05,
      "loss": 0.9049,
      "num_input_tokens_seen": 10393784,
      "step": 18755
    },
    {
      "epoch": 329.12389380530976,
      "grad_norm": 8.9375,
      "learning_rate": 2.7432841531753994e-05,
      "loss": 1.0554,
      "num_input_tokens_seen": 10396504,
      "step": 18760
    },
    {
      "epoch": 329.21238938053096,
      "grad_norm": 7.71875,
      "learning_rate": 2.7423070463455147e-05,
      "loss": 1.1812,
      "num_input_tokens_seen": 10399272,
      "step": 18765
    },
    {
      "epoch": 329.3008849557522,
      "grad_norm": 8.625,
      "learning_rate": 2.7413299021488397e-05,
      "loss": 0.9487,
      "num_input_tokens_seen": 10401912,
      "step": 18770
    },
    {
      "epoch": 329.3893805309734,
      "grad_norm": 6.65625,
      "learning_rate": 2.7403527207360615e-05,
      "loss": 0.9276,
      "num_input_tokens_seen": 10404728,
      "step": 18775
    },
    {
      "epoch": 329.4778761061947,
      "grad_norm": 7.4375,
      "learning_rate": 2.7393755022578722e-05,
      "loss": 1.0383,
      "num_input_tokens_seen": 10407496,
      "step": 18780
    },
    {
      "epoch": 329.56637168141594,
      "grad_norm": 11.0,
      "learning_rate": 2.7383982468649714e-05,
      "loss": 0.9546,
      "num_input_tokens_seen": 10410600,
      "step": 18785
    },
    {
      "epoch": 329.65486725663715,
      "grad_norm": 8.125,
      "learning_rate": 2.7374209547080665e-05,
      "loss": 1.1021,
      "num_input_tokens_seen": 10413752,
      "step": 18790
    },
    {
      "epoch": 329.7433628318584,
      "grad_norm": 9.25,
      "learning_rate": 2.7364436259378663e-05,
      "loss": 1.11,
      "num_input_tokens_seen": 10416312,
      "step": 18795
    },
    {
      "epoch": 329.83185840707966,
      "grad_norm": 9.9375,
      "learning_rate": 2.735466260705088e-05,
      "loss": 0.9223,
      "num_input_tokens_seen": 10419256,
      "step": 18800
    },
    {
      "epoch": 329.83185840707966,
      "eval_loss": 1.0274866819381714,
      "eval_runtime": 0.9397,
      "eval_samples_per_second": 26.604,
      "eval_steps_per_second": 13.834,
      "num_input_tokens_seen": 10419256,
      "step": 18800
    },
    {
      "epoch": 329.92035398230087,
      "grad_norm": 10.5,
      "learning_rate": 2.7344888591604524e-05,
      "loss": 1.2907,
      "num_input_tokens_seen": 10422296,
      "step": 18805
    },
    {
      "epoch": 330.0,
      "grad_norm": 35.0,
      "learning_rate": 2.7335114214546893e-05,
      "loss": 1.0517,
      "num_input_tokens_seen": 10424464,
      "step": 18810
    },
    {
      "epoch": 330.08849557522126,
      "grad_norm": 8.3125,
      "learning_rate": 2.7325339477385293e-05,
      "loss": 0.9253,
      "num_input_tokens_seen": 10427200,
      "step": 18815
    },
    {
      "epoch": 330.17699115044246,
      "grad_norm": 6.15625,
      "learning_rate": 2.7315564381627128e-05,
      "loss": 1.0536,
      "num_input_tokens_seen": 10430176,
      "step": 18820
    },
    {
      "epoch": 330.2654867256637,
      "grad_norm": 8.4375,
      "learning_rate": 2.7305788928779835e-05,
      "loss": 1.064,
      "num_input_tokens_seen": 10432800,
      "step": 18825
    },
    {
      "epoch": 330.353982300885,
      "grad_norm": 7.125,
      "learning_rate": 2.729601312035091e-05,
      "loss": 1.146,
      "num_input_tokens_seen": 10435680,
      "step": 18830
    },
    {
      "epoch": 330.4424778761062,
      "grad_norm": 7.75,
      "learning_rate": 2.7286236957847915e-05,
      "loss": 1.2711,
      "num_input_tokens_seen": 10438192,
      "step": 18835
    },
    {
      "epoch": 330.53097345132744,
      "grad_norm": 8.375,
      "learning_rate": 2.7276460442778446e-05,
      "loss": 0.9406,
      "num_input_tokens_seen": 10440464,
      "step": 18840
    },
    {
      "epoch": 330.6194690265487,
      "grad_norm": 9.4375,
      "learning_rate": 2.726668357665017e-05,
      "loss": 1.0871,
      "num_input_tokens_seen": 10443504,
      "step": 18845
    },
    {
      "epoch": 330.7079646017699,
      "grad_norm": 9.125,
      "learning_rate": 2.7256906360970808e-05,
      "loss": 1.2025,
      "num_input_tokens_seen": 10446464,
      "step": 18850
    },
    {
      "epoch": 330.79646017699116,
      "grad_norm": 7.84375,
      "learning_rate": 2.7247128797248117e-05,
      "loss": 0.9858,
      "num_input_tokens_seen": 10449456,
      "step": 18855
    },
    {
      "epoch": 330.88495575221236,
      "grad_norm": 9.0625,
      "learning_rate": 2.7237350886989925e-05,
      "loss": 0.8731,
      "num_input_tokens_seen": 10451888,
      "step": 18860
    },
    {
      "epoch": 330.9734513274336,
      "grad_norm": 10.0,
      "learning_rate": 2.7227572631704107e-05,
      "loss": 1.0624,
      "num_input_tokens_seen": 10455024,
      "step": 18865
    },
    {
      "epoch": 331.05309734513276,
      "grad_norm": 8.0625,
      "learning_rate": 2.7217794032898596e-05,
      "loss": 1.091,
      "num_input_tokens_seen": 10457448,
      "step": 18870
    },
    {
      "epoch": 331.14159292035396,
      "grad_norm": 11.5,
      "learning_rate": 2.7208015092081384e-05,
      "loss": 0.9844,
      "num_input_tokens_seen": 10460344,
      "step": 18875
    },
    {
      "epoch": 331.2300884955752,
      "grad_norm": 9.9375,
      "learning_rate": 2.719823581076049e-05,
      "loss": 1.1647,
      "num_input_tokens_seen": 10462936,
      "step": 18880
    },
    {
      "epoch": 331.3185840707965,
      "grad_norm": 5.875,
      "learning_rate": 2.718845619044401e-05,
      "loss": 1.0572,
      "num_input_tokens_seen": 10466008,
      "step": 18885
    },
    {
      "epoch": 331.4070796460177,
      "grad_norm": 8.375,
      "learning_rate": 2.7178676232640088e-05,
      "loss": 1.0076,
      "num_input_tokens_seen": 10468824,
      "step": 18890
    },
    {
      "epoch": 331.49557522123894,
      "grad_norm": 7.65625,
      "learning_rate": 2.716889593885691e-05,
      "loss": 0.9887,
      "num_input_tokens_seen": 10472376,
      "step": 18895
    },
    {
      "epoch": 331.5840707964602,
      "grad_norm": 9.4375,
      "learning_rate": 2.7159115310602716e-05,
      "loss": 1.252,
      "num_input_tokens_seen": 10474984,
      "step": 18900
    },
    {
      "epoch": 331.6725663716814,
      "grad_norm": 8.125,
      "learning_rate": 2.7149334349385814e-05,
      "loss": 1.3462,
      "num_input_tokens_seen": 10477816,
      "step": 18905
    },
    {
      "epoch": 331.76106194690266,
      "grad_norm": 6.8125,
      "learning_rate": 2.713955305671454e-05,
      "loss": 0.9957,
      "num_input_tokens_seen": 10480376,
      "step": 18910
    },
    {
      "epoch": 331.8495575221239,
      "grad_norm": 10.125,
      "learning_rate": 2.71297714340973e-05,
      "loss": 0.8338,
      "num_input_tokens_seen": 10483112,
      "step": 18915
    },
    {
      "epoch": 331.9380530973451,
      "grad_norm": 10.625,
      "learning_rate": 2.7119989483042545e-05,
      "loss": 0.9884,
      "num_input_tokens_seen": 10485448,
      "step": 18920
    },
    {
      "epoch": 332.01769911504425,
      "grad_norm": 10.0625,
      "learning_rate": 2.7110207205058768e-05,
      "loss": 1.0637,
      "num_input_tokens_seen": 10487656,
      "step": 18925
    },
    {
      "epoch": 332.1061946902655,
      "grad_norm": 7.3125,
      "learning_rate": 2.7100424601654517e-05,
      "loss": 1.1028,
      "num_input_tokens_seen": 10490024,
      "step": 18930
    },
    {
      "epoch": 332.1946902654867,
      "grad_norm": 9.0625,
      "learning_rate": 2.7090641674338403e-05,
      "loss": 0.9758,
      "num_input_tokens_seen": 10493064,
      "step": 18935
    },
    {
      "epoch": 332.283185840708,
      "grad_norm": 12.5625,
      "learning_rate": 2.7080858424619072e-05,
      "loss": 1.0996,
      "num_input_tokens_seen": 10496088,
      "step": 18940
    },
    {
      "epoch": 332.37168141592923,
      "grad_norm": 8.1875,
      "learning_rate": 2.707107485400521e-05,
      "loss": 0.8885,
      "num_input_tokens_seen": 10498888,
      "step": 18945
    },
    {
      "epoch": 332.46017699115043,
      "grad_norm": 8.125,
      "learning_rate": 2.7061290964005586e-05,
      "loss": 1.1634,
      "num_input_tokens_seen": 10501544,
      "step": 18950
    },
    {
      "epoch": 332.5486725663717,
      "grad_norm": 7.34375,
      "learning_rate": 2.7051506756129e-05,
      "loss": 1.0683,
      "num_input_tokens_seen": 10504200,
      "step": 18955
    },
    {
      "epoch": 332.6371681415929,
      "grad_norm": 6.71875,
      "learning_rate": 2.704172223188428e-05,
      "loss": 1.0343,
      "num_input_tokens_seen": 10507032,
      "step": 18960
    },
    {
      "epoch": 332.72566371681415,
      "grad_norm": 6.8125,
      "learning_rate": 2.7031937392780334e-05,
      "loss": 1.0911,
      "num_input_tokens_seen": 10510232,
      "step": 18965
    },
    {
      "epoch": 332.8141592920354,
      "grad_norm": 7.90625,
      "learning_rate": 2.702215224032611e-05,
      "loss": 1.093,
      "num_input_tokens_seen": 10513032,
      "step": 18970
    },
    {
      "epoch": 332.9026548672566,
      "grad_norm": 11.1875,
      "learning_rate": 2.70123667760306e-05,
      "loss": 1.036,
      "num_input_tokens_seen": 10515688,
      "step": 18975
    },
    {
      "epoch": 332.9911504424779,
      "grad_norm": 9.6875,
      "learning_rate": 2.7002581001402845e-05,
      "loss": 0.9057,
      "num_input_tokens_seen": 10518504,
      "step": 18980
    },
    {
      "epoch": 333.070796460177,
      "grad_norm": 7.53125,
      "learning_rate": 2.6992794917951923e-05,
      "loss": 1.3952,
      "num_input_tokens_seen": 10520784,
      "step": 18985
    },
    {
      "epoch": 333.1592920353982,
      "grad_norm": 8.875,
      "learning_rate": 2.6983008527187e-05,
      "loss": 0.9835,
      "num_input_tokens_seen": 10523520,
      "step": 18990
    },
    {
      "epoch": 333.24778761061947,
      "grad_norm": 9.875,
      "learning_rate": 2.697322183061723e-05,
      "loss": 1.2079,
      "num_input_tokens_seen": 10526704,
      "step": 18995
    },
    {
      "epoch": 333.3362831858407,
      "grad_norm": 12.125,
      "learning_rate": 2.696343482975186e-05,
      "loss": 0.9825,
      "num_input_tokens_seen": 10529488,
      "step": 19000
    },
    {
      "epoch": 333.3362831858407,
      "eval_loss": 1.025775671005249,
      "eval_runtime": 0.9374,
      "eval_samples_per_second": 26.668,
      "eval_steps_per_second": 13.868,
      "num_input_tokens_seen": 10529488,
      "step": 19000
    },
    {
      "epoch": 333.42477876106193,
      "grad_norm": 7.65625,
      "learning_rate": 2.695364752610016e-05,
      "loss": 1.0472,
      "num_input_tokens_seen": 10532240,
      "step": 19005
    },
    {
      "epoch": 333.5132743362832,
      "grad_norm": 7.46875,
      "learning_rate": 2.6943859921171467e-05,
      "loss": 1.06,
      "num_input_tokens_seen": 10535312,
      "step": 19010
    },
    {
      "epoch": 333.60176991150445,
      "grad_norm": 8.0625,
      "learning_rate": 2.6934072016475143e-05,
      "loss": 1.1846,
      "num_input_tokens_seen": 10537872,
      "step": 19015
    },
    {
      "epoch": 333.69026548672565,
      "grad_norm": 8.5625,
      "learning_rate": 2.6924283813520606e-05,
      "loss": 1.2073,
      "num_input_tokens_seen": 10540208,
      "step": 19020
    },
    {
      "epoch": 333.7787610619469,
      "grad_norm": 9.125,
      "learning_rate": 2.691449531381733e-05,
      "loss": 1.0799,
      "num_input_tokens_seen": 10543152,
      "step": 19025
    },
    {
      "epoch": 333.86725663716817,
      "grad_norm": 7.15625,
      "learning_rate": 2.6904706518874816e-05,
      "loss": 0.9484,
      "num_input_tokens_seen": 10546176,
      "step": 19030
    },
    {
      "epoch": 333.95575221238937,
      "grad_norm": 10.3125,
      "learning_rate": 2.6894917430202615e-05,
      "loss": 0.8116,
      "num_input_tokens_seen": 10549168,
      "step": 19035
    },
    {
      "epoch": 334.0353982300885,
      "grad_norm": 6.59375,
      "learning_rate": 2.6885128049310343e-05,
      "loss": 1.0013,
      "num_input_tokens_seen": 10551816,
      "step": 19040
    },
    {
      "epoch": 334.12389380530976,
      "grad_norm": 9.375,
      "learning_rate": 2.687533837770762e-05,
      "loss": 0.9592,
      "num_input_tokens_seen": 10554520,
      "step": 19045
    },
    {
      "epoch": 334.21238938053096,
      "grad_norm": 8.25,
      "learning_rate": 2.6865548416904162e-05,
      "loss": 1.1765,
      "num_input_tokens_seen": 10557496,
      "step": 19050
    },
    {
      "epoch": 334.3008849557522,
      "grad_norm": 8.125,
      "learning_rate": 2.68557581684097e-05,
      "loss": 0.8765,
      "num_input_tokens_seen": 10560744,
      "step": 19055
    },
    {
      "epoch": 334.3893805309734,
      "grad_norm": 8.875,
      "learning_rate": 2.6845967633733998e-05,
      "loss": 1.0904,
      "num_input_tokens_seen": 10563384,
      "step": 19060
    },
    {
      "epoch": 334.4778761061947,
      "grad_norm": 10.875,
      "learning_rate": 2.683617681438689e-05,
      "loss": 1.1597,
      "num_input_tokens_seen": 10566088,
      "step": 19065
    },
    {
      "epoch": 334.56637168141594,
      "grad_norm": 9.0625,
      "learning_rate": 2.682638571187825e-05,
      "loss": 1.1846,
      "num_input_tokens_seen": 10569080,
      "step": 19070
    },
    {
      "epoch": 334.65486725663715,
      "grad_norm": 8.125,
      "learning_rate": 2.6816594327717976e-05,
      "loss": 1.1465,
      "num_input_tokens_seen": 10571416,
      "step": 19075
    },
    {
      "epoch": 334.7433628318584,
      "grad_norm": 8.0625,
      "learning_rate": 2.680680266341603e-05,
      "loss": 1.1231,
      "num_input_tokens_seen": 10574280,
      "step": 19080
    },
    {
      "epoch": 334.83185840707966,
      "grad_norm": 7.875,
      "learning_rate": 2.67970107204824e-05,
      "loss": 0.9985,
      "num_input_tokens_seen": 10576840,
      "step": 19085
    },
    {
      "epoch": 334.92035398230087,
      "grad_norm": 9.75,
      "learning_rate": 2.6787218500427142e-05,
      "loss": 1.0755,
      "num_input_tokens_seen": 10579464,
      "step": 19090
    },
    {
      "epoch": 335.0,
      "grad_norm": 15.0625,
      "learning_rate": 2.6777426004760332e-05,
      "loss": 0.7655,
      "num_input_tokens_seen": 10581672,
      "step": 19095
    },
    {
      "epoch": 335.08849557522126,
      "grad_norm": 8.875,
      "learning_rate": 2.6767633234992094e-05,
      "loss": 0.9721,
      "num_input_tokens_seen": 10584520,
      "step": 19100
    },
    {
      "epoch": 335.17699115044246,
      "grad_norm": 7.09375,
      "learning_rate": 2.6757840192632598e-05,
      "loss": 1.2427,
      "num_input_tokens_seen": 10587752,
      "step": 19105
    },
    {
      "epoch": 335.2654867256637,
      "grad_norm": 7.96875,
      "learning_rate": 2.6748046879192052e-05,
      "loss": 1.1486,
      "num_input_tokens_seen": 10590424,
      "step": 19110
    },
    {
      "epoch": 335.353982300885,
      "grad_norm": 11.0,
      "learning_rate": 2.673825329618071e-05,
      "loss": 1.009,
      "num_input_tokens_seen": 10592920,
      "step": 19115
    },
    {
      "epoch": 335.4424778761062,
      "grad_norm": 6.9375,
      "learning_rate": 2.6728459445108866e-05,
      "loss": 1.0662,
      "num_input_tokens_seen": 10595464,
      "step": 19120
    },
    {
      "epoch": 335.53097345132744,
      "grad_norm": 8.6875,
      "learning_rate": 2.6718665327486854e-05,
      "loss": 1.0276,
      "num_input_tokens_seen": 10598216,
      "step": 19125
    },
    {
      "epoch": 335.6194690265487,
      "grad_norm": 8.1875,
      "learning_rate": 2.6708870944825048e-05,
      "loss": 1.165,
      "num_input_tokens_seen": 10601416,
      "step": 19130
    },
    {
      "epoch": 335.7079646017699,
      "grad_norm": 8.4375,
      "learning_rate": 2.6699076298633874e-05,
      "loss": 0.9231,
      "num_input_tokens_seen": 10604184,
      "step": 19135
    },
    {
      "epoch": 335.79646017699116,
      "grad_norm": 9.625,
      "learning_rate": 2.6689281390423788e-05,
      "loss": 0.8668,
      "num_input_tokens_seen": 10607448,
      "step": 19140
    },
    {
      "epoch": 335.88495575221236,
      "grad_norm": 8.5625,
      "learning_rate": 2.667948622170527e-05,
      "loss": 1.0736,
      "num_input_tokens_seen": 10610296,
      "step": 19145
    },
    {
      "epoch": 335.9734513274336,
      "grad_norm": 7.59375,
      "learning_rate": 2.6669690793988873e-05,
      "loss": 1.1033,
      "num_input_tokens_seen": 10612712,
      "step": 19150
    },
    {
      "epoch": 336.05309734513276,
      "grad_norm": 10.625,
      "learning_rate": 2.665989510878518e-05,
      "loss": 1.1628,
      "num_input_tokens_seen": 10615304,
      "step": 19155
    },
    {
      "epoch": 336.14159292035396,
      "grad_norm": 10.5625,
      "learning_rate": 2.6650099167604793e-05,
      "loss": 1.3106,
      "num_input_tokens_seen": 10618152,
      "step": 19160
    },
    {
      "epoch": 336.2300884955752,
      "grad_norm": 10.75,
      "learning_rate": 2.6640302971958376e-05,
      "loss": 0.9268,
      "num_input_tokens_seen": 10621400,
      "step": 19165
    },
    {
      "epoch": 336.3185840707965,
      "grad_norm": 8.125,
      "learning_rate": 2.6630506523356635e-05,
      "loss": 1.0643,
      "num_input_tokens_seen": 10624264,
      "step": 19170
    },
    {
      "epoch": 336.4070796460177,
      "grad_norm": 9.5,
      "learning_rate": 2.6620709823310297e-05,
      "loss": 0.9934,
      "num_input_tokens_seen": 10626840,
      "step": 19175
    },
    {
      "epoch": 336.49557522123894,
      "grad_norm": 7.4375,
      "learning_rate": 2.661091287333014e-05,
      "loss": 0.9139,
      "num_input_tokens_seen": 10629720,
      "step": 19180
    },
    {
      "epoch": 336.5840707964602,
      "grad_norm": 9.375,
      "learning_rate": 2.660111567492696e-05,
      "loss": 1.078,
      "num_input_tokens_seen": 10632440,
      "step": 19185
    },
    {
      "epoch": 336.6725663716814,
      "grad_norm": 6.09375,
      "learning_rate": 2.6591318229611635e-05,
      "loss": 1.015,
      "num_input_tokens_seen": 10635064,
      "step": 19190
    },
    {
      "epoch": 336.76106194690266,
      "grad_norm": 7.40625,
      "learning_rate": 2.6581520538895037e-05,
      "loss": 1.1141,
      "num_input_tokens_seen": 10637768,
      "step": 19195
    },
    {
      "epoch": 336.8495575221239,
      "grad_norm": 8.5625,
      "learning_rate": 2.6571722604288102e-05,
      "loss": 1.0664,
      "num_input_tokens_seen": 10640296,
      "step": 19200
    },
    {
      "epoch": 336.8495575221239,
      "eval_loss": 1.0257378816604614,
      "eval_runtime": 0.9445,
      "eval_samples_per_second": 26.47,
      "eval_steps_per_second": 13.765,
      "num_input_tokens_seen": 10640296,
      "step": 19200
    },
    {
      "epoch": 336.9380530973451,
      "grad_norm": 7.875,
      "learning_rate": 2.656192442730179e-05,
      "loss": 0.9363,
      "num_input_tokens_seen": 10643560,
      "step": 19205
    },
    {
      "epoch": 337.01769911504425,
      "grad_norm": 7.3125,
      "learning_rate": 2.6552126009447098e-05,
      "loss": 1.0801,
      "num_input_tokens_seen": 10645600,
      "step": 19210
    },
    {
      "epoch": 337.1061946902655,
      "grad_norm": 6.21875,
      "learning_rate": 2.654232735223507e-05,
      "loss": 0.9726,
      "num_input_tokens_seen": 10648512,
      "step": 19215
    },
    {
      "epoch": 337.1946902654867,
      "grad_norm": 8.1875,
      "learning_rate": 2.6532528457176787e-05,
      "loss": 1.1284,
      "num_input_tokens_seen": 10651232,
      "step": 19220
    },
    {
      "epoch": 337.283185840708,
      "grad_norm": 8.375,
      "learning_rate": 2.6522729325783348e-05,
      "loss": 1.1188,
      "num_input_tokens_seen": 10654592,
      "step": 19225
    },
    {
      "epoch": 337.37168141592923,
      "grad_norm": 11.3125,
      "learning_rate": 2.6512929959565914e-05,
      "loss": 1.1233,
      "num_input_tokens_seen": 10657456,
      "step": 19230
    },
    {
      "epoch": 337.46017699115043,
      "grad_norm": 9.125,
      "learning_rate": 2.6503130360035673e-05,
      "loss": 1.1091,
      "num_input_tokens_seen": 10660128,
      "step": 19235
    },
    {
      "epoch": 337.5486725663717,
      "grad_norm": 6.46875,
      "learning_rate": 2.6493330528703835e-05,
      "loss": 0.9431,
      "num_input_tokens_seen": 10663088,
      "step": 19240
    },
    {
      "epoch": 337.6371681415929,
      "grad_norm": 6.90625,
      "learning_rate": 2.648353046708167e-05,
      "loss": 0.8249,
      "num_input_tokens_seen": 10666000,
      "step": 19245
    },
    {
      "epoch": 337.72566371681415,
      "grad_norm": 7.96875,
      "learning_rate": 2.647373017668046e-05,
      "loss": 1.1235,
      "num_input_tokens_seen": 10668752,
      "step": 19250
    },
    {
      "epoch": 337.8141592920354,
      "grad_norm": 8.375,
      "learning_rate": 2.6463929659011537e-05,
      "loss": 0.9846,
      "num_input_tokens_seen": 10671280,
      "step": 19255
    },
    {
      "epoch": 337.9026548672566,
      "grad_norm": 7.03125,
      "learning_rate": 2.6454128915586262e-05,
      "loss": 1.073,
      "num_input_tokens_seen": 10673952,
      "step": 19260
    },
    {
      "epoch": 337.9911504424779,
      "grad_norm": 7.78125,
      "learning_rate": 2.6444327947916036e-05,
      "loss": 1.107,
      "num_input_tokens_seen": 10676512,
      "step": 19265
    },
    {
      "epoch": 338.070796460177,
      "grad_norm": 7.34375,
      "learning_rate": 2.6434526757512292e-05,
      "loss": 1.1915,
      "num_input_tokens_seen": 10678800,
      "step": 19270
    },
    {
      "epoch": 338.1592920353982,
      "grad_norm": 10.625,
      "learning_rate": 2.6424725345886486e-05,
      "loss": 0.966,
      "num_input_tokens_seen": 10681984,
      "step": 19275
    },
    {
      "epoch": 338.24778761061947,
      "grad_norm": 6.65625,
      "learning_rate": 2.641492371455014e-05,
      "loss": 0.9879,
      "num_input_tokens_seen": 10684576,
      "step": 19280
    },
    {
      "epoch": 338.3362831858407,
      "grad_norm": 7.1875,
      "learning_rate": 2.640512186501477e-05,
      "loss": 0.9906,
      "num_input_tokens_seen": 10687808,
      "step": 19285
    },
    {
      "epoch": 338.42477876106193,
      "grad_norm": 7.75,
      "learning_rate": 2.639531979879195e-05,
      "loss": 1.0993,
      "num_input_tokens_seen": 10690320,
      "step": 19290
    },
    {
      "epoch": 338.5132743362832,
      "grad_norm": 7.28125,
      "learning_rate": 2.638551751739328e-05,
      "loss": 1.0034,
      "num_input_tokens_seen": 10693200,
      "step": 19295
    },
    {
      "epoch": 338.60176991150445,
      "grad_norm": 7.9375,
      "learning_rate": 2.6375715022330404e-05,
      "loss": 1.0954,
      "num_input_tokens_seen": 10695968,
      "step": 19300
    },
    {
      "epoch": 338.69026548672565,
      "grad_norm": 6.21875,
      "learning_rate": 2.6365912315114976e-05,
      "loss": 1.0576,
      "num_input_tokens_seen": 10698480,
      "step": 19305
    },
    {
      "epoch": 338.7787610619469,
      "grad_norm": 10.8125,
      "learning_rate": 2.6356109397258704e-05,
      "loss": 1.0703,
      "num_input_tokens_seen": 10701552,
      "step": 19310
    },
    {
      "epoch": 338.86725663716817,
      "grad_norm": 8.4375,
      "learning_rate": 2.6346306270273325e-05,
      "loss": 1.0098,
      "num_input_tokens_seen": 10704256,
      "step": 19315
    },
    {
      "epoch": 338.95575221238937,
      "grad_norm": 12.0625,
      "learning_rate": 2.6336502935670608e-05,
      "loss": 1.1254,
      "num_input_tokens_seen": 10707200,
      "step": 19320
    },
    {
      "epoch": 339.0353982300885,
      "grad_norm": 7.28125,
      "learning_rate": 2.6326699394962333e-05,
      "loss": 1.0598,
      "num_input_tokens_seen": 10709344,
      "step": 19325
    },
    {
      "epoch": 339.12389380530976,
      "grad_norm": 7.46875,
      "learning_rate": 2.6316895649660334e-05,
      "loss": 1.1435,
      "num_input_tokens_seen": 10712048,
      "step": 19330
    },
    {
      "epoch": 339.21238938053096,
      "grad_norm": 6.28125,
      "learning_rate": 2.6307091701276486e-05,
      "loss": 0.9544,
      "num_input_tokens_seen": 10714608,
      "step": 19335
    },
    {
      "epoch": 339.3008849557522,
      "grad_norm": 7.28125,
      "learning_rate": 2.629728755132267e-05,
      "loss": 1.0226,
      "num_input_tokens_seen": 10717536,
      "step": 19340
    },
    {
      "epoch": 339.3893805309734,
      "grad_norm": 7.5,
      "learning_rate": 2.628748320131081e-05,
      "loss": 1.118,
      "num_input_tokens_seen": 10720528,
      "step": 19345
    },
    {
      "epoch": 339.4778761061947,
      "grad_norm": 7.90625,
      "learning_rate": 2.6277678652752856e-05,
      "loss": 1.1203,
      "num_input_tokens_seen": 10723984,
      "step": 19350
    },
    {
      "epoch": 339.56637168141594,
      "grad_norm": 11.625,
      "learning_rate": 2.6267873907160807e-05,
      "loss": 1.2229,
      "num_input_tokens_seen": 10726704,
      "step": 19355
    },
    {
      "epoch": 339.65486725663715,
      "grad_norm": 7.8125,
      "learning_rate": 2.6258068966046668e-05,
      "loss": 0.9738,
      "num_input_tokens_seen": 10729344,
      "step": 19360
    },
    {
      "epoch": 339.7433628318584,
      "grad_norm": 8.875,
      "learning_rate": 2.6248263830922475e-05,
      "loss": 0.9768,
      "num_input_tokens_seen": 10732400,
      "step": 19365
    },
    {
      "epoch": 339.83185840707966,
      "grad_norm": 6.28125,
      "learning_rate": 2.6238458503300318e-05,
      "loss": 0.9379,
      "num_input_tokens_seen": 10735024,
      "step": 19370
    },
    {
      "epoch": 339.92035398230087,
      "grad_norm": 8.3125,
      "learning_rate": 2.6228652984692292e-05,
      "loss": 1.031,
      "num_input_tokens_seen": 10737456,
      "step": 19375
    },
    {
      "epoch": 340.0,
      "grad_norm": 11.1875,
      "learning_rate": 2.621884727661054e-05,
      "loss": 1.1023,
      "num_input_tokens_seen": 10739656,
      "step": 19380
    },
    {
      "epoch": 340.08849557522126,
      "grad_norm": 10.4375,
      "learning_rate": 2.6209041380567222e-05,
      "loss": 1.1286,
      "num_input_tokens_seen": 10742552,
      "step": 19385
    },
    {
      "epoch": 340.17699115044246,
      "grad_norm": 8.0,
      "learning_rate": 2.6199235298074527e-05,
      "loss": 0.9554,
      "num_input_tokens_seen": 10745400,
      "step": 19390
    },
    {
      "epoch": 340.2654867256637,
      "grad_norm": 7.0,
      "learning_rate": 2.618942903064468e-05,
      "loss": 1.0598,
      "num_input_tokens_seen": 10748072,
      "step": 19395
    },
    {
      "epoch": 340.353982300885,
      "grad_norm": 13.3125,
      "learning_rate": 2.6179622579789932e-05,
      "loss": 1.022,
      "num_input_tokens_seen": 10750776,
      "step": 19400
    },
    {
      "epoch": 340.353982300885,
      "eval_loss": 1.026552677154541,
      "eval_runtime": 0.9347,
      "eval_samples_per_second": 26.746,
      "eval_steps_per_second": 13.908,
      "num_input_tokens_seen": 10750776,
      "step": 19400
    },
    {
      "epoch": 340.4424778761062,
      "grad_norm": 8.625,
      "learning_rate": 2.6169815947022553e-05,
      "loss": 1.1528,
      "num_input_tokens_seen": 10753768,
      "step": 19405
    },
    {
      "epoch": 340.53097345132744,
      "grad_norm": 11.375,
      "learning_rate": 2.6160009133854853e-05,
      "loss": 1.0931,
      "num_input_tokens_seen": 10756808,
      "step": 19410
    },
    {
      "epoch": 340.6194690265487,
      "grad_norm": 10.9375,
      "learning_rate": 2.6150202141799168e-05,
      "loss": 0.9727,
      "num_input_tokens_seen": 10759560,
      "step": 19415
    },
    {
      "epoch": 340.7079646017699,
      "grad_norm": 7.4375,
      "learning_rate": 2.614039497236786e-05,
      "loss": 1.0044,
      "num_input_tokens_seen": 10762248,
      "step": 19420
    },
    {
      "epoch": 340.79646017699116,
      "grad_norm": 9.875,
      "learning_rate": 2.6130587627073315e-05,
      "loss": 1.0802,
      "num_input_tokens_seen": 10764968,
      "step": 19425
    },
    {
      "epoch": 340.88495575221236,
      "grad_norm": 6.71875,
      "learning_rate": 2.6120780107427956e-05,
      "loss": 1.018,
      "num_input_tokens_seen": 10767944,
      "step": 19430
    },
    {
      "epoch": 340.9734513274336,
      "grad_norm": 7.875,
      "learning_rate": 2.6110972414944214e-05,
      "loss": 1.0761,
      "num_input_tokens_seen": 10770616,
      "step": 19435
    },
    {
      "epoch": 341.05309734513276,
      "grad_norm": 6.78125,
      "learning_rate": 2.6101164551134565e-05,
      "loss": 1.01,
      "num_input_tokens_seen": 10772768,
      "step": 19440
    },
    {
      "epoch": 341.14159292035396,
      "grad_norm": 7.875,
      "learning_rate": 2.6091356517511505e-05,
      "loss": 0.8984,
      "num_input_tokens_seen": 10775248,
      "step": 19445
    },
    {
      "epoch": 341.2300884955752,
      "grad_norm": 8.5,
      "learning_rate": 2.608154831558755e-05,
      "loss": 1.014,
      "num_input_tokens_seen": 10778432,
      "step": 19450
    },
    {
      "epoch": 341.3185840707965,
      "grad_norm": 7.21875,
      "learning_rate": 2.607173994687526e-05,
      "loss": 0.9646,
      "num_input_tokens_seen": 10781424,
      "step": 19455
    },
    {
      "epoch": 341.4070796460177,
      "grad_norm": 9.375,
      "learning_rate": 2.6061931412887196e-05,
      "loss": 0.9442,
      "num_input_tokens_seen": 10784320,
      "step": 19460
    },
    {
      "epoch": 341.49557522123894,
      "grad_norm": 8.1875,
      "learning_rate": 2.6052122715135973e-05,
      "loss": 1.0503,
      "num_input_tokens_seen": 10787264,
      "step": 19465
    },
    {
      "epoch": 341.5840707964602,
      "grad_norm": 9.625,
      "learning_rate": 2.60423138551342e-05,
      "loss": 1.0598,
      "num_input_tokens_seen": 10789680,
      "step": 19470
    },
    {
      "epoch": 341.6725663716814,
      "grad_norm": 8.3125,
      "learning_rate": 2.6032504834394527e-05,
      "loss": 1.1615,
      "num_input_tokens_seen": 10792304,
      "step": 19475
    },
    {
      "epoch": 341.76106194690266,
      "grad_norm": 9.0,
      "learning_rate": 2.602269565442964e-05,
      "loss": 1.0755,
      "num_input_tokens_seen": 10795584,
      "step": 19480
    },
    {
      "epoch": 341.8495575221239,
      "grad_norm": 8.6875,
      "learning_rate": 2.6012886316752227e-05,
      "loss": 1.1211,
      "num_input_tokens_seen": 10798624,
      "step": 19485
    },
    {
      "epoch": 341.9380530973451,
      "grad_norm": 10.375,
      "learning_rate": 2.6003076822875018e-05,
      "loss": 1.1021,
      "num_input_tokens_seen": 10801392,
      "step": 19490
    },
    {
      "epoch": 342.01769911504425,
      "grad_norm": 7.34375,
      "learning_rate": 2.5993267174310755e-05,
      "loss": 1.254,
      "num_input_tokens_seen": 10803656,
      "step": 19495
    },
    {
      "epoch": 342.1061946902655,
      "grad_norm": 8.6875,
      "learning_rate": 2.5983457372572218e-05,
      "loss": 0.9922,
      "num_input_tokens_seen": 10806632,
      "step": 19500
    },
    {
      "epoch": 342.1946902654867,
      "grad_norm": 9.9375,
      "learning_rate": 2.597364741917219e-05,
      "loss": 1.1095,
      "num_input_tokens_seen": 10809592,
      "step": 19505
    },
    {
      "epoch": 342.283185840708,
      "grad_norm": 8.5,
      "learning_rate": 2.5963837315623492e-05,
      "loss": 1.0738,
      "num_input_tokens_seen": 10812360,
      "step": 19510
    },
    {
      "epoch": 342.37168141592923,
      "grad_norm": 8.0,
      "learning_rate": 2.595402706343897e-05,
      "loss": 0.8815,
      "num_input_tokens_seen": 10815208,
      "step": 19515
    },
    {
      "epoch": 342.46017699115043,
      "grad_norm": 8.5625,
      "learning_rate": 2.594421666413148e-05,
      "loss": 0.9917,
      "num_input_tokens_seen": 10818152,
      "step": 19520
    },
    {
      "epoch": 342.5486725663717,
      "grad_norm": 8.9375,
      "learning_rate": 2.5934406119213928e-05,
      "loss": 0.9629,
      "num_input_tokens_seen": 10820760,
      "step": 19525
    },
    {
      "epoch": 342.6371681415929,
      "grad_norm": 8.1875,
      "learning_rate": 2.5924595430199193e-05,
      "loss": 0.9346,
      "num_input_tokens_seen": 10823256,
      "step": 19530
    },
    {
      "epoch": 342.72566371681415,
      "grad_norm": 10.8125,
      "learning_rate": 2.5914784598600238e-05,
      "loss": 1.3944,
      "num_input_tokens_seen": 10825960,
      "step": 19535
    },
    {
      "epoch": 342.8141592920354,
      "grad_norm": 8.125,
      "learning_rate": 2.5904973625930002e-05,
      "loss": 1.0848,
      "num_input_tokens_seen": 10828776,
      "step": 19540
    },
    {
      "epoch": 342.9026548672566,
      "grad_norm": 8.25,
      "learning_rate": 2.5895162513701456e-05,
      "loss": 1.0047,
      "num_input_tokens_seen": 10831592,
      "step": 19545
    },
    {
      "epoch": 342.9911504424779,
      "grad_norm": 10.75,
      "learning_rate": 2.5885351263427593e-05,
      "loss": 1.232,
      "num_input_tokens_seen": 10834376,
      "step": 19550
    },
    {
      "epoch": 343.070796460177,
      "grad_norm": 8.0625,
      "learning_rate": 2.5875539876621448e-05,
      "loss": 0.9459,
      "num_input_tokens_seen": 10836880,
      "step": 19555
    },
    {
      "epoch": 343.1592920353982,
      "grad_norm": 9.875,
      "learning_rate": 2.586572835479605e-05,
      "loss": 1.1383,
      "num_input_tokens_seen": 10839584,
      "step": 19560
    },
    {
      "epoch": 343.24778761061947,
      "grad_norm": 9.25,
      "learning_rate": 2.585591669946446e-05,
      "loss": 0.9953,
      "num_input_tokens_seen": 10842176,
      "step": 19565
    },
    {
      "epoch": 343.3362831858407,
      "grad_norm": 7.875,
      "learning_rate": 2.5846104912139756e-05,
      "loss": 1.1483,
      "num_input_tokens_seen": 10844896,
      "step": 19570
    },
    {
      "epoch": 343.42477876106193,
      "grad_norm": 7.4375,
      "learning_rate": 2.583629299433505e-05,
      "loss": 1.1293,
      "num_input_tokens_seen": 10847824,
      "step": 19575
    },
    {
      "epoch": 343.5132743362832,
      "grad_norm": 9.125,
      "learning_rate": 2.582648094756345e-05,
      "loss": 1.0833,
      "num_input_tokens_seen": 10850480,
      "step": 19580
    },
    {
      "epoch": 343.60176991150445,
      "grad_norm": 9.75,
      "learning_rate": 2.5816668773338098e-05,
      "loss": 1.2875,
      "num_input_tokens_seen": 10853344,
      "step": 19585
    },
    {
      "epoch": 343.69026548672565,
      "grad_norm": 6.75,
      "learning_rate": 2.580685647317216e-05,
      "loss": 0.8911,
      "num_input_tokens_seen": 10855840,
      "step": 19590
    },
    {
      "epoch": 343.7787610619469,
      "grad_norm": 6.15625,
      "learning_rate": 2.5797044048578818e-05,
      "loss": 0.9497,
      "num_input_tokens_seen": 10858656,
      "step": 19595
    },
    {
      "epoch": 343.86725663716817,
      "grad_norm": 8.4375,
      "learning_rate": 2.5787231501071262e-05,
      "loss": 0.9859,
      "num_input_tokens_seen": 10861648,
      "step": 19600
    },
    {
      "epoch": 343.86725663716817,
      "eval_loss": 1.031451940536499,
      "eval_runtime": 0.9395,
      "eval_samples_per_second": 26.611,
      "eval_steps_per_second": 13.838,
      "num_input_tokens_seen": 10861648,
      "step": 19600
    },
    {
      "epoch": 343.95575221238937,
      "grad_norm": 10.0,
      "learning_rate": 2.577741883216272e-05,
      "loss": 1.0686,
      "num_input_tokens_seen": 10864720,
      "step": 19605
    },
    {
      "epoch": 344.0353982300885,
      "grad_norm": 8.25,
      "learning_rate": 2.576760604336642e-05,
      "loss": 0.8602,
      "num_input_tokens_seen": 10867312,
      "step": 19610
    },
    {
      "epoch": 344.12389380530976,
      "grad_norm": 8.5,
      "learning_rate": 2.575779313619563e-05,
      "loss": 0.9562,
      "num_input_tokens_seen": 10869904,
      "step": 19615
    },
    {
      "epoch": 344.21238938053096,
      "grad_norm": 10.625,
      "learning_rate": 2.5747980112163605e-05,
      "loss": 0.9448,
      "num_input_tokens_seen": 10872496,
      "step": 19620
    },
    {
      "epoch": 344.3008849557522,
      "grad_norm": 8.25,
      "learning_rate": 2.5738166972783656e-05,
      "loss": 0.9038,
      "num_input_tokens_seen": 10875760,
      "step": 19625
    },
    {
      "epoch": 344.3893805309734,
      "grad_norm": 8.125,
      "learning_rate": 2.5728353719569075e-05,
      "loss": 1.0567,
      "num_input_tokens_seen": 10879072,
      "step": 19630
    },
    {
      "epoch": 344.4778761061947,
      "grad_norm": 8.1875,
      "learning_rate": 2.57185403540332e-05,
      "loss": 1.0083,
      "num_input_tokens_seen": 10881616,
      "step": 19635
    },
    {
      "epoch": 344.56637168141594,
      "grad_norm": 10.125,
      "learning_rate": 2.5708726877689375e-05,
      "loss": 1.24,
      "num_input_tokens_seen": 10884144,
      "step": 19640
    },
    {
      "epoch": 344.65486725663715,
      "grad_norm": 9.5625,
      "learning_rate": 2.5698913292050964e-05,
      "loss": 1.1547,
      "num_input_tokens_seen": 10886752,
      "step": 19645
    },
    {
      "epoch": 344.7433628318584,
      "grad_norm": 7.21875,
      "learning_rate": 2.568909959863133e-05,
      "loss": 0.9883,
      "num_input_tokens_seen": 10889808,
      "step": 19650
    },
    {
      "epoch": 344.83185840707966,
      "grad_norm": 6.90625,
      "learning_rate": 2.5679285798943887e-05,
      "loss": 1.0556,
      "num_input_tokens_seen": 10892608,
      "step": 19655
    },
    {
      "epoch": 344.92035398230087,
      "grad_norm": 7.0625,
      "learning_rate": 2.5669471894502035e-05,
      "loss": 1.1572,
      "num_input_tokens_seen": 10895216,
      "step": 19660
    },
    {
      "epoch": 345.0,
      "grad_norm": 7.90625,
      "learning_rate": 2.56596578868192e-05,
      "loss": 1.148,
      "num_input_tokens_seen": 10897432,
      "step": 19665
    },
    {
      "epoch": 345.08849557522126,
      "grad_norm": 9.0625,
      "learning_rate": 2.564984377740883e-05,
      "loss": 0.9879,
      "num_input_tokens_seen": 10899864,
      "step": 19670
    },
    {
      "epoch": 345.17699115044246,
      "grad_norm": 7.4375,
      "learning_rate": 2.564002956778438e-05,
      "loss": 0.9929,
      "num_input_tokens_seen": 10902472,
      "step": 19675
    },
    {
      "epoch": 345.2654867256637,
      "grad_norm": 8.5625,
      "learning_rate": 2.563021525945934e-05,
      "loss": 1.1194,
      "num_input_tokens_seen": 10905144,
      "step": 19680
    },
    {
      "epoch": 345.353982300885,
      "grad_norm": 8.6875,
      "learning_rate": 2.562040085394718e-05,
      "loss": 1.1809,
      "num_input_tokens_seen": 10908136,
      "step": 19685
    },
    {
      "epoch": 345.4424778761062,
      "grad_norm": 8.8125,
      "learning_rate": 2.56105863527614e-05,
      "loss": 0.9662,
      "num_input_tokens_seen": 10910968,
      "step": 19690
    },
    {
      "epoch": 345.53097345132744,
      "grad_norm": 6.6875,
      "learning_rate": 2.5600771757415548e-05,
      "loss": 0.9083,
      "num_input_tokens_seen": 10914200,
      "step": 19695
    },
    {
      "epoch": 345.6194690265487,
      "grad_norm": 6.125,
      "learning_rate": 2.5590957069423134e-05,
      "loss": 0.9082,
      "num_input_tokens_seen": 10917192,
      "step": 19700
    },
    {
      "epoch": 345.7079646017699,
      "grad_norm": 7.28125,
      "learning_rate": 2.5581142290297716e-05,
      "loss": 1.0961,
      "num_input_tokens_seen": 10920200,
      "step": 19705
    },
    {
      "epoch": 345.79646017699116,
      "grad_norm": 8.4375,
      "learning_rate": 2.557132742155285e-05,
      "loss": 1.1588,
      "num_input_tokens_seen": 10922904,
      "step": 19710
    },
    {
      "epoch": 345.88495575221236,
      "grad_norm": 7.90625,
      "learning_rate": 2.556151246470212e-05,
      "loss": 1.0429,
      "num_input_tokens_seen": 10925736,
      "step": 19715
    },
    {
      "epoch": 345.9734513274336,
      "grad_norm": 9.25,
      "learning_rate": 2.5551697421259114e-05,
      "loss": 1.2517,
      "num_input_tokens_seen": 10928568,
      "step": 19720
    },
    {
      "epoch": 346.05309734513276,
      "grad_norm": 7.25,
      "learning_rate": 2.554188229273743e-05,
      "loss": 0.9211,
      "num_input_tokens_seen": 10930616,
      "step": 19725
    },
    {
      "epoch": 346.14159292035396,
      "grad_norm": 7.625,
      "learning_rate": 2.5532067080650678e-05,
      "loss": 1.051,
      "num_input_tokens_seen": 10933064,
      "step": 19730
    },
    {
      "epoch": 346.2300884955752,
      "grad_norm": 5.96875,
      "learning_rate": 2.55222517865125e-05,
      "loss": 0.8819,
      "num_input_tokens_seen": 10936216,
      "step": 19735
    },
    {
      "epoch": 346.3185840707965,
      "grad_norm": 8.5,
      "learning_rate": 2.5512436411836538e-05,
      "loss": 1.1499,
      "num_input_tokens_seen": 10939400,
      "step": 19740
    },
    {
      "epoch": 346.4070796460177,
      "grad_norm": 8.5625,
      "learning_rate": 2.5502620958136443e-05,
      "loss": 1.0786,
      "num_input_tokens_seen": 10942472,
      "step": 19745
    },
    {
      "epoch": 346.49557522123894,
      "grad_norm": 9.0625,
      "learning_rate": 2.5492805426925874e-05,
      "loss": 1.0832,
      "num_input_tokens_seen": 10944856,
      "step": 19750
    },
    {
      "epoch": 346.5840707964602,
      "grad_norm": 7.25,
      "learning_rate": 2.5482989819718523e-05,
      "loss": 1.0135,
      "num_input_tokens_seen": 10947944,
      "step": 19755
    },
    {
      "epoch": 346.6725663716814,
      "grad_norm": 6.3125,
      "learning_rate": 2.5473174138028065e-05,
      "loss": 1.1063,
      "num_input_tokens_seen": 10950568,
      "step": 19760
    },
    {
      "epoch": 346.76106194690266,
      "grad_norm": 6.9375,
      "learning_rate": 2.5463358383368212e-05,
      "loss": 1.234,
      "num_input_tokens_seen": 10953464,
      "step": 19765
    },
    {
      "epoch": 346.8495575221239,
      "grad_norm": 8.875,
      "learning_rate": 2.545354255725267e-05,
      "loss": 1.0492,
      "num_input_tokens_seen": 10956200,
      "step": 19770
    },
    {
      "epoch": 346.9380530973451,
      "grad_norm": 11.9375,
      "learning_rate": 2.5443726661195165e-05,
      "loss": 1.1123,
      "num_input_tokens_seen": 10959192,
      "step": 19775
    },
    {
      "epoch": 347.01769911504425,
      "grad_norm": 8.0625,
      "learning_rate": 2.543391069670944e-05,
      "loss": 0.9453,
      "num_input_tokens_seen": 10961272,
      "step": 19780
    },
    {
      "epoch": 347.1061946902655,
      "grad_norm": 9.5,
      "learning_rate": 2.5424094665309228e-05,
      "loss": 0.9674,
      "num_input_tokens_seen": 10963848,
      "step": 19785
    },
    {
      "epoch": 347.1946902654867,
      "grad_norm": 8.125,
      "learning_rate": 2.5414278568508292e-05,
      "loss": 0.9834,
      "num_input_tokens_seen": 10966600,
      "step": 19790
    },
    {
      "epoch": 347.283185840708,
      "grad_norm": 7.65625,
      "learning_rate": 2.540446240782039e-05,
      "loss": 1.0284,
      "num_input_tokens_seen": 10970040,
      "step": 19795
    },
    {
      "epoch": 347.37168141592923,
      "grad_norm": 8.9375,
      "learning_rate": 2.5394646184759307e-05,
      "loss": 1.1735,
      "num_input_tokens_seen": 10972808,
      "step": 19800
    },
    {
      "epoch": 347.37168141592923,
      "eval_loss": 1.0318105220794678,
      "eval_runtime": 0.9382,
      "eval_samples_per_second": 26.647,
      "eval_steps_per_second": 13.856,
      "num_input_tokens_seen": 10972808,
      "step": 19800
    },
    {
      "epoch": 347.46017699115043,
      "grad_norm": 6.75,
      "learning_rate": 2.538482990083882e-05,
      "loss": 0.9315,
      "num_input_tokens_seen": 10975576,
      "step": 19805
    },
    {
      "epoch": 347.5486725663717,
      "grad_norm": 9.3125,
      "learning_rate": 2.5375013557572725e-05,
      "loss": 0.8641,
      "num_input_tokens_seen": 10978200,
      "step": 19810
    },
    {
      "epoch": 347.6371681415929,
      "grad_norm": 8.8125,
      "learning_rate": 2.536519715647483e-05,
      "loss": 1.252,
      "num_input_tokens_seen": 10981160,
      "step": 19815
    },
    {
      "epoch": 347.72566371681415,
      "grad_norm": 9.75,
      "learning_rate": 2.535538069905894e-05,
      "loss": 1.2949,
      "num_input_tokens_seen": 10983880,
      "step": 19820
    },
    {
      "epoch": 347.8141592920354,
      "grad_norm": 8.375,
      "learning_rate": 2.534556418683888e-05,
      "loss": 1.0635,
      "num_input_tokens_seen": 10986568,
      "step": 19825
    },
    {
      "epoch": 347.9026548672566,
      "grad_norm": 8.375,
      "learning_rate": 2.5335747621328486e-05,
      "loss": 1.0796,
      "num_input_tokens_seen": 10989448,
      "step": 19830
    },
    {
      "epoch": 347.9911504424779,
      "grad_norm": 8.75,
      "learning_rate": 2.5325931004041586e-05,
      "loss": 0.9228,
      "num_input_tokens_seen": 10992008,
      "step": 19835
    },
    {
      "epoch": 348.070796460177,
      "grad_norm": 10.8125,
      "learning_rate": 2.5316114336492032e-05,
      "loss": 1.0938,
      "num_input_tokens_seen": 10994536,
      "step": 19840
    },
    {
      "epoch": 348.1592920353982,
      "grad_norm": 8.0,
      "learning_rate": 2.530629762019367e-05,
      "loss": 1.0597,
      "num_input_tokens_seen": 10997064,
      "step": 19845
    },
    {
      "epoch": 348.24778761061947,
      "grad_norm": 7.34375,
      "learning_rate": 2.5296480856660364e-05,
      "loss": 0.9736,
      "num_input_tokens_seen": 10999992,
      "step": 19850
    },
    {
      "epoch": 348.3362831858407,
      "grad_norm": 6.875,
      "learning_rate": 2.528666404740599e-05,
      "loss": 1.0728,
      "num_input_tokens_seen": 11002840,
      "step": 19855
    },
    {
      "epoch": 348.42477876106193,
      "grad_norm": 9.625,
      "learning_rate": 2.527684719394442e-05,
      "loss": 0.9716,
      "num_input_tokens_seen": 11005832,
      "step": 19860
    },
    {
      "epoch": 348.5132743362832,
      "grad_norm": 7.9375,
      "learning_rate": 2.526703029778953e-05,
      "loss": 1.2549,
      "num_input_tokens_seen": 11008360,
      "step": 19865
    },
    {
      "epoch": 348.60176991150445,
      "grad_norm": 6.4375,
      "learning_rate": 2.5257213360455208e-05,
      "loss": 1.0129,
      "num_input_tokens_seen": 11011560,
      "step": 19870
    },
    {
      "epoch": 348.69026548672565,
      "grad_norm": 7.9375,
      "learning_rate": 2.5247396383455353e-05,
      "loss": 0.9308,
      "num_input_tokens_seen": 11014456,
      "step": 19875
    },
    {
      "epoch": 348.7787610619469,
      "grad_norm": 9.25,
      "learning_rate": 2.523757936830387e-05,
      "loss": 1.0976,
      "num_input_tokens_seen": 11017160,
      "step": 19880
    },
    {
      "epoch": 348.86725663716817,
      "grad_norm": 7.125,
      "learning_rate": 2.5227762316514662e-05,
      "loss": 1.0852,
      "num_input_tokens_seen": 11019704,
      "step": 19885
    },
    {
      "epoch": 348.95575221238937,
      "grad_norm": 7.40625,
      "learning_rate": 2.5217945229601648e-05,
      "loss": 1.0399,
      "num_input_tokens_seen": 11022776,
      "step": 19890
    },
    {
      "epoch": 349.0353982300885,
      "grad_norm": 7.4375,
      "learning_rate": 2.5208128109078738e-05,
      "loss": 1.1602,
      "num_input_tokens_seen": 11025056,
      "step": 19895
    },
    {
      "epoch": 349.12389380530976,
      "grad_norm": 9.0625,
      "learning_rate": 2.5198310956459853e-05,
      "loss": 0.9133,
      "num_input_tokens_seen": 11028112,
      "step": 19900
    },
    {
      "epoch": 349.21238938053096,
      "grad_norm": 7.84375,
      "learning_rate": 2.518849377325893e-05,
      "loss": 1.1475,
      "num_input_tokens_seen": 11030752,
      "step": 19905
    },
    {
      "epoch": 349.3008849557522,
      "grad_norm": 5.28125,
      "learning_rate": 2.51786765609899e-05,
      "loss": 1.0199,
      "num_input_tokens_seen": 11033920,
      "step": 19910
    },
    {
      "epoch": 349.3893805309734,
      "grad_norm": 9.8125,
      "learning_rate": 2.5168859321166694e-05,
      "loss": 1.0725,
      "num_input_tokens_seen": 11036832,
      "step": 19915
    },
    {
      "epoch": 349.4778761061947,
      "grad_norm": 7.4375,
      "learning_rate": 2.515904205530326e-05,
      "loss": 0.9938,
      "num_input_tokens_seen": 11038992,
      "step": 19920
    },
    {
      "epoch": 349.56637168141594,
      "grad_norm": 7.59375,
      "learning_rate": 2.514922476491355e-05,
      "loss": 1.0573,
      "num_input_tokens_seen": 11041888,
      "step": 19925
    },
    {
      "epoch": 349.65486725663715,
      "grad_norm": 9.5625,
      "learning_rate": 2.51394074515115e-05,
      "loss": 1.2024,
      "num_input_tokens_seen": 11044256,
      "step": 19930
    },
    {
      "epoch": 349.7433628318584,
      "grad_norm": 6.75,
      "learning_rate": 2.5129590116611067e-05,
      "loss": 0.9491,
      "num_input_tokens_seen": 11047680,
      "step": 19935
    },
    {
      "epoch": 349.83185840707966,
      "grad_norm": 8.8125,
      "learning_rate": 2.5119772761726212e-05,
      "loss": 1.0085,
      "num_input_tokens_seen": 11050352,
      "step": 19940
    },
    {
      "epoch": 349.92035398230087,
      "grad_norm": 10.6875,
      "learning_rate": 2.5109955388370893e-05,
      "loss": 1.1603,
      "num_input_tokens_seen": 11052896,
      "step": 19945
    },
    {
      "epoch": 350.0,
      "grad_norm": 14.875,
      "learning_rate": 2.510013799805907e-05,
      "loss": 1.1008,
      "num_input_tokens_seen": 11055552,
      "step": 19950
    },
    {
      "epoch": 350.08849557522126,
      "grad_norm": 8.5625,
      "learning_rate": 2.5090320592304706e-05,
      "loss": 0.9196,
      "num_input_tokens_seen": 11058128,
      "step": 19955
    },
    {
      "epoch": 350.17699115044246,
      "grad_norm": 8.125,
      "learning_rate": 2.5080503172621777e-05,
      "loss": 1.1581,
      "num_input_tokens_seen": 11060800,
      "step": 19960
    },
    {
      "epoch": 350.2654867256637,
      "grad_norm": 9.5,
      "learning_rate": 2.5070685740524246e-05,
      "loss": 1.0032,
      "num_input_tokens_seen": 11063904,
      "step": 19965
    },
    {
      "epoch": 350.353982300885,
      "grad_norm": 12.625,
      "learning_rate": 2.5060868297526084e-05,
      "loss": 1.2757,
      "num_input_tokens_seen": 11066368,
      "step": 19970
    },
    {
      "epoch": 350.4424778761062,
      "grad_norm": 9.875,
      "learning_rate": 2.5051050845141267e-05,
      "loss": 0.841,
      "num_input_tokens_seen": 11068944,
      "step": 19975
    },
    {
      "epoch": 350.53097345132744,
      "grad_norm": 7.75,
      "learning_rate": 2.5041233384883765e-05,
      "loss": 1.2781,
      "num_input_tokens_seen": 11071536,
      "step": 19980
    },
    {
      "epoch": 350.6194690265487,
      "grad_norm": 8.5,
      "learning_rate": 2.5031415918267564e-05,
      "loss": 0.9147,
      "num_input_tokens_seen": 11074448,
      "step": 19985
    },
    {
      "epoch": 350.7079646017699,
      "grad_norm": 8.3125,
      "learning_rate": 2.5021598446806626e-05,
      "loss": 1.0895,
      "num_input_tokens_seen": 11077216,
      "step": 19990
    },
    {
      "epoch": 350.79646017699116,
      "grad_norm": 8.6875,
      "learning_rate": 2.5011780972014937e-05,
      "loss": 1.1353,
      "num_input_tokens_seen": 11080720,
      "step": 19995
    },
    {
      "epoch": 350.88495575221236,
      "grad_norm": 7.15625,
      "learning_rate": 2.5001963495406478e-05,
      "loss": 0.9339,
      "num_input_tokens_seen": 11083136,
      "step": 20000
    },
    {
      "epoch": 350.88495575221236,
      "eval_loss": 1.0204646587371826,
      "eval_runtime": 0.9358,
      "eval_samples_per_second": 26.716,
      "eval_steps_per_second": 13.892,
      "num_input_tokens_seen": 11083136,
      "step": 20000
    },
    {
      "epoch": 350.9734513274336,
      "grad_norm": 9.375,
      "learning_rate": 2.499214601849522e-05,
      "loss": 1.085,
      "num_input_tokens_seen": 11086320,
      "step": 20005
    },
    {
      "epoch": 351.05309734513276,
      "grad_norm": 6.90625,
      "learning_rate": 2.4982328542795148e-05,
      "loss": 0.8212,
      "num_input_tokens_seen": 11088688,
      "step": 20010
    },
    {
      "epoch": 351.14159292035396,
      "grad_norm": 8.375,
      "learning_rate": 2.497251106982024e-05,
      "loss": 1.03,
      "num_input_tokens_seen": 11091504,
      "step": 20015
    },
    {
      "epoch": 351.2300884955752,
      "grad_norm": 8.0,
      "learning_rate": 2.4962693601084458e-05,
      "loss": 1.0079,
      "num_input_tokens_seen": 11094464,
      "step": 20020
    },
    {
      "epoch": 351.3185840707965,
      "grad_norm": 6.8125,
      "learning_rate": 2.4952876138101794e-05,
      "loss": 1.052,
      "num_input_tokens_seen": 11097296,
      "step": 20025
    },
    {
      "epoch": 351.4070796460177,
      "grad_norm": 12.0625,
      "learning_rate": 2.4943058682386233e-05,
      "loss": 1.0331,
      "num_input_tokens_seen": 11099920,
      "step": 20030
    },
    {
      "epoch": 351.49557522123894,
      "grad_norm": 7.96875,
      "learning_rate": 2.493324123545173e-05,
      "loss": 1.086,
      "num_input_tokens_seen": 11103056,
      "step": 20035
    },
    {
      "epoch": 351.5840707964602,
      "grad_norm": 8.125,
      "learning_rate": 2.4923423798812272e-05,
      "loss": 0.9102,
      "num_input_tokens_seen": 11106160,
      "step": 20040
    },
    {
      "epoch": 351.6725663716814,
      "grad_norm": 9.5625,
      "learning_rate": 2.4913606373981825e-05,
      "loss": 1.1453,
      "num_input_tokens_seen": 11109120,
      "step": 20045
    },
    {
      "epoch": 351.76106194690266,
      "grad_norm": 10.0625,
      "learning_rate": 2.4903788962474357e-05,
      "loss": 1.172,
      "num_input_tokens_seen": 11111632,
      "step": 20050
    },
    {
      "epoch": 351.8495575221239,
      "grad_norm": 8.3125,
      "learning_rate": 2.489397156580385e-05,
      "loss": 1.0916,
      "num_input_tokens_seen": 11114416,
      "step": 20055
    },
    {
      "epoch": 351.9380530973451,
      "grad_norm": 8.125,
      "learning_rate": 2.4884154185484246e-05,
      "loss": 1.0653,
      "num_input_tokens_seen": 11117120,
      "step": 20060
    },
    {
      "epoch": 352.01769911504425,
      "grad_norm": 6.75,
      "learning_rate": 2.4874336823029526e-05,
      "loss": 1.144,
      "num_input_tokens_seen": 11119816,
      "step": 20065
    },
    {
      "epoch": 352.1061946902655,
      "grad_norm": 11.3125,
      "learning_rate": 2.4864519479953656e-05,
      "loss": 1.1209,
      "num_input_tokens_seen": 11122248,
      "step": 20070
    },
    {
      "epoch": 352.1946902654867,
      "grad_norm": 10.1875,
      "learning_rate": 2.485470215777058e-05,
      "loss": 1.3793,
      "num_input_tokens_seen": 11124888,
      "step": 20075
    },
    {
      "epoch": 352.283185840708,
      "grad_norm": 17.0,
      "learning_rate": 2.4844884857994258e-05,
      "loss": 1.1826,
      "num_input_tokens_seen": 11127640,
      "step": 20080
    },
    {
      "epoch": 352.37168141592923,
      "grad_norm": 8.75,
      "learning_rate": 2.4835067582138638e-05,
      "loss": 1.0444,
      "num_input_tokens_seen": 11130856,
      "step": 20085
    },
    {
      "epoch": 352.46017699115043,
      "grad_norm": 7.5625,
      "learning_rate": 2.4825250331717666e-05,
      "loss": 0.9859,
      "num_input_tokens_seen": 11134152,
      "step": 20090
    },
    {
      "epoch": 352.5486725663717,
      "grad_norm": 7.59375,
      "learning_rate": 2.4815433108245298e-05,
      "loss": 1.0724,
      "num_input_tokens_seen": 11136568,
      "step": 20095
    },
    {
      "epoch": 352.6371681415929,
      "grad_norm": 8.125,
      "learning_rate": 2.4805615913235456e-05,
      "loss": 0.9958,
      "num_input_tokens_seen": 11139496,
      "step": 20100
    },
    {
      "epoch": 352.72566371681415,
      "grad_norm": 8.0,
      "learning_rate": 2.479579874820208e-05,
      "loss": 0.9402,
      "num_input_tokens_seen": 11142472,
      "step": 20105
    },
    {
      "epoch": 352.8141592920354,
      "grad_norm": 9.25,
      "learning_rate": 2.4785981614659115e-05,
      "loss": 1.125,
      "num_input_tokens_seen": 11145336,
      "step": 20110
    },
    {
      "epoch": 352.9026548672566,
      "grad_norm": 10.0,
      "learning_rate": 2.477616451412047e-05,
      "loss": 1.0171,
      "num_input_tokens_seen": 11148232,
      "step": 20115
    },
    {
      "epoch": 352.9911504424779,
      "grad_norm": 8.625,
      "learning_rate": 2.476634744810007e-05,
      "loss": 0.8266,
      "num_input_tokens_seen": 11150696,
      "step": 20120
    },
    {
      "epoch": 353.070796460177,
      "grad_norm": 8.875,
      "learning_rate": 2.475653041811183e-05,
      "loss": 1.2613,
      "num_input_tokens_seen": 11152904,
      "step": 20125
    },
    {
      "epoch": 353.1592920353982,
      "grad_norm": 6.25,
      "learning_rate": 2.4746713425669652e-05,
      "loss": 1.0136,
      "num_input_tokens_seen": 11155576,
      "step": 20130
    },
    {
      "epoch": 353.24778761061947,
      "grad_norm": 7.5,
      "learning_rate": 2.4736896472287458e-05,
      "loss": 1.063,
      "num_input_tokens_seen": 11158024,
      "step": 20135
    },
    {
      "epoch": 353.3362831858407,
      "grad_norm": 7.21875,
      "learning_rate": 2.4727079559479124e-05,
      "loss": 0.9128,
      "num_input_tokens_seen": 11160632,
      "step": 20140
    },
    {
      "epoch": 353.42477876106193,
      "grad_norm": 8.0625,
      "learning_rate": 2.4717262688758557e-05,
      "loss": 1.0192,
      "num_input_tokens_seen": 11163592,
      "step": 20145
    },
    {
      "epoch": 353.5132743362832,
      "grad_norm": 6.0,
      "learning_rate": 2.4707445861639637e-05,
      "loss": 1.0909,
      "num_input_tokens_seen": 11166648,
      "step": 20150
    },
    {
      "epoch": 353.60176991150445,
      "grad_norm": 7.40625,
      "learning_rate": 2.4697629079636244e-05,
      "loss": 1.1538,
      "num_input_tokens_seen": 11169608,
      "step": 20155
    },
    {
      "epoch": 353.69026548672565,
      "grad_norm": 9.6875,
      "learning_rate": 2.4687812344262244e-05,
      "loss": 0.9789,
      "num_input_tokens_seen": 11172872,
      "step": 20160
    },
    {
      "epoch": 353.7787610619469,
      "grad_norm": 10.0,
      "learning_rate": 2.46779956570315e-05,
      "loss": 1.1967,
      "num_input_tokens_seen": 11175240,
      "step": 20165
    },
    {
      "epoch": 353.86725663716817,
      "grad_norm": 7.125,
      "learning_rate": 2.466817901945787e-05,
      "loss": 0.9968,
      "num_input_tokens_seen": 11178392,
      "step": 20170
    },
    {
      "epoch": 353.95575221238937,
      "grad_norm": 10.0,
      "learning_rate": 2.4658362433055217e-05,
      "loss": 1.1031,
      "num_input_tokens_seen": 11181336,
      "step": 20175
    },
    {
      "epoch": 354.0353982300885,
      "grad_norm": 8.25,
      "learning_rate": 2.4648545899337356e-05,
      "loss": 0.7889,
      "num_input_tokens_seen": 11183416,
      "step": 20180
    },
    {
      "epoch": 354.12389380530976,
      "grad_norm": 8.3125,
      "learning_rate": 2.4638729419818143e-05,
      "loss": 1.1169,
      "num_input_tokens_seen": 11185960,
      "step": 20185
    },
    {
      "epoch": 354.21238938053096,
      "grad_norm": 9.75,
      "learning_rate": 2.46289129960114e-05,
      "loss": 0.9674,
      "num_input_tokens_seen": 11188696,
      "step": 20190
    },
    {
      "epoch": 354.3008849557522,
      "grad_norm": 9.8125,
      "learning_rate": 2.4619096629430924e-05,
      "loss": 1.021,
      "num_input_tokens_seen": 11191032,
      "step": 20195
    },
    {
      "epoch": 354.3893805309734,
      "grad_norm": 8.125,
      "learning_rate": 2.4609280321590543e-05,
      "loss": 1.0463,
      "num_input_tokens_seen": 11193448,
      "step": 20200
    },
    {
      "epoch": 354.3893805309734,
      "eval_loss": 1.0251762866973877,
      "eval_runtime": 0.9341,
      "eval_samples_per_second": 26.764,
      "eval_steps_per_second": 13.917,
      "num_input_tokens_seen": 11193448,
      "step": 20200
    },
    {
      "epoch": 354.4778761061947,
      "grad_norm": 8.8125,
      "learning_rate": 2.4599464074004037e-05,
      "loss": 0.9699,
      "num_input_tokens_seen": 11196648,
      "step": 20205
    },
    {
      "epoch": 354.56637168141594,
      "grad_norm": 6.6875,
      "learning_rate": 2.4589647888185204e-05,
      "loss": 1.0439,
      "num_input_tokens_seen": 11199624,
      "step": 20210
    },
    {
      "epoch": 354.65486725663715,
      "grad_norm": 6.125,
      "learning_rate": 2.4579831765647836e-05,
      "loss": 0.9456,
      "num_input_tokens_seen": 11202776,
      "step": 20215
    },
    {
      "epoch": 354.7433628318584,
      "grad_norm": 8.25,
      "learning_rate": 2.4570015707905676e-05,
      "loss": 1.0122,
      "num_input_tokens_seen": 11205832,
      "step": 20220
    },
    {
      "epoch": 354.83185840707966,
      "grad_norm": 9.5625,
      "learning_rate": 2.4560199716472508e-05,
      "loss": 1.2104,
      "num_input_tokens_seen": 11208984,
      "step": 20225
    },
    {
      "epoch": 354.92035398230087,
      "grad_norm": 7.0625,
      "learning_rate": 2.455038379286207e-05,
      "loss": 1.1444,
      "num_input_tokens_seen": 11211816,
      "step": 20230
    },
    {
      "epoch": 355.0,
      "grad_norm": 11.125,
      "learning_rate": 2.4540567938588095e-05,
      "loss": 1.1284,
      "num_input_tokens_seen": 11214024,
      "step": 20235
    },
    {
      "epoch": 355.08849557522126,
      "grad_norm": 9.5,
      "learning_rate": 2.4530752155164328e-05,
      "loss": 0.8633,
      "num_input_tokens_seen": 11217336,
      "step": 20240
    },
    {
      "epoch": 355.17699115044246,
      "grad_norm": 8.875,
      "learning_rate": 2.4520936444104463e-05,
      "loss": 1.1289,
      "num_input_tokens_seen": 11220072,
      "step": 20245
    },
    {
      "epoch": 355.2654867256637,
      "grad_norm": 7.875,
      "learning_rate": 2.4511120806922218e-05,
      "loss": 1.04,
      "num_input_tokens_seen": 11222472,
      "step": 20250
    },
    {
      "epoch": 355.353982300885,
      "grad_norm": 8.8125,
      "learning_rate": 2.45013052451313e-05,
      "loss": 1.0066,
      "num_input_tokens_seen": 11225464,
      "step": 20255
    },
    {
      "epoch": 355.4424778761062,
      "grad_norm": 7.125,
      "learning_rate": 2.4491489760245376e-05,
      "loss": 0.9807,
      "num_input_tokens_seen": 11228024,
      "step": 20260
    },
    {
      "epoch": 355.53097345132744,
      "grad_norm": 8.5,
      "learning_rate": 2.4481674353778115e-05,
      "loss": 1.0602,
      "num_input_tokens_seen": 11230840,
      "step": 20265
    },
    {
      "epoch": 355.6194690265487,
      "grad_norm": 8.75,
      "learning_rate": 2.447185902724319e-05,
      "loss": 1.1655,
      "num_input_tokens_seen": 11233768,
      "step": 20270
    },
    {
      "epoch": 355.7079646017699,
      "grad_norm": 8.5625,
      "learning_rate": 2.4462043782154233e-05,
      "loss": 0.9655,
      "num_input_tokens_seen": 11236424,
      "step": 20275
    },
    {
      "epoch": 355.79646017699116,
      "grad_norm": 9.1875,
      "learning_rate": 2.4452228620024895e-05,
      "loss": 1.191,
      "num_input_tokens_seen": 11239640,
      "step": 20280
    },
    {
      "epoch": 355.88495575221236,
      "grad_norm": 8.5625,
      "learning_rate": 2.4442413542368776e-05,
      "loss": 1.0449,
      "num_input_tokens_seen": 11241944,
      "step": 20285
    },
    {
      "epoch": 355.9734513274336,
      "grad_norm": 7.125,
      "learning_rate": 2.4432598550699502e-05,
      "loss": 1.1527,
      "num_input_tokens_seen": 11244584,
      "step": 20290
    },
    {
      "epoch": 356.05309734513276,
      "grad_norm": 6.90625,
      "learning_rate": 2.4422783646530663e-05,
      "loss": 1.0407,
      "num_input_tokens_seen": 11247448,
      "step": 20295
    },
    {
      "epoch": 356.14159292035396,
      "grad_norm": 8.4375,
      "learning_rate": 2.441296883137584e-05,
      "loss": 1.1909,
      "num_input_tokens_seen": 11250024,
      "step": 20300
    },
    {
      "epoch": 356.2300884955752,
      "grad_norm": 8.4375,
      "learning_rate": 2.4403154106748592e-05,
      "loss": 0.9576,
      "num_input_tokens_seen": 11252728,
      "step": 20305
    },
    {
      "epoch": 356.3185840707965,
      "grad_norm": 8.125,
      "learning_rate": 2.4393339474162494e-05,
      "loss": 0.8987,
      "num_input_tokens_seen": 11255432,
      "step": 20310
    },
    {
      "epoch": 356.4070796460177,
      "grad_norm": 8.0625,
      "learning_rate": 2.4383524935131062e-05,
      "loss": 1.2015,
      "num_input_tokens_seen": 11257992,
      "step": 20315
    },
    {
      "epoch": 356.49557522123894,
      "grad_norm": 8.75,
      "learning_rate": 2.437371049116784e-05,
      "loss": 1.1143,
      "num_input_tokens_seen": 11261272,
      "step": 20320
    },
    {
      "epoch": 356.5840707964602,
      "grad_norm": 7.875,
      "learning_rate": 2.436389614378632e-05,
      "loss": 0.9099,
      "num_input_tokens_seen": 11263944,
      "step": 20325
    },
    {
      "epoch": 356.6725663716814,
      "grad_norm": 7.90625,
      "learning_rate": 2.435408189450002e-05,
      "loss": 0.9365,
      "num_input_tokens_seen": 11266792,
      "step": 20330
    },
    {
      "epoch": 356.76106194690266,
      "grad_norm": 7.15625,
      "learning_rate": 2.4344267744822406e-05,
      "loss": 1.1007,
      "num_input_tokens_seen": 11269896,
      "step": 20335
    },
    {
      "epoch": 356.8495575221239,
      "grad_norm": 9.0,
      "learning_rate": 2.4334453696266944e-05,
      "loss": 1.2299,
      "num_input_tokens_seen": 11272888,
      "step": 20340
    },
    {
      "epoch": 356.9380530973451,
      "grad_norm": 8.4375,
      "learning_rate": 2.432463975034708e-05,
      "loss": 1.187,
      "num_input_tokens_seen": 11275656,
      "step": 20345
    },
    {
      "epoch": 357.01769911504425,
      "grad_norm": 8.4375,
      "learning_rate": 2.4314825908576265e-05,
      "loss": 0.958,
      "num_input_tokens_seen": 11277808,
      "step": 20350
    },
    {
      "epoch": 357.1061946902655,
      "grad_norm": 7.375,
      "learning_rate": 2.4305012172467897e-05,
      "loss": 1.0035,
      "num_input_tokens_seen": 11280912,
      "step": 20355
    },
    {
      "epoch": 357.1946902654867,
      "grad_norm": 9.375,
      "learning_rate": 2.4295198543535393e-05,
      "loss": 1.0296,
      "num_input_tokens_seen": 11283360,
      "step": 20360
    },
    {
      "epoch": 357.283185840708,
      "grad_norm": 6.6875,
      "learning_rate": 2.4285385023292124e-05,
      "loss": 1.0311,
      "num_input_tokens_seen": 11285632,
      "step": 20365
    },
    {
      "epoch": 357.37168141592923,
      "grad_norm": 7.9375,
      "learning_rate": 2.427557161325147e-05,
      "loss": 1.1114,
      "num_input_tokens_seen": 11288624,
      "step": 20370
    },
    {
      "epoch": 357.46017699115043,
      "grad_norm": 8.5625,
      "learning_rate": 2.4265758314926778e-05,
      "loss": 0.9587,
      "num_input_tokens_seen": 11291536,
      "step": 20375
    },
    {
      "epoch": 357.5486725663717,
      "grad_norm": 7.03125,
      "learning_rate": 2.4255945129831373e-05,
      "loss": 1.1054,
      "num_input_tokens_seen": 11294288,
      "step": 20380
    },
    {
      "epoch": 357.6371681415929,
      "grad_norm": 7.9375,
      "learning_rate": 2.4246132059478578e-05,
      "loss": 1.1767,
      "num_input_tokens_seen": 11296864,
      "step": 20385
    },
    {
      "epoch": 357.72566371681415,
      "grad_norm": 7.125,
      "learning_rate": 2.4236319105381706e-05,
      "loss": 0.9022,
      "num_input_tokens_seen": 11299216,
      "step": 20390
    },
    {
      "epoch": 357.8141592920354,
      "grad_norm": 6.65625,
      "learning_rate": 2.422650626905401e-05,
      "loss": 1.0164,
      "num_input_tokens_seen": 11301920,
      "step": 20395
    },
    {
      "epoch": 357.9026548672566,
      "grad_norm": 9.75,
      "learning_rate": 2.4216693552008785e-05,
      "loss": 1.0393,
      "num_input_tokens_seen": 11305168,
      "step": 20400
    },
    {
      "epoch": 357.9026548672566,
      "eval_loss": 1.0222878456115723,
      "eval_runtime": 0.9361,
      "eval_samples_per_second": 26.708,
      "eval_steps_per_second": 13.888,
      "num_input_tokens_seen": 11305168,
      "step": 20400
    },
    {
      "epoch": 357.9911504424779,
      "grad_norm": 10.625,
      "learning_rate": 2.4206880955759247e-05,
      "loss": 1.1431,
      "num_input_tokens_seen": 11308240,
      "step": 20405
    },
    {
      "epoch": 358.070796460177,
      "grad_norm": 8.125,
      "learning_rate": 2.419706848181863e-05,
      "loss": 0.8668,
      "num_input_tokens_seen": 11310736,
      "step": 20410
    },
    {
      "epoch": 358.1592920353982,
      "grad_norm": 7.03125,
      "learning_rate": 2.4187256131700153e-05,
      "loss": 1.1008,
      "num_input_tokens_seen": 11313840,
      "step": 20415
    },
    {
      "epoch": 358.24778761061947,
      "grad_norm": 7.375,
      "learning_rate": 2.4177443906916985e-05,
      "loss": 1.1674,
      "num_input_tokens_seen": 11316224,
      "step": 20420
    },
    {
      "epoch": 358.3362831858407,
      "grad_norm": 7.125,
      "learning_rate": 2.4167631808982303e-05,
      "loss": 0.9309,
      "num_input_tokens_seen": 11319392,
      "step": 20425
    },
    {
      "epoch": 358.42477876106193,
      "grad_norm": 10.75,
      "learning_rate": 2.4157819839409264e-05,
      "loss": 0.952,
      "num_input_tokens_seen": 11322032,
      "step": 20430
    },
    {
      "epoch": 358.5132743362832,
      "grad_norm": 7.125,
      "learning_rate": 2.414800799971098e-05,
      "loss": 1.1104,
      "num_input_tokens_seen": 11324608,
      "step": 20435
    },
    {
      "epoch": 358.60176991150445,
      "grad_norm": 7.125,
      "learning_rate": 2.4138196291400582e-05,
      "loss": 1.1538,
      "num_input_tokens_seen": 11327664,
      "step": 20440
    },
    {
      "epoch": 358.69026548672565,
      "grad_norm": 7.8125,
      "learning_rate": 2.412838471599114e-05,
      "loss": 1.074,
      "num_input_tokens_seen": 11330768,
      "step": 20445
    },
    {
      "epoch": 358.7787610619469,
      "grad_norm": 6.28125,
      "learning_rate": 2.411857327499572e-05,
      "loss": 1.008,
      "num_input_tokens_seen": 11333184,
      "step": 20450
    },
    {
      "epoch": 358.86725663716817,
      "grad_norm": 8.125,
      "learning_rate": 2.410876196992739e-05,
      "loss": 0.9676,
      "num_input_tokens_seen": 11336320,
      "step": 20455
    },
    {
      "epoch": 358.95575221238937,
      "grad_norm": 12.3125,
      "learning_rate": 2.4098950802299156e-05,
      "loss": 1.1045,
      "num_input_tokens_seen": 11339008,
      "step": 20460
    },
    {
      "epoch": 359.0353982300885,
      "grad_norm": 11.75,
      "learning_rate": 2.4089139773624027e-05,
      "loss": 1.0253,
      "num_input_tokens_seen": 11341360,
      "step": 20465
    },
    {
      "epoch": 359.12389380530976,
      "grad_norm": 9.5625,
      "learning_rate": 2.4079328885415007e-05,
      "loss": 1.0813,
      "num_input_tokens_seen": 11344016,
      "step": 20470
    },
    {
      "epoch": 359.21238938053096,
      "grad_norm": 7.6875,
      "learning_rate": 2.4069518139185036e-05,
      "loss": 0.9438,
      "num_input_tokens_seen": 11346960,
      "step": 20475
    },
    {
      "epoch": 359.3008849557522,
      "grad_norm": 8.375,
      "learning_rate": 2.405970753644706e-05,
      "loss": 1.0873,
      "num_input_tokens_seen": 11350000,
      "step": 20480
    },
    {
      "epoch": 359.3893805309734,
      "grad_norm": 12.375,
      "learning_rate": 2.4049897078714e-05,
      "loss": 1.2661,
      "num_input_tokens_seen": 11352576,
      "step": 20485
    },
    {
      "epoch": 359.4778761061947,
      "grad_norm": 7.625,
      "learning_rate": 2.404008676749874e-05,
      "loss": 1.0904,
      "num_input_tokens_seen": 11354880,
      "step": 20490
    },
    {
      "epoch": 359.56637168141594,
      "grad_norm": 10.8125,
      "learning_rate": 2.403027660431418e-05,
      "loss": 0.9844,
      "num_input_tokens_seen": 11358176,
      "step": 20495
    },
    {
      "epoch": 359.65486725663715,
      "grad_norm": 6.46875,
      "learning_rate": 2.402046659067314e-05,
      "loss": 0.9993,
      "num_input_tokens_seen": 11361136,
      "step": 20500
    },
    {
      "epoch": 359.7433628318584,
      "grad_norm": 8.25,
      "learning_rate": 2.401065672808847e-05,
      "loss": 1.087,
      "num_input_tokens_seen": 11363920,
      "step": 20505
    },
    {
      "epoch": 359.83185840707966,
      "grad_norm": 7.34375,
      "learning_rate": 2.400084701807296e-05,
      "loss": 1.1708,
      "num_input_tokens_seen": 11366752,
      "step": 20510
    },
    {
      "epoch": 359.92035398230087,
      "grad_norm": 8.6875,
      "learning_rate": 2.39910374621394e-05,
      "loss": 1.0802,
      "num_input_tokens_seen": 11369632,
      "step": 20515
    },
    {
      "epoch": 360.0,
      "grad_norm": 7.75,
      "learning_rate": 2.3981228061800544e-05,
      "loss": 0.9536,
      "num_input_tokens_seen": 11372432,
      "step": 20520
    },
    {
      "epoch": 360.08849557522126,
      "grad_norm": 7.03125,
      "learning_rate": 2.3971418818569115e-05,
      "loss": 1.1667,
      "num_input_tokens_seen": 11374832,
      "step": 20525
    },
    {
      "epoch": 360.17699115044246,
      "grad_norm": 8.0,
      "learning_rate": 2.3961609733957832e-05,
      "loss": 1.2504,
      "num_input_tokens_seen": 11377648,
      "step": 20530
    },
    {
      "epoch": 360.2654867256637,
      "grad_norm": 7.0,
      "learning_rate": 2.395180080947939e-05,
      "loss": 0.976,
      "num_input_tokens_seen": 11380224,
      "step": 20535
    },
    {
      "epoch": 360.353982300885,
      "grad_norm": 7.5,
      "learning_rate": 2.394199204664642e-05,
      "loss": 0.9375,
      "num_input_tokens_seen": 11383216,
      "step": 20540
    },
    {
      "epoch": 360.4424778761062,
      "grad_norm": 8.0,
      "learning_rate": 2.3932183446971583e-05,
      "loss": 1.1247,
      "num_input_tokens_seen": 11385744,
      "step": 20545
    },
    {
      "epoch": 360.53097345132744,
      "grad_norm": 7.5,
      "learning_rate": 2.3922375011967473e-05,
      "loss": 0.9187,
      "num_input_tokens_seen": 11388768,
      "step": 20550
    },
    {
      "epoch": 360.6194690265487,
      "grad_norm": 8.125,
      "learning_rate": 2.3912566743146676e-05,
      "loss": 1.0072,
      "num_input_tokens_seen": 11392096,
      "step": 20555
    },
    {
      "epoch": 360.7079646017699,
      "grad_norm": 7.5,
      "learning_rate": 2.390275864202176e-05,
      "loss": 1.0726,
      "num_input_tokens_seen": 11394784,
      "step": 20560
    },
    {
      "epoch": 360.79646017699116,
      "grad_norm": 8.5,
      "learning_rate": 2.3892950710105243e-05,
      "loss": 1.056,
      "num_input_tokens_seen": 11397888,
      "step": 20565
    },
    {
      "epoch": 360.88495575221236,
      "grad_norm": 6.4375,
      "learning_rate": 2.3883142948909635e-05,
      "loss": 1.0035,
      "num_input_tokens_seen": 11400640,
      "step": 20570
    },
    {
      "epoch": 360.9734513274336,
      "grad_norm": 7.75,
      "learning_rate": 2.3873335359947433e-05,
      "loss": 1.1562,
      "num_input_tokens_seen": 11403312,
      "step": 20575
    },
    {
      "epoch": 361.05309734513276,
      "grad_norm": 6.96875,
      "learning_rate": 2.3863527944731066e-05,
      "loss": 1.3069,
      "num_input_tokens_seen": 11405184,
      "step": 20580
    },
    {
      "epoch": 361.14159292035396,
      "grad_norm": 8.1875,
      "learning_rate": 2.385372070477298e-05,
      "loss": 1.0549,
      "num_input_tokens_seen": 11408400,
      "step": 20585
    },
    {
      "epoch": 361.2300884955752,
      "grad_norm": 10.0,
      "learning_rate": 2.384391364158556e-05,
      "loss": 1.0894,
      "num_input_tokens_seen": 11410912,
      "step": 20590
    },
    {
      "epoch": 361.3185840707965,
      "grad_norm": 8.25,
      "learning_rate": 2.3834106756681185e-05,
      "loss": 0.9157,
      "num_input_tokens_seen": 11413680,
      "step": 20595
    },
    {
      "epoch": 361.4070796460177,
      "grad_norm": 7.1875,
      "learning_rate": 2.3824300051572206e-05,
      "loss": 0.8809,
      "num_input_tokens_seen": 11416112,
      "step": 20600
    },
    {
      "epoch": 361.4070796460177,
      "eval_loss": 1.0299760103225708,
      "eval_runtime": 0.942,
      "eval_samples_per_second": 26.54,
      "eval_steps_per_second": 13.801,
      "num_input_tokens_seen": 11416112,
      "step": 20600
    },
    {
      "epoch": 361.49557522123894,
      "grad_norm": 9.5625,
      "learning_rate": 2.3814493527770923e-05,
      "loss": 0.9911,
      "num_input_tokens_seen": 11419152,
      "step": 20605
    },
    {
      "epoch": 361.5840707964602,
      "grad_norm": 6.875,
      "learning_rate": 2.3804687186789637e-05,
      "loss": 1.007,
      "num_input_tokens_seen": 11421616,
      "step": 20610
    },
    {
      "epoch": 361.6725663716814,
      "grad_norm": 6.625,
      "learning_rate": 2.379488103014062e-05,
      "loss": 0.9846,
      "num_input_tokens_seen": 11424224,
      "step": 20615
    },
    {
      "epoch": 361.76106194690266,
      "grad_norm": 9.8125,
      "learning_rate": 2.3785075059336086e-05,
      "loss": 1.1964,
      "num_input_tokens_seen": 11427392,
      "step": 20620
    },
    {
      "epoch": 361.8495575221239,
      "grad_norm": 13.4375,
      "learning_rate": 2.3775269275888248e-05,
      "loss": 1.0925,
      "num_input_tokens_seen": 11430448,
      "step": 20625
    },
    {
      "epoch": 361.9380530973451,
      "grad_norm": 14.3125,
      "learning_rate": 2.3765463681309274e-05,
      "loss": 1.1221,
      "num_input_tokens_seen": 11433904,
      "step": 20630
    },
    {
      "epoch": 362.01769911504425,
      "grad_norm": 9.25,
      "learning_rate": 2.3755658277111313e-05,
      "loss": 1.1338,
      "num_input_tokens_seen": 11436168,
      "step": 20635
    },
    {
      "epoch": 362.1061946902655,
      "grad_norm": 6.6875,
      "learning_rate": 2.374585306480649e-05,
      "loss": 0.9405,
      "num_input_tokens_seen": 11438504,
      "step": 20640
    },
    {
      "epoch": 362.1946902654867,
      "grad_norm": 8.0,
      "learning_rate": 2.3736048045906877e-05,
      "loss": 1.121,
      "num_input_tokens_seen": 11441272,
      "step": 20645
    },
    {
      "epoch": 362.283185840708,
      "grad_norm": 11.0,
      "learning_rate": 2.372624322192454e-05,
      "loss": 1.1265,
      "num_input_tokens_seen": 11444280,
      "step": 20650
    },
    {
      "epoch": 362.37168141592923,
      "grad_norm": 8.1875,
      "learning_rate": 2.3716438594371516e-05,
      "loss": 1.0424,
      "num_input_tokens_seen": 11447416,
      "step": 20655
    },
    {
      "epoch": 362.46017699115043,
      "grad_norm": 11.25,
      "learning_rate": 2.3706634164759784e-05,
      "loss": 1.0959,
      "num_input_tokens_seen": 11450568,
      "step": 20660
    },
    {
      "epoch": 362.5486725663717,
      "grad_norm": 9.375,
      "learning_rate": 2.3696829934601323e-05,
      "loss": 0.9686,
      "num_input_tokens_seen": 11453224,
      "step": 20665
    },
    {
      "epoch": 362.6371681415929,
      "grad_norm": 7.8125,
      "learning_rate": 2.3687025905408053e-05,
      "loss": 0.9811,
      "num_input_tokens_seen": 11455544,
      "step": 20670
    },
    {
      "epoch": 362.72566371681415,
      "grad_norm": 8.375,
      "learning_rate": 2.3677222078691886e-05,
      "loss": 1.0358,
      "num_input_tokens_seen": 11458312,
      "step": 20675
    },
    {
      "epoch": 362.8141592920354,
      "grad_norm": 8.3125,
      "learning_rate": 2.366741845596471e-05,
      "loss": 1.1261,
      "num_input_tokens_seen": 11461432,
      "step": 20680
    },
    {
      "epoch": 362.9026548672566,
      "grad_norm": 9.125,
      "learning_rate": 2.3657615038738343e-05,
      "loss": 1.0093,
      "num_input_tokens_seen": 11464632,
      "step": 20685
    },
    {
      "epoch": 362.9911504424779,
      "grad_norm": 10.0,
      "learning_rate": 2.3647811828524614e-05,
      "loss": 1.0635,
      "num_input_tokens_seen": 11467080,
      "step": 20690
    },
    {
      "epoch": 363.070796460177,
      "grad_norm": 8.3125,
      "learning_rate": 2.363800882683529e-05,
      "loss": 1.0905,
      "num_input_tokens_seen": 11468984,
      "step": 20695
    },
    {
      "epoch": 363.1592920353982,
      "grad_norm": 7.59375,
      "learning_rate": 2.3628206035182125e-05,
      "loss": 1.0109,
      "num_input_tokens_seen": 11471448,
      "step": 20700
    },
    {
      "epoch": 363.24778761061947,
      "grad_norm": 6.375,
      "learning_rate": 2.361840345507683e-05,
      "loss": 1.0605,
      "num_input_tokens_seen": 11474472,
      "step": 20705
    },
    {
      "epoch": 363.3362831858407,
      "grad_norm": 8.0625,
      "learning_rate": 2.3608601088031073e-05,
      "loss": 1.0542,
      "num_input_tokens_seen": 11477352,
      "step": 20710
    },
    {
      "epoch": 363.42477876106193,
      "grad_norm": 7.9375,
      "learning_rate": 2.3598798935556516e-05,
      "loss": 0.8988,
      "num_input_tokens_seen": 11480616,
      "step": 20715
    },
    {
      "epoch": 363.5132743362832,
      "grad_norm": 8.3125,
      "learning_rate": 2.3588996999164784e-05,
      "loss": 1.1549,
      "num_input_tokens_seen": 11483496,
      "step": 20720
    },
    {
      "epoch": 363.60176991150445,
      "grad_norm": 6.96875,
      "learning_rate": 2.3579195280367434e-05,
      "loss": 1.029,
      "num_input_tokens_seen": 11485912,
      "step": 20725
    },
    {
      "epoch": 363.69026548672565,
      "grad_norm": 8.75,
      "learning_rate": 2.356939378067603e-05,
      "loss": 0.9812,
      "num_input_tokens_seen": 11488552,
      "step": 20730
    },
    {
      "epoch": 363.7787610619469,
      "grad_norm": 8.1875,
      "learning_rate": 2.3559592501602092e-05,
      "loss": 1.1537,
      "num_input_tokens_seen": 11492008,
      "step": 20735
    },
    {
      "epoch": 363.86725663716817,
      "grad_norm": 8.5,
      "learning_rate": 2.3549791444657076e-05,
      "loss": 1.1551,
      "num_input_tokens_seen": 11495000,
      "step": 20740
    },
    {
      "epoch": 363.95575221238937,
      "grad_norm": 8.25,
      "learning_rate": 2.353999061135246e-05,
      "loss": 0.964,
      "num_input_tokens_seen": 11497720,
      "step": 20745
    },
    {
      "epoch": 364.0353982300885,
      "grad_norm": 8.875,
      "learning_rate": 2.3530190003199626e-05,
      "loss": 1.7258,
      "num_input_tokens_seen": 11499824,
      "step": 20750
    },
    {
      "epoch": 364.12389380530976,
      "grad_norm": 9.4375,
      "learning_rate": 2.3520389621709965e-05,
      "loss": 1.1665,
      "num_input_tokens_seen": 11502464,
      "step": 20755
    },
    {
      "epoch": 364.21238938053096,
      "grad_norm": 6.46875,
      "learning_rate": 2.351058946839483e-05,
      "loss": 0.9267,
      "num_input_tokens_seen": 11505312,
      "step": 20760
    },
    {
      "epoch": 364.3008849557522,
      "grad_norm": 7.0625,
      "learning_rate": 2.350078954476551e-05,
      "loss": 0.9276,
      "num_input_tokens_seen": 11507888,
      "step": 20765
    },
    {
      "epoch": 364.3893805309734,
      "grad_norm": 13.25,
      "learning_rate": 2.3490989852333272e-05,
      "loss": 1.1853,
      "num_input_tokens_seen": 11510304,
      "step": 20770
    },
    {
      "epoch": 364.4778761061947,
      "grad_norm": 10.0625,
      "learning_rate": 2.3481190392609377e-05,
      "loss": 0.9236,
      "num_input_tokens_seen": 11513376,
      "step": 20775
    },
    {
      "epoch": 364.56637168141594,
      "grad_norm": 7.8125,
      "learning_rate": 2.3471391167105e-05,
      "loss": 1.0683,
      "num_input_tokens_seen": 11515664,
      "step": 20780
    },
    {
      "epoch": 364.65486725663715,
      "grad_norm": 10.125,
      "learning_rate": 2.3461592177331325e-05,
      "loss": 1.0511,
      "num_input_tokens_seen": 11518768,
      "step": 20785
    },
    {
      "epoch": 364.7433628318584,
      "grad_norm": 6.625,
      "learning_rate": 2.345179342479946e-05,
      "loss": 0.9402,
      "num_input_tokens_seen": 11521616,
      "step": 20790
    },
    {
      "epoch": 364.83185840707966,
      "grad_norm": 8.4375,
      "learning_rate": 2.3441994911020503e-05,
      "loss": 1.1078,
      "num_input_tokens_seen": 11524480,
      "step": 20795
    },
    {
      "epoch": 364.92035398230087,
      "grad_norm": 6.90625,
      "learning_rate": 2.3432196637505522e-05,
      "loss": 1.1824,
      "num_input_tokens_seen": 11527424,
      "step": 20800
    },
    {
      "epoch": 364.92035398230087,
      "eval_loss": 1.0257512331008911,
      "eval_runtime": 0.9339,
      "eval_samples_per_second": 26.77,
      "eval_steps_per_second": 13.92,
      "num_input_tokens_seen": 11527424,
      "step": 20800
    },
    {
      "epoch": 365.0,
      "grad_norm": 23.75,
      "learning_rate": 2.3422398605765515e-05,
      "loss": 0.9786,
      "num_input_tokens_seen": 11530208,
      "step": 20805
    },
    {
      "epoch": 365.08849557522126,
      "grad_norm": 8.25,
      "learning_rate": 2.3412600817311462e-05,
      "loss": 1.1048,
      "num_input_tokens_seen": 11532800,
      "step": 20810
    },
    {
      "epoch": 365.17699115044246,
      "grad_norm": 8.3125,
      "learning_rate": 2.3402803273654326e-05,
      "loss": 0.9797,
      "num_input_tokens_seen": 11535456,
      "step": 20815
    },
    {
      "epoch": 365.2654867256637,
      "grad_norm": 9.125,
      "learning_rate": 2.3393005976304983e-05,
      "loss": 1.1075,
      "num_input_tokens_seen": 11538144,
      "step": 20820
    },
    {
      "epoch": 365.353982300885,
      "grad_norm": 10.0625,
      "learning_rate": 2.338320892677432e-05,
      "loss": 1.1107,
      "num_input_tokens_seen": 11541152,
      "step": 20825
    },
    {
      "epoch": 365.4424778761062,
      "grad_norm": 7.53125,
      "learning_rate": 2.3373412126573155e-05,
      "loss": 1.032,
      "num_input_tokens_seen": 11543904,
      "step": 20830
    },
    {
      "epoch": 365.53097345132744,
      "grad_norm": 12.0,
      "learning_rate": 2.3363615577212285e-05,
      "loss": 0.9796,
      "num_input_tokens_seen": 11546784,
      "step": 20835
    },
    {
      "epoch": 365.6194690265487,
      "grad_norm": 8.25,
      "learning_rate": 2.3353819280202455e-05,
      "loss": 1.0478,
      "num_input_tokens_seen": 11550000,
      "step": 20840
    },
    {
      "epoch": 365.7079646017699,
      "grad_norm": 7.90625,
      "learning_rate": 2.334402323705438e-05,
      "loss": 0.9182,
      "num_input_tokens_seen": 11552704,
      "step": 20845
    },
    {
      "epoch": 365.79646017699116,
      "grad_norm": 11.0,
      "learning_rate": 2.3334227449278725e-05,
      "loss": 1.0574,
      "num_input_tokens_seen": 11555168,
      "step": 20850
    },
    {
      "epoch": 365.88495575221236,
      "grad_norm": 7.25,
      "learning_rate": 2.3324431918386143e-05,
      "loss": 0.9714,
      "num_input_tokens_seen": 11558448,
      "step": 20855
    },
    {
      "epoch": 365.9734513274336,
      "grad_norm": 7.46875,
      "learning_rate": 2.3314636645887207e-05,
      "loss": 1.1687,
      "num_input_tokens_seen": 11561408,
      "step": 20860
    },
    {
      "epoch": 366.05309734513276,
      "grad_norm": 9.375,
      "learning_rate": 2.3304841633292487e-05,
      "loss": 1.1374,
      "num_input_tokens_seen": 11563704,
      "step": 20865
    },
    {
      "epoch": 366.14159292035396,
      "grad_norm": 8.3125,
      "learning_rate": 2.329504688211248e-05,
      "loss": 0.9185,
      "num_input_tokens_seen": 11566680,
      "step": 20870
    },
    {
      "epoch": 366.2300884955752,
      "grad_norm": 9.9375,
      "learning_rate": 2.3285252393857677e-05,
      "loss": 0.9408,
      "num_input_tokens_seen": 11569512,
      "step": 20875
    },
    {
      "epoch": 366.3185840707965,
      "grad_norm": 12.0,
      "learning_rate": 2.327545817003851e-05,
      "loss": 1.0796,
      "num_input_tokens_seen": 11572456,
      "step": 20880
    },
    {
      "epoch": 366.4070796460177,
      "grad_norm": 8.5625,
      "learning_rate": 2.326566421216535e-05,
      "loss": 1.1184,
      "num_input_tokens_seen": 11575112,
      "step": 20885
    },
    {
      "epoch": 366.49557522123894,
      "grad_norm": 8.625,
      "learning_rate": 2.3255870521748565e-05,
      "loss": 1.1541,
      "num_input_tokens_seen": 11577704,
      "step": 20890
    },
    {
      "epoch": 366.5840707964602,
      "grad_norm": 7.96875,
      "learning_rate": 2.3246077100298474e-05,
      "loss": 1.1962,
      "num_input_tokens_seen": 11580456,
      "step": 20895
    },
    {
      "epoch": 366.6725663716814,
      "grad_norm": 9.125,
      "learning_rate": 2.3236283949325328e-05,
      "loss": 1.0574,
      "num_input_tokens_seen": 11583416,
      "step": 20900
    },
    {
      "epoch": 366.76106194690266,
      "grad_norm": 9.4375,
      "learning_rate": 2.3226491070339368e-05,
      "loss": 1.1303,
      "num_input_tokens_seen": 11586040,
      "step": 20905
    },
    {
      "epoch": 366.8495575221239,
      "grad_norm": 9.25,
      "learning_rate": 2.3216698464850762e-05,
      "loss": 1.0028,
      "num_input_tokens_seen": 11588744,
      "step": 20910
    },
    {
      "epoch": 366.9380530973451,
      "grad_norm": 5.96875,
      "learning_rate": 2.320690613436967e-05,
      "loss": 0.9168,
      "num_input_tokens_seen": 11591704,
      "step": 20915
    },
    {
      "epoch": 367.01769911504425,
      "grad_norm": 9.0625,
      "learning_rate": 2.3197114080406192e-05,
      "loss": 1.3943,
      "num_input_tokens_seen": 11594304,
      "step": 20920
    },
    {
      "epoch": 367.1061946902655,
      "grad_norm": 7.40625,
      "learning_rate": 2.3187322304470365e-05,
      "loss": 1.0462,
      "num_input_tokens_seen": 11597104,
      "step": 20925
    },
    {
      "epoch": 367.1946902654867,
      "grad_norm": 9.5,
      "learning_rate": 2.3177530808072222e-05,
      "loss": 1.0078,
      "num_input_tokens_seen": 11599616,
      "step": 20930
    },
    {
      "epoch": 367.283185840708,
      "grad_norm": 8.6875,
      "learning_rate": 2.316773959272174e-05,
      "loss": 1.0632,
      "num_input_tokens_seen": 11602352,
      "step": 20935
    },
    {
      "epoch": 367.37168141592923,
      "grad_norm": 8.6875,
      "learning_rate": 2.3157948659928823e-05,
      "loss": 1.0839,
      "num_input_tokens_seen": 11605360,
      "step": 20940
    },
    {
      "epoch": 367.46017699115043,
      "grad_norm": 6.9375,
      "learning_rate": 2.3148158011203388e-05,
      "loss": 0.9389,
      "num_input_tokens_seen": 11608000,
      "step": 20945
    },
    {
      "epoch": 367.5486725663717,
      "grad_norm": 8.75,
      "learning_rate": 2.3138367648055253e-05,
      "loss": 1.0092,
      "num_input_tokens_seen": 11611088,
      "step": 20950
    },
    {
      "epoch": 367.6371681415929,
      "grad_norm": 8.3125,
      "learning_rate": 2.312857757199422e-05,
      "loss": 1.1525,
      "num_input_tokens_seen": 11614288,
      "step": 20955
    },
    {
      "epoch": 367.72566371681415,
      "grad_norm": 10.5,
      "learning_rate": 2.3118787784530048e-05,
      "loss": 1.1553,
      "num_input_tokens_seen": 11616832,
      "step": 20960
    },
    {
      "epoch": 367.8141592920354,
      "grad_norm": 9.25,
      "learning_rate": 2.310899828717243e-05,
      "loss": 1.2569,
      "num_input_tokens_seen": 11619616,
      "step": 20965
    },
    {
      "epoch": 367.9026548672566,
      "grad_norm": 6.25,
      "learning_rate": 2.309920908143104e-05,
      "loss": 0.8588,
      "num_input_tokens_seen": 11622352,
      "step": 20970
    },
    {
      "epoch": 367.9911504424779,
      "grad_norm": 6.90625,
      "learning_rate": 2.308942016881551e-05,
      "loss": 0.9579,
      "num_input_tokens_seen": 11624816,
      "step": 20975
    },
    {
      "epoch": 368.070796460177,
      "grad_norm": 11.625,
      "learning_rate": 2.307963155083539e-05,
      "loss": 1.1741,
      "num_input_tokens_seen": 11626856,
      "step": 20980
    },
    {
      "epoch": 368.1592920353982,
      "grad_norm": 9.75,
      "learning_rate": 2.306984322900022e-05,
      "loss": 1.0739,
      "num_input_tokens_seen": 11629736,
      "step": 20985
    },
    {
      "epoch": 368.24778761061947,
      "grad_norm": 9.1875,
      "learning_rate": 2.3060055204819482e-05,
      "loss": 1.0003,
      "num_input_tokens_seen": 11632744,
      "step": 20990
    },
    {
      "epoch": 368.3362831858407,
      "grad_norm": 8.4375,
      "learning_rate": 2.3050267479802604e-05,
      "loss": 0.9413,
      "num_input_tokens_seen": 11635240,
      "step": 20995
    },
    {
      "epoch": 368.42477876106193,
      "grad_norm": 7.75,
      "learning_rate": 2.304048005545899e-05,
      "loss": 0.8426,
      "num_input_tokens_seen": 11637784,
      "step": 21000
    },
    {
      "epoch": 368.42477876106193,
      "eval_loss": 1.0242201089859009,
      "eval_runtime": 0.9354,
      "eval_samples_per_second": 26.726,
      "eval_steps_per_second": 13.898,
      "num_input_tokens_seen": 11637784,
      "step": 21000
    },
    {
      "epoch": 368.5132743362832,
      "grad_norm": 7.6875,
      "learning_rate": 2.3030692933297972e-05,
      "loss": 1.1983,
      "num_input_tokens_seen": 11641096,
      "step": 21005
    },
    {
      "epoch": 368.60176991150445,
      "grad_norm": 7.4375,
      "learning_rate": 2.3020906114828843e-05,
      "loss": 0.9265,
      "num_input_tokens_seen": 11644056,
      "step": 21010
    },
    {
      "epoch": 368.69026548672565,
      "grad_norm": 9.5,
      "learning_rate": 2.301111960156088e-05,
      "loss": 0.9797,
      "num_input_tokens_seen": 11646904,
      "step": 21015
    },
    {
      "epoch": 368.7787610619469,
      "grad_norm": 9.125,
      "learning_rate": 2.300133339500326e-05,
      "loss": 1.1999,
      "num_input_tokens_seen": 11649592,
      "step": 21020
    },
    {
      "epoch": 368.86725663716817,
      "grad_norm": 12.3125,
      "learning_rate": 2.2991547496665148e-05,
      "loss": 1.0443,
      "num_input_tokens_seen": 11652424,
      "step": 21025
    },
    {
      "epoch": 368.95575221238937,
      "grad_norm": 8.75,
      "learning_rate": 2.298176190805565e-05,
      "loss": 1.2767,
      "num_input_tokens_seen": 11655016,
      "step": 21030
    },
    {
      "epoch": 369.0353982300885,
      "grad_norm": 8.3125,
      "learning_rate": 2.2971976630683826e-05,
      "loss": 0.9884,
      "num_input_tokens_seen": 11657208,
      "step": 21035
    },
    {
      "epoch": 369.12389380530976,
      "grad_norm": 11.1875,
      "learning_rate": 2.29621916660587e-05,
      "loss": 0.9886,
      "num_input_tokens_seen": 11660824,
      "step": 21040
    },
    {
      "epoch": 369.21238938053096,
      "grad_norm": 7.28125,
      "learning_rate": 2.295240701568922e-05,
      "loss": 1.0395,
      "num_input_tokens_seen": 11663864,
      "step": 21045
    },
    {
      "epoch": 369.3008849557522,
      "grad_norm": 9.125,
      "learning_rate": 2.2942622681084312e-05,
      "loss": 1.1533,
      "num_input_tokens_seen": 11666280,
      "step": 21050
    },
    {
      "epoch": 369.3893805309734,
      "grad_norm": 7.8125,
      "learning_rate": 2.293283866375284e-05,
      "loss": 1.1259,
      "num_input_tokens_seen": 11669112,
      "step": 21055
    },
    {
      "epoch": 369.4778761061947,
      "grad_norm": 11.3125,
      "learning_rate": 2.2923054965203627e-05,
      "loss": 0.9363,
      "num_input_tokens_seen": 11671944,
      "step": 21060
    },
    {
      "epoch": 369.56637168141594,
      "grad_norm": 7.0,
      "learning_rate": 2.2913271586945443e-05,
      "loss": 1.0348,
      "num_input_tokens_seen": 11674904,
      "step": 21065
    },
    {
      "epoch": 369.65486725663715,
      "grad_norm": 7.28125,
      "learning_rate": 2.290348853048699e-05,
      "loss": 0.9371,
      "num_input_tokens_seen": 11677880,
      "step": 21070
    },
    {
      "epoch": 369.7433628318584,
      "grad_norm": 7.375,
      "learning_rate": 2.2893705797336956e-05,
      "loss": 0.9826,
      "num_input_tokens_seen": 11680520,
      "step": 21075
    },
    {
      "epoch": 369.83185840707966,
      "grad_norm": 6.375,
      "learning_rate": 2.288392338900397e-05,
      "loss": 0.9492,
      "num_input_tokens_seen": 11683080,
      "step": 21080
    },
    {
      "epoch": 369.92035398230087,
      "grad_norm": 9.9375,
      "learning_rate": 2.2874141306996576e-05,
      "loss": 1.3116,
      "num_input_tokens_seen": 11685784,
      "step": 21085
    },
    {
      "epoch": 370.0,
      "grad_norm": 37.5,
      "learning_rate": 2.2864359552823312e-05,
      "loss": 1.0212,
      "num_input_tokens_seen": 11687856,
      "step": 21090
    },
    {
      "epoch": 370.08849557522126,
      "grad_norm": 7.34375,
      "learning_rate": 2.2854578127992648e-05,
      "loss": 1.2089,
      "num_input_tokens_seen": 11690240,
      "step": 21095
    },
    {
      "epoch": 370.17699115044246,
      "grad_norm": 9.875,
      "learning_rate": 2.2844797034012988e-05,
      "loss": 1.0121,
      "num_input_tokens_seen": 11693344,
      "step": 21100
    },
    {
      "epoch": 370.2654867256637,
      "grad_norm": 8.25,
      "learning_rate": 2.2835016272392722e-05,
      "loss": 1.1039,
      "num_input_tokens_seen": 11695984,
      "step": 21105
    },
    {
      "epoch": 370.353982300885,
      "grad_norm": 7.75,
      "learning_rate": 2.2825235844640142e-05,
      "loss": 1.0097,
      "num_input_tokens_seen": 11699296,
      "step": 21110
    },
    {
      "epoch": 370.4424778761062,
      "grad_norm": 6.4375,
      "learning_rate": 2.2815455752263522e-05,
      "loss": 0.9052,
      "num_input_tokens_seen": 11701952,
      "step": 21115
    },
    {
      "epoch": 370.53097345132744,
      "grad_norm": 9.375,
      "learning_rate": 2.2805675996771092e-05,
      "loss": 1.1696,
      "num_input_tokens_seen": 11704560,
      "step": 21120
    },
    {
      "epoch": 370.6194690265487,
      "grad_norm": 8.9375,
      "learning_rate": 2.2795896579670987e-05,
      "loss": 0.9709,
      "num_input_tokens_seen": 11707088,
      "step": 21125
    },
    {
      "epoch": 370.7079646017699,
      "grad_norm": 12.25,
      "learning_rate": 2.2786117502471337e-05,
      "loss": 1.2623,
      "num_input_tokens_seen": 11709824,
      "step": 21130
    },
    {
      "epoch": 370.79646017699116,
      "grad_norm": 10.5,
      "learning_rate": 2.2776338766680185e-05,
      "loss": 0.9453,
      "num_input_tokens_seen": 11713024,
      "step": 21135
    },
    {
      "epoch": 370.88495575221236,
      "grad_norm": 9.625,
      "learning_rate": 2.2766560373805533e-05,
      "loss": 0.9852,
      "num_input_tokens_seen": 11715760,
      "step": 21140
    },
    {
      "epoch": 370.9734513274336,
      "grad_norm": 9.0,
      "learning_rate": 2.2756782325355353e-05,
      "loss": 0.9906,
      "num_input_tokens_seen": 11718480,
      "step": 21145
    },
    {
      "epoch": 371.05309734513276,
      "grad_norm": 9.1875,
      "learning_rate": 2.2747004622837514e-05,
      "loss": 0.9441,
      "num_input_tokens_seen": 11720736,
      "step": 21150
    },
    {
      "epoch": 371.14159292035396,
      "grad_norm": 10.125,
      "learning_rate": 2.2737227267759878e-05,
      "loss": 1.0688,
      "num_input_tokens_seen": 11724256,
      "step": 21155
    },
    {
      "epoch": 371.2300884955752,
      "grad_norm": 6.59375,
      "learning_rate": 2.272745026163024e-05,
      "loss": 0.8957,
      "num_input_tokens_seen": 11726880,
      "step": 21160
    },
    {
      "epoch": 371.3185840707965,
      "grad_norm": 10.25,
      "learning_rate": 2.271767360595633e-05,
      "loss": 1.2686,
      "num_input_tokens_seen": 11729104,
      "step": 21165
    },
    {
      "epoch": 371.4070796460177,
      "grad_norm": 10.8125,
      "learning_rate": 2.270789730224583e-05,
      "loss": 1.0406,
      "num_input_tokens_seen": 11731632,
      "step": 21170
    },
    {
      "epoch": 371.49557522123894,
      "grad_norm": 7.75,
      "learning_rate": 2.2698121352006367e-05,
      "loss": 0.8926,
      "num_input_tokens_seen": 11734416,
      "step": 21175
    },
    {
      "epoch": 371.5840707964602,
      "grad_norm": 8.3125,
      "learning_rate": 2.2688345756745517e-05,
      "loss": 1.0421,
      "num_input_tokens_seen": 11737072,
      "step": 21180
    },
    {
      "epoch": 371.6725663716814,
      "grad_norm": 9.1875,
      "learning_rate": 2.267857051797081e-05,
      "loss": 1.0933,
      "num_input_tokens_seen": 11739744,
      "step": 21185
    },
    {
      "epoch": 371.76106194690266,
      "grad_norm": 7.875,
      "learning_rate": 2.2668795637189695e-05,
      "loss": 1.187,
      "num_input_tokens_seen": 11742560,
      "step": 21190
    },
    {
      "epoch": 371.8495575221239,
      "grad_norm": 9.75,
      "learning_rate": 2.2659021115909586e-05,
      "loss": 1.0586,
      "num_input_tokens_seen": 11745504,
      "step": 21195
    },
    {
      "epoch": 371.9380530973451,
      "grad_norm": 8.6875,
      "learning_rate": 2.2649246955637847e-05,
      "loss": 1.1234,
      "num_input_tokens_seen": 11748768,
      "step": 21200
    },
    {
      "epoch": 371.9380530973451,
      "eval_loss": 1.0325586795806885,
      "eval_runtime": 0.9345,
      "eval_samples_per_second": 26.752,
      "eval_steps_per_second": 13.911,
      "num_input_tokens_seen": 11748768,
      "step": 21200
    },
    {
      "epoch": 372.01769911504425,
      "grad_norm": 8.9375,
      "learning_rate": 2.2639473157881766e-05,
      "loss": 1.0475,
      "num_input_tokens_seen": 11751336,
      "step": 21205
    },
    {
      "epoch": 372.1061946902655,
      "grad_norm": 10.25,
      "learning_rate": 2.2629699724148594e-05,
      "loss": 0.9885,
      "num_input_tokens_seen": 11753960,
      "step": 21210
    },
    {
      "epoch": 372.1946902654867,
      "grad_norm": 9.8125,
      "learning_rate": 2.26199266559455e-05,
      "loss": 1.3335,
      "num_input_tokens_seen": 11756568,
      "step": 21215
    },
    {
      "epoch": 372.283185840708,
      "grad_norm": 9.6875,
      "learning_rate": 2.2610153954779625e-05,
      "loss": 1.025,
      "num_input_tokens_seen": 11759608,
      "step": 21220
    },
    {
      "epoch": 372.37168141592923,
      "grad_norm": 10.25,
      "learning_rate": 2.2600381622158056e-05,
      "loss": 1.051,
      "num_input_tokens_seen": 11762472,
      "step": 21225
    },
    {
      "epoch": 372.46017699115043,
      "grad_norm": 8.1875,
      "learning_rate": 2.2590609659587783e-05,
      "loss": 1.0936,
      "num_input_tokens_seen": 11765160,
      "step": 21230
    },
    {
      "epoch": 372.5486725663717,
      "grad_norm": 8.5,
      "learning_rate": 2.2580838068575787e-05,
      "loss": 1.0994,
      "num_input_tokens_seen": 11767640,
      "step": 21235
    },
    {
      "epoch": 372.6371681415929,
      "grad_norm": 7.1875,
      "learning_rate": 2.257106685062896e-05,
      "loss": 1.0482,
      "num_input_tokens_seen": 11770664,
      "step": 21240
    },
    {
      "epoch": 372.72566371681415,
      "grad_norm": 7.21875,
      "learning_rate": 2.256129600725415e-05,
      "loss": 0.935,
      "num_input_tokens_seen": 11773608,
      "step": 21245
    },
    {
      "epoch": 372.8141592920354,
      "grad_norm": 10.9375,
      "learning_rate": 2.2551525539958145e-05,
      "loss": 1.1132,
      "num_input_tokens_seen": 11776728,
      "step": 21250
    },
    {
      "epoch": 372.9026548672566,
      "grad_norm": 7.21875,
      "learning_rate": 2.2541755450247663e-05,
      "loss": 0.9404,
      "num_input_tokens_seen": 11779192,
      "step": 21255
    },
    {
      "epoch": 372.9911504424779,
      "grad_norm": 9.5625,
      "learning_rate": 2.2531985739629382e-05,
      "loss": 1.0374,
      "num_input_tokens_seen": 11781896,
      "step": 21260
    },
    {
      "epoch": 373.070796460177,
      "grad_norm": 9.4375,
      "learning_rate": 2.2522216409609924e-05,
      "loss": 1.1646,
      "num_input_tokens_seen": 11784120,
      "step": 21265
    },
    {
      "epoch": 373.1592920353982,
      "grad_norm": 12.625,
      "learning_rate": 2.2512447461695826e-05,
      "loss": 1.1279,
      "num_input_tokens_seen": 11786792,
      "step": 21270
    },
    {
      "epoch": 373.24778761061947,
      "grad_norm": 6.75,
      "learning_rate": 2.2502678897393593e-05,
      "loss": 0.9044,
      "num_input_tokens_seen": 11789448,
      "step": 21275
    },
    {
      "epoch": 373.3362831858407,
      "grad_norm": 9.0625,
      "learning_rate": 2.2492910718209665e-05,
      "loss": 0.9879,
      "num_input_tokens_seen": 11792376,
      "step": 21280
    },
    {
      "epoch": 373.42477876106193,
      "grad_norm": 8.875,
      "learning_rate": 2.2483142925650398e-05,
      "loss": 1.1414,
      "num_input_tokens_seen": 11794792,
      "step": 21285
    },
    {
      "epoch": 373.5132743362832,
      "grad_norm": 9.1875,
      "learning_rate": 2.247337552122213e-05,
      "loss": 1.2015,
      "num_input_tokens_seen": 11797480,
      "step": 21290
    },
    {
      "epoch": 373.60176991150445,
      "grad_norm": 9.625,
      "learning_rate": 2.24636085064311e-05,
      "loss": 1.2969,
      "num_input_tokens_seen": 11800424,
      "step": 21295
    },
    {
      "epoch": 373.69026548672565,
      "grad_norm": 6.59375,
      "learning_rate": 2.245384188278351e-05,
      "loss": 0.9348,
      "num_input_tokens_seen": 11803144,
      "step": 21300
    },
    {
      "epoch": 373.7787610619469,
      "grad_norm": 7.375,
      "learning_rate": 2.2444075651785513e-05,
      "loss": 1.0605,
      "num_input_tokens_seen": 11806344,
      "step": 21305
    },
    {
      "epoch": 373.86725663716817,
      "grad_norm": 8.375,
      "learning_rate": 2.243430981494316e-05,
      "loss": 0.8566,
      "num_input_tokens_seen": 11809064,
      "step": 21310
    },
    {
      "epoch": 373.95575221238937,
      "grad_norm": 6.71875,
      "learning_rate": 2.2424544373762475e-05,
      "loss": 1.0202,
      "num_input_tokens_seen": 11811880,
      "step": 21315
    },
    {
      "epoch": 374.0353982300885,
      "grad_norm": 8.1875,
      "learning_rate": 2.2414779329749418e-05,
      "loss": 1.2201,
      "num_input_tokens_seen": 11813936,
      "step": 21320
    },
    {
      "epoch": 374.12389380530976,
      "grad_norm": 8.6875,
      "learning_rate": 2.2405014684409873e-05,
      "loss": 0.8931,
      "num_input_tokens_seen": 11817296,
      "step": 21325
    },
    {
      "epoch": 374.21238938053096,
      "grad_norm": 8.9375,
      "learning_rate": 2.239525043924968e-05,
      "loss": 1.1057,
      "num_input_tokens_seen": 11820496,
      "step": 21330
    },
    {
      "epoch": 374.3008849557522,
      "grad_norm": 8.625,
      "learning_rate": 2.2385486595774592e-05,
      "loss": 1.1785,
      "num_input_tokens_seen": 11823152,
      "step": 21335
    },
    {
      "epoch": 374.3893805309734,
      "grad_norm": 10.4375,
      "learning_rate": 2.237572315549033e-05,
      "loss": 1.0573,
      "num_input_tokens_seen": 11825552,
      "step": 21340
    },
    {
      "epoch": 374.4778761061947,
      "grad_norm": 9.875,
      "learning_rate": 2.2365960119902545e-05,
      "loss": 1.2173,
      "num_input_tokens_seen": 11827952,
      "step": 21345
    },
    {
      "epoch": 374.56637168141594,
      "grad_norm": 8.75,
      "learning_rate": 2.2356197490516806e-05,
      "loss": 0.9677,
      "num_input_tokens_seen": 11830976,
      "step": 21350
    },
    {
      "epoch": 374.65486725663715,
      "grad_norm": 6.90625,
      "learning_rate": 2.234643526883863e-05,
      "loss": 0.9597,
      "num_input_tokens_seen": 11833680,
      "step": 21355
    },
    {
      "epoch": 374.7433628318584,
      "grad_norm": 9.5625,
      "learning_rate": 2.2336673456373497e-05,
      "loss": 1.1329,
      "num_input_tokens_seen": 11836432,
      "step": 21360
    },
    {
      "epoch": 374.83185840707966,
      "grad_norm": 7.5,
      "learning_rate": 2.2326912054626772e-05,
      "loss": 1.078,
      "num_input_tokens_seen": 11839072,
      "step": 21365
    },
    {
      "epoch": 374.92035398230087,
      "grad_norm": 8.9375,
      "learning_rate": 2.2317151065103813e-05,
      "loss": 1.0491,
      "num_input_tokens_seen": 11842000,
      "step": 21370
    },
    {
      "epoch": 375.0,
      "grad_norm": 11.625,
      "learning_rate": 2.2307390489309865e-05,
      "loss": 0.8958,
      "num_input_tokens_seen": 11844288,
      "step": 21375
    },
    {
      "epoch": 375.08849557522126,
      "grad_norm": 13.5,
      "learning_rate": 2.2297630328750146e-05,
      "loss": 0.8216,
      "num_input_tokens_seen": 11847472,
      "step": 21380
    },
    {
      "epoch": 375.17699115044246,
      "grad_norm": 8.4375,
      "learning_rate": 2.228787058492979e-05,
      "loss": 1.0352,
      "num_input_tokens_seen": 11850192,
      "step": 21385
    },
    {
      "epoch": 375.2654867256637,
      "grad_norm": 7.09375,
      "learning_rate": 2.2278111259353875e-05,
      "loss": 0.9831,
      "num_input_tokens_seen": 11853088,
      "step": 21390
    },
    {
      "epoch": 375.353982300885,
      "grad_norm": 7.0,
      "learning_rate": 2.2268352353527395e-05,
      "loss": 1.089,
      "num_input_tokens_seen": 11855536,
      "step": 21395
    },
    {
      "epoch": 375.4424778761062,
      "grad_norm": 7.625,
      "learning_rate": 2.225859386895533e-05,
      "loss": 1.1642,
      "num_input_tokens_seen": 11857872,
      "step": 21400
    },
    {
      "epoch": 375.4424778761062,
      "eval_loss": 1.0289069414138794,
      "eval_runtime": 0.9373,
      "eval_samples_per_second": 26.674,
      "eval_steps_per_second": 13.87,
      "num_input_tokens_seen": 11857872,
      "step": 21400
    },
    {
      "epoch": 375.53097345132744,
      "grad_norm": 8.3125,
      "learning_rate": 2.2248835807142525e-05,
      "loss": 1.2242,
      "num_input_tokens_seen": 11860256,
      "step": 21405
    },
    {
      "epoch": 375.6194690265487,
      "grad_norm": 7.78125,
      "learning_rate": 2.2239078169593826e-05,
      "loss": 0.9753,
      "num_input_tokens_seen": 11863536,
      "step": 21410
    },
    {
      "epoch": 375.7079646017699,
      "grad_norm": 8.3125,
      "learning_rate": 2.222932095781396e-05,
      "loss": 1.0518,
      "num_input_tokens_seen": 11866720,
      "step": 21415
    },
    {
      "epoch": 375.79646017699116,
      "grad_norm": 8.4375,
      "learning_rate": 2.221956417330762e-05,
      "loss": 1.1145,
      "num_input_tokens_seen": 11869552,
      "step": 21420
    },
    {
      "epoch": 375.88495575221236,
      "grad_norm": 7.1875,
      "learning_rate": 2.2209807817579438e-05,
      "loss": 1.1221,
      "num_input_tokens_seen": 11871984,
      "step": 21425
    },
    {
      "epoch": 375.9734513274336,
      "grad_norm": 7.96875,
      "learning_rate": 2.220005189213394e-05,
      "loss": 1.0334,
      "num_input_tokens_seen": 11875024,
      "step": 21430
    },
    {
      "epoch": 376.05309734513276,
      "grad_norm": 9.8125,
      "learning_rate": 2.2190296398475624e-05,
      "loss": 1.0031,
      "num_input_tokens_seen": 11877280,
      "step": 21435
    },
    {
      "epoch": 376.14159292035396,
      "grad_norm": 7.625,
      "learning_rate": 2.2180541338108926e-05,
      "loss": 1.1572,
      "num_input_tokens_seen": 11879712,
      "step": 21440
    },
    {
      "epoch": 376.2300884955752,
      "grad_norm": 8.5,
      "learning_rate": 2.2170786712538176e-05,
      "loss": 0.8854,
      "num_input_tokens_seen": 11882544,
      "step": 21445
    },
    {
      "epoch": 376.3185840707965,
      "grad_norm": 8.625,
      "learning_rate": 2.216103252326768e-05,
      "loss": 1.0689,
      "num_input_tokens_seen": 11884976,
      "step": 21450
    },
    {
      "epoch": 376.4070796460177,
      "grad_norm": 6.8125,
      "learning_rate": 2.2151278771801635e-05,
      "loss": 1.0519,
      "num_input_tokens_seen": 11887536,
      "step": 21455
    },
    {
      "epoch": 376.49557522123894,
      "grad_norm": 8.5,
      "learning_rate": 2.21415254596442e-05,
      "loss": 0.9148,
      "num_input_tokens_seen": 11890880,
      "step": 21460
    },
    {
      "epoch": 376.5840707964602,
      "grad_norm": 7.875,
      "learning_rate": 2.213177258829947e-05,
      "loss": 1.0097,
      "num_input_tokens_seen": 11893792,
      "step": 21465
    },
    {
      "epoch": 376.6725663716814,
      "grad_norm": 12.9375,
      "learning_rate": 2.2122020159271445e-05,
      "loss": 1.215,
      "num_input_tokens_seen": 11897040,
      "step": 21470
    },
    {
      "epoch": 376.76106194690266,
      "grad_norm": 9.4375,
      "learning_rate": 2.2112268174064075e-05,
      "loss": 1.1031,
      "num_input_tokens_seen": 11899792,
      "step": 21475
    },
    {
      "epoch": 376.8495575221239,
      "grad_norm": 9.125,
      "learning_rate": 2.2102516634181253e-05,
      "loss": 1.1326,
      "num_input_tokens_seen": 11902400,
      "step": 21480
    },
    {
      "epoch": 376.9380530973451,
      "grad_norm": 7.25,
      "learning_rate": 2.209276554112677e-05,
      "loss": 0.8463,
      "num_input_tokens_seen": 11905584,
      "step": 21485
    },
    {
      "epoch": 377.01769911504425,
      "grad_norm": 7.5625,
      "learning_rate": 2.2083014896404384e-05,
      "loss": 1.0574,
      "num_input_tokens_seen": 11907952,
      "step": 21490
    },
    {
      "epoch": 377.1061946902655,
      "grad_norm": 6.90625,
      "learning_rate": 2.207326470151775e-05,
      "loss": 1.1458,
      "num_input_tokens_seen": 11910976,
      "step": 21495
    },
    {
      "epoch": 377.1946902654867,
      "grad_norm": 11.0,
      "learning_rate": 2.2063514957970477e-05,
      "loss": 1.033,
      "num_input_tokens_seen": 11914032,
      "step": 21500
    },
    {
      "epoch": 377.283185840708,
      "grad_norm": 9.4375,
      "learning_rate": 2.205376566726611e-05,
      "loss": 1.0084,
      "num_input_tokens_seen": 11917168,
      "step": 21505
    },
    {
      "epoch": 377.37168141592923,
      "grad_norm": 8.125,
      "learning_rate": 2.204401683090809e-05,
      "loss": 0.9919,
      "num_input_tokens_seen": 11919824,
      "step": 21510
    },
    {
      "epoch": 377.46017699115043,
      "grad_norm": 7.75,
      "learning_rate": 2.203426845039982e-05,
      "loss": 1.0051,
      "num_input_tokens_seen": 11922688,
      "step": 21515
    },
    {
      "epoch": 377.5486725663717,
      "grad_norm": 7.0625,
      "learning_rate": 2.202452052724464e-05,
      "loss": 1.0443,
      "num_input_tokens_seen": 11925808,
      "step": 21520
    },
    {
      "epoch": 377.6371681415929,
      "grad_norm": 10.75,
      "learning_rate": 2.2014773062945777e-05,
      "loss": 1.1165,
      "num_input_tokens_seen": 11928096,
      "step": 21525
    },
    {
      "epoch": 377.72566371681415,
      "grad_norm": 10.1875,
      "learning_rate": 2.2005026059006427e-05,
      "loss": 1.1457,
      "num_input_tokens_seen": 11930768,
      "step": 21530
    },
    {
      "epoch": 377.8141592920354,
      "grad_norm": 12.0625,
      "learning_rate": 2.1995279516929695e-05,
      "loss": 1.104,
      "num_input_tokens_seen": 11933600,
      "step": 21535
    },
    {
      "epoch": 377.9026548672566,
      "grad_norm": 6.09375,
      "learning_rate": 2.1985533438218613e-05,
      "loss": 0.9877,
      "num_input_tokens_seen": 11936560,
      "step": 21540
    },
    {
      "epoch": 377.9911504424779,
      "grad_norm": 10.5625,
      "learning_rate": 2.197578782437617e-05,
      "loss": 1.058,
      "num_input_tokens_seen": 11939104,
      "step": 21545
    },
    {
      "epoch": 378.070796460177,
      "grad_norm": 7.125,
      "learning_rate": 2.196604267690524e-05,
      "loss": 1.2309,
      "num_input_tokens_seen": 11941264,
      "step": 21550
    },
    {
      "epoch": 378.1592920353982,
      "grad_norm": 9.0,
      "learning_rate": 2.195629799730865e-05,
      "loss": 0.9463,
      "num_input_tokens_seen": 11943648,
      "step": 21555
    },
    {
      "epoch": 378.24778761061947,
      "grad_norm": 10.0,
      "learning_rate": 2.1946553787089173e-05,
      "loss": 1.1121,
      "num_input_tokens_seen": 11947008,
      "step": 21560
    },
    {
      "epoch": 378.3362831858407,
      "grad_norm": 8.4375,
      "learning_rate": 2.193681004774947e-05,
      "loss": 0.8864,
      "num_input_tokens_seen": 11949728,
      "step": 21565
    },
    {
      "epoch": 378.42477876106193,
      "grad_norm": 10.875,
      "learning_rate": 2.1927066780792154e-05,
      "loss": 1.1192,
      "num_input_tokens_seen": 11952368,
      "step": 21570
    },
    {
      "epoch": 378.5132743362832,
      "grad_norm": 7.75,
      "learning_rate": 2.191732398771975e-05,
      "loss": 1.0332,
      "num_input_tokens_seen": 11955184,
      "step": 21575
    },
    {
      "epoch": 378.60176991150445,
      "grad_norm": 12.8125,
      "learning_rate": 2.1907581670034725e-05,
      "loss": 0.9535,
      "num_input_tokens_seen": 11958032,
      "step": 21580
    },
    {
      "epoch": 378.69026548672565,
      "grad_norm": 11.4375,
      "learning_rate": 2.189783982923948e-05,
      "loss": 1.0761,
      "num_input_tokens_seen": 11961040,
      "step": 21585
    },
    {
      "epoch": 378.7787610619469,
      "grad_norm": 10.1875,
      "learning_rate": 2.1888098466836303e-05,
      "loss": 1.2943,
      "num_input_tokens_seen": 11963728,
      "step": 21590
    },
    {
      "epoch": 378.86725663716817,
      "grad_norm": 7.78125,
      "learning_rate": 2.1878357584327457e-05,
      "loss": 1.0237,
      "num_input_tokens_seen": 11966800,
      "step": 21595
    },
    {
      "epoch": 378.95575221238937,
      "grad_norm": 7.625,
      "learning_rate": 2.1868617183215103e-05,
      "loss": 1.027,
      "num_input_tokens_seen": 11969696,
      "step": 21600
    },
    {
      "epoch": 378.95575221238937,
      "eval_loss": 1.0271248817443848,
      "eval_runtime": 0.9361,
      "eval_samples_per_second": 26.706,
      "eval_steps_per_second": 13.887,
      "num_input_tokens_seen": 11969696,
      "step": 21600
    },
    {
      "epoch": 379.0353982300885,
      "grad_norm": 8.9375,
      "learning_rate": 2.1858877265001327e-05,
      "loss": 1.0102,
      "num_input_tokens_seen": 11972272,
      "step": 21605
    },
    {
      "epoch": 379.12389380530976,
      "grad_norm": 10.125,
      "learning_rate": 2.184913783118816e-05,
      "loss": 1.1229,
      "num_input_tokens_seen": 11975440,
      "step": 21610
    },
    {
      "epoch": 379.21238938053096,
      "grad_norm": 7.875,
      "learning_rate": 2.1839398883277522e-05,
      "loss": 1.0934,
      "num_input_tokens_seen": 11978304,
      "step": 21615
    },
    {
      "epoch": 379.3008849557522,
      "grad_norm": 8.375,
      "learning_rate": 2.182966042277129e-05,
      "loss": 0.8639,
      "num_input_tokens_seen": 11981152,
      "step": 21620
    },
    {
      "epoch": 379.3893805309734,
      "grad_norm": 8.5625,
      "learning_rate": 2.181992245117128e-05,
      "loss": 1.1954,
      "num_input_tokens_seen": 11983824,
      "step": 21625
    },
    {
      "epoch": 379.4778761061947,
      "grad_norm": 11.5,
      "learning_rate": 2.181018496997918e-05,
      "loss": 1.2627,
      "num_input_tokens_seen": 11986512,
      "step": 21630
    },
    {
      "epoch": 379.56637168141594,
      "grad_norm": 6.84375,
      "learning_rate": 2.1800447980696648e-05,
      "loss": 1.1377,
      "num_input_tokens_seen": 11989040,
      "step": 21635
    },
    {
      "epoch": 379.65486725663715,
      "grad_norm": 7.15625,
      "learning_rate": 2.1790711484825248e-05,
      "loss": 1.144,
      "num_input_tokens_seen": 11991584,
      "step": 21640
    },
    {
      "epoch": 379.7433628318584,
      "grad_norm": 7.0,
      "learning_rate": 2.178097548386646e-05,
      "loss": 0.9709,
      "num_input_tokens_seen": 11994560,
      "step": 21645
    },
    {
      "epoch": 379.83185840707966,
      "grad_norm": 8.125,
      "learning_rate": 2.1771239979321712e-05,
      "loss": 0.994,
      "num_input_tokens_seen": 11996992,
      "step": 21650
    },
    {
      "epoch": 379.92035398230087,
      "grad_norm": 7.375,
      "learning_rate": 2.1761504972692327e-05,
      "loss": 0.9208,
      "num_input_tokens_seen": 11999632,
      "step": 21655
    },
    {
      "epoch": 380.0,
      "grad_norm": 9.0,
      "learning_rate": 2.1751770465479572e-05,
      "loss": 0.8893,
      "num_input_tokens_seen": 12002368,
      "step": 21660
    },
    {
      "epoch": 380.08849557522126,
      "grad_norm": 7.8125,
      "learning_rate": 2.174203645918464e-05,
      "loss": 0.9371,
      "num_input_tokens_seen": 12005728,
      "step": 21665
    },
    {
      "epoch": 380.17699115044246,
      "grad_norm": 10.125,
      "learning_rate": 2.1732302955308624e-05,
      "loss": 1.2502,
      "num_input_tokens_seen": 12008528,
      "step": 21670
    },
    {
      "epoch": 380.2654867256637,
      "grad_norm": 8.625,
      "learning_rate": 2.172256995535255e-05,
      "loss": 1.012,
      "num_input_tokens_seen": 12011408,
      "step": 21675
    },
    {
      "epoch": 380.353982300885,
      "grad_norm": 7.59375,
      "learning_rate": 2.171283746081739e-05,
      "loss": 1.1305,
      "num_input_tokens_seen": 12014112,
      "step": 21680
    },
    {
      "epoch": 380.4424778761062,
      "grad_norm": 9.25,
      "learning_rate": 2.1703105473203988e-05,
      "loss": 1.2972,
      "num_input_tokens_seen": 12016784,
      "step": 21685
    },
    {
      "epoch": 380.53097345132744,
      "grad_norm": 7.53125,
      "learning_rate": 2.1693373994013168e-05,
      "loss": 1.0676,
      "num_input_tokens_seen": 12019440,
      "step": 21690
    },
    {
      "epoch": 380.6194690265487,
      "grad_norm": 9.3125,
      "learning_rate": 2.168364302474562e-05,
      "loss": 1.0422,
      "num_input_tokens_seen": 12022096,
      "step": 21695
    },
    {
      "epoch": 380.7079646017699,
      "grad_norm": 8.4375,
      "learning_rate": 2.167391256690199e-05,
      "loss": 0.9903,
      "num_input_tokens_seen": 12024832,
      "step": 21700
    },
    {
      "epoch": 380.79646017699116,
      "grad_norm": 6.875,
      "learning_rate": 2.1664182621982855e-05,
      "loss": 1.0303,
      "num_input_tokens_seen": 12027728,
      "step": 21705
    },
    {
      "epoch": 380.88495575221236,
      "grad_norm": 6.0625,
      "learning_rate": 2.1654453191488673e-05,
      "loss": 0.8496,
      "num_input_tokens_seen": 12030704,
      "step": 21710
    },
    {
      "epoch": 380.9734513274336,
      "grad_norm": 7.5625,
      "learning_rate": 2.1644724276919846e-05,
      "loss": 0.9659,
      "num_input_tokens_seen": 12033520,
      "step": 21715
    },
    {
      "epoch": 381.05309734513276,
      "grad_norm": 7.78125,
      "learning_rate": 2.1634995879776715e-05,
      "loss": 1.0148,
      "num_input_tokens_seen": 12035592,
      "step": 21720
    },
    {
      "epoch": 381.14159292035396,
      "grad_norm": 7.84375,
      "learning_rate": 2.162526800155949e-05,
      "loss": 0.936,
      "num_input_tokens_seen": 12038536,
      "step": 21725
    },
    {
      "epoch": 381.2300884955752,
      "grad_norm": 10.0625,
      "learning_rate": 2.1615540643768363e-05,
      "loss": 1.0878,
      "num_input_tokens_seen": 12041448,
      "step": 21730
    },
    {
      "epoch": 381.3185840707965,
      "grad_norm": 8.0,
      "learning_rate": 2.160581380790339e-05,
      "loss": 0.987,
      "num_input_tokens_seen": 12044520,
      "step": 21735
    },
    {
      "epoch": 381.4070796460177,
      "grad_norm": 8.75,
      "learning_rate": 2.1596087495464586e-05,
      "loss": 0.9155,
      "num_input_tokens_seen": 12047432,
      "step": 21740
    },
    {
      "epoch": 381.49557522123894,
      "grad_norm": 9.125,
      "learning_rate": 2.1586361707951866e-05,
      "loss": 1.1287,
      "num_input_tokens_seen": 12050616,
      "step": 21745
    },
    {
      "epoch": 381.5840707964602,
      "grad_norm": 8.75,
      "learning_rate": 2.157663644686507e-05,
      "loss": 1.0178,
      "num_input_tokens_seen": 12053048,
      "step": 21750
    },
    {
      "epoch": 381.6725663716814,
      "grad_norm": 6.6875,
      "learning_rate": 2.156691171370396e-05,
      "loss": 1.0244,
      "num_input_tokens_seen": 12055896,
      "step": 21755
    },
    {
      "epoch": 381.76106194690266,
      "grad_norm": 11.5625,
      "learning_rate": 2.1557187509968195e-05,
      "loss": 1.1918,
      "num_input_tokens_seen": 12058776,
      "step": 21760
    },
    {
      "epoch": 381.8495575221239,
      "grad_norm": 7.34375,
      "learning_rate": 2.1547463837157382e-05,
      "loss": 0.8836,
      "num_input_tokens_seen": 12061448,
      "step": 21765
    },
    {
      "epoch": 381.9380530973451,
      "grad_norm": 8.8125,
      "learning_rate": 2.1537740696771045e-05,
      "loss": 1.1912,
      "num_input_tokens_seen": 12063976,
      "step": 21770
    },
    {
      "epoch": 382.01769911504425,
      "grad_norm": 8.1875,
      "learning_rate": 2.1528018090308587e-05,
      "loss": 1.2411,
      "num_input_tokens_seen": 12066432,
      "step": 21775
    },
    {
      "epoch": 382.1061946902655,
      "grad_norm": 7.9375,
      "learning_rate": 2.151829601926938e-05,
      "loss": 1.1246,
      "num_input_tokens_seen": 12069040,
      "step": 21780
    },
    {
      "epoch": 382.1946902654867,
      "grad_norm": 6.84375,
      "learning_rate": 2.1508574485152684e-05,
      "loss": 0.8891,
      "num_input_tokens_seen": 12072080,
      "step": 21785
    },
    {
      "epoch": 382.283185840708,
      "grad_norm": 8.5625,
      "learning_rate": 2.1498853489457667e-05,
      "loss": 1.2206,
      "num_input_tokens_seen": 12074864,
      "step": 21790
    },
    {
      "epoch": 382.37168141592923,
      "grad_norm": 6.3125,
      "learning_rate": 2.1489133033683455e-05,
      "loss": 0.9761,
      "num_input_tokens_seen": 12077712,
      "step": 21795
    },
    {
      "epoch": 382.46017699115043,
      "grad_norm": 5.9375,
      "learning_rate": 2.1479413119329038e-05,
      "loss": 1.0781,
      "num_input_tokens_seen": 12080592,
      "step": 21800
    },
    {
      "epoch": 382.46017699115043,
      "eval_loss": 1.021244764328003,
      "eval_runtime": 0.9402,
      "eval_samples_per_second": 26.591,
      "eval_steps_per_second": 13.827,
      "num_input_tokens_seen": 12080592,
      "step": 21800
    },
    {
      "epoch": 382.5486725663717,
      "grad_norm": 14.125,
      "learning_rate": 2.1469693747893355e-05,
      "loss": 1.1575,
      "num_input_tokens_seen": 12083328,
      "step": 21805
    },
    {
      "epoch": 382.6371681415929,
      "grad_norm": 9.4375,
      "learning_rate": 2.1459974920875274e-05,
      "loss": 1.1115,
      "num_input_tokens_seen": 12086080,
      "step": 21810
    },
    {
      "epoch": 382.72566371681415,
      "grad_norm": 9.875,
      "learning_rate": 2.145025663977354e-05,
      "loss": 1.106,
      "num_input_tokens_seen": 12088768,
      "step": 21815
    },
    {
      "epoch": 382.8141592920354,
      "grad_norm": 9.0,
      "learning_rate": 2.1440538906086844e-05,
      "loss": 0.8967,
      "num_input_tokens_seen": 12091312,
      "step": 21820
    },
    {
      "epoch": 382.9026548672566,
      "grad_norm": 7.5,
      "learning_rate": 2.1430821721313782e-05,
      "loss": 1.1037,
      "num_input_tokens_seen": 12094304,
      "step": 21825
    },
    {
      "epoch": 382.9911504424779,
      "grad_norm": 5.9375,
      "learning_rate": 2.142110508695286e-05,
      "loss": 0.9628,
      "num_input_tokens_seen": 12097120,
      "step": 21830
    },
    {
      "epoch": 383.070796460177,
      "grad_norm": 10.625,
      "learning_rate": 2.1411389004502515e-05,
      "loss": 1.0394,
      "num_input_tokens_seen": 12099392,
      "step": 21835
    },
    {
      "epoch": 383.1592920353982,
      "grad_norm": 9.1875,
      "learning_rate": 2.140167347546107e-05,
      "loss": 0.9982,
      "num_input_tokens_seen": 12102400,
      "step": 21840
    },
    {
      "epoch": 383.24778761061947,
      "grad_norm": 8.75,
      "learning_rate": 2.1391958501326793e-05,
      "loss": 0.9729,
      "num_input_tokens_seen": 12105168,
      "step": 21845
    },
    {
      "epoch": 383.3362831858407,
      "grad_norm": 7.75,
      "learning_rate": 2.1382244083597873e-05,
      "loss": 1.0805,
      "num_input_tokens_seen": 12107632,
      "step": 21850
    },
    {
      "epoch": 383.42477876106193,
      "grad_norm": 7.34375,
      "learning_rate": 2.137253022377237e-05,
      "loss": 0.9263,
      "num_input_tokens_seen": 12110400,
      "step": 21855
    },
    {
      "epoch": 383.5132743362832,
      "grad_norm": 11.375,
      "learning_rate": 2.136281692334829e-05,
      "loss": 1.2433,
      "num_input_tokens_seen": 12113776,
      "step": 21860
    },
    {
      "epoch": 383.60176991150445,
      "grad_norm": 8.6875,
      "learning_rate": 2.135310418382356e-05,
      "loss": 1.0238,
      "num_input_tokens_seen": 12116528,
      "step": 21865
    },
    {
      "epoch": 383.69026548672565,
      "grad_norm": 8.0,
      "learning_rate": 2.134339200669598e-05,
      "loss": 1.241,
      "num_input_tokens_seen": 12119344,
      "step": 21870
    },
    {
      "epoch": 383.7787610619469,
      "grad_norm": 9.5,
      "learning_rate": 2.133368039346331e-05,
      "loss": 1.0591,
      "num_input_tokens_seen": 12122368,
      "step": 21875
    },
    {
      "epoch": 383.86725663716817,
      "grad_norm": 8.0,
      "learning_rate": 2.1323969345623195e-05,
      "loss": 1.0061,
      "num_input_tokens_seen": 12124832,
      "step": 21880
    },
    {
      "epoch": 383.95575221238937,
      "grad_norm": 8.1875,
      "learning_rate": 2.1314258864673207e-05,
      "loss": 0.8933,
      "num_input_tokens_seen": 12127344,
      "step": 21885
    },
    {
      "epoch": 384.0353982300885,
      "grad_norm": 6.90625,
      "learning_rate": 2.130454895211082e-05,
      "loss": 1.0751,
      "num_input_tokens_seen": 12129824,
      "step": 21890
    },
    {
      "epoch": 384.12389380530976,
      "grad_norm": 10.3125,
      "learning_rate": 2.129483960943342e-05,
      "loss": 1.155,
      "num_input_tokens_seen": 12132656,
      "step": 21895
    },
    {
      "epoch": 384.21238938053096,
      "grad_norm": 6.40625,
      "learning_rate": 2.128513083813831e-05,
      "loss": 0.9255,
      "num_input_tokens_seen": 12135664,
      "step": 21900
    },
    {
      "epoch": 384.3008849557522,
      "grad_norm": 8.6875,
      "learning_rate": 2.1275422639722724e-05,
      "loss": 1.0521,
      "num_input_tokens_seen": 12138224,
      "step": 21905
    },
    {
      "epoch": 384.3893805309734,
      "grad_norm": 8.1875,
      "learning_rate": 2.126571501568376e-05,
      "loss": 1.0606,
      "num_input_tokens_seen": 12141504,
      "step": 21910
    },
    {
      "epoch": 384.4778761061947,
      "grad_norm": 9.8125,
      "learning_rate": 2.1256007967518478e-05,
      "loss": 1.0843,
      "num_input_tokens_seen": 12144512,
      "step": 21915
    },
    {
      "epoch": 384.56637168141594,
      "grad_norm": 8.3125,
      "learning_rate": 2.124630149672381e-05,
      "loss": 1.1246,
      "num_input_tokens_seen": 12146944,
      "step": 21920
    },
    {
      "epoch": 384.65486725663715,
      "grad_norm": 10.1875,
      "learning_rate": 2.1236595604796624e-05,
      "loss": 1.1002,
      "num_input_tokens_seen": 12149872,
      "step": 21925
    },
    {
      "epoch": 384.7433628318584,
      "grad_norm": 8.375,
      "learning_rate": 2.1226890293233693e-05,
      "loss": 0.9912,
      "num_input_tokens_seen": 12152832,
      "step": 21930
    },
    {
      "epoch": 384.83185840707966,
      "grad_norm": 8.4375,
      "learning_rate": 2.1217185563531694e-05,
      "loss": 1.0091,
      "num_input_tokens_seen": 12155360,
      "step": 21935
    },
    {
      "epoch": 384.92035398230087,
      "grad_norm": 7.1875,
      "learning_rate": 2.120748141718721e-05,
      "loss": 1.1036,
      "num_input_tokens_seen": 12157856,
      "step": 21940
    },
    {
      "epoch": 385.0,
      "grad_norm": 6.5625,
      "learning_rate": 2.1197777855696765e-05,
      "loss": 0.935,
      "num_input_tokens_seen": 12160032,
      "step": 21945
    },
    {
      "epoch": 385.08849557522126,
      "grad_norm": 9.0625,
      "learning_rate": 2.1188074880556746e-05,
      "loss": 1.1196,
      "num_input_tokens_seen": 12163216,
      "step": 21950
    },
    {
      "epoch": 385.17699115044246,
      "grad_norm": 9.3125,
      "learning_rate": 2.1178372493263495e-05,
      "loss": 1.0617,
      "num_input_tokens_seen": 12165456,
      "step": 21955
    },
    {
      "epoch": 385.2654867256637,
      "grad_norm": 6.75,
      "learning_rate": 2.116867069531322e-05,
      "loss": 0.9939,
      "num_input_tokens_seen": 12167888,
      "step": 21960
    },
    {
      "epoch": 385.353982300885,
      "grad_norm": 6.9375,
      "learning_rate": 2.1158969488202073e-05,
      "loss": 1.0717,
      "num_input_tokens_seen": 12170832,
      "step": 21965
    },
    {
      "epoch": 385.4424778761062,
      "grad_norm": 8.1875,
      "learning_rate": 2.114926887342611e-05,
      "loss": 0.7829,
      "num_input_tokens_seen": 12173632,
      "step": 21970
    },
    {
      "epoch": 385.53097345132744,
      "grad_norm": 8.3125,
      "learning_rate": 2.113956885248127e-05,
      "loss": 1.0031,
      "num_input_tokens_seen": 12176448,
      "step": 21975
    },
    {
      "epoch": 385.6194690265487,
      "grad_norm": 7.34375,
      "learning_rate": 2.112986942686342e-05,
      "loss": 1.1465,
      "num_input_tokens_seen": 12179152,
      "step": 21980
    },
    {
      "epoch": 385.7079646017699,
      "grad_norm": 7.34375,
      "learning_rate": 2.112017059806835e-05,
      "loss": 0.9471,
      "num_input_tokens_seen": 12182384,
      "step": 21985
    },
    {
      "epoch": 385.79646017699116,
      "grad_norm": 9.25,
      "learning_rate": 2.1110472367591724e-05,
      "loss": 1.1238,
      "num_input_tokens_seen": 12185056,
      "step": 21990
    },
    {
      "epoch": 385.88495575221236,
      "grad_norm": 10.125,
      "learning_rate": 2.1100774736929145e-05,
      "loss": 1.1304,
      "num_input_tokens_seen": 12187776,
      "step": 21995
    },
    {
      "epoch": 385.9734513274336,
      "grad_norm": 8.625,
      "learning_rate": 2.10910777075761e-05,
      "loss": 1.1651,
      "num_input_tokens_seen": 12190608,
      "step": 22000
    },
    {
      "epoch": 385.9734513274336,
      "eval_loss": 1.031529188156128,
      "eval_runtime": 0.9472,
      "eval_samples_per_second": 26.392,
      "eval_steps_per_second": 13.724,
      "num_input_tokens_seen": 12190608,
      "step": 22000
    },
    {
      "epoch": 386.05309734513276,
      "grad_norm": 7.40625,
      "learning_rate": 2.108138128102799e-05,
      "loss": 0.9711,
      "num_input_tokens_seen": 12193176,
      "step": 22005
    },
    {
      "epoch": 386.14159292035396,
      "grad_norm": 7.0,
      "learning_rate": 2.107168545878014e-05,
      "loss": 0.9158,
      "num_input_tokens_seen": 12196040,
      "step": 22010
    },
    {
      "epoch": 386.2300884955752,
      "grad_norm": 7.90625,
      "learning_rate": 2.106199024232775e-05,
      "loss": 0.9826,
      "num_input_tokens_seen": 12198856,
      "step": 22015
    },
    {
      "epoch": 386.3185840707965,
      "grad_norm": 7.25,
      "learning_rate": 2.105229563316595e-05,
      "loss": 1.1071,
      "num_input_tokens_seen": 12201576,
      "step": 22020
    },
    {
      "epoch": 386.4070796460177,
      "grad_norm": 8.4375,
      "learning_rate": 2.1042601632789784e-05,
      "loss": 1.0817,
      "num_input_tokens_seen": 12203928,
      "step": 22025
    },
    {
      "epoch": 386.49557522123894,
      "grad_norm": 9.0,
      "learning_rate": 2.103290824269417e-05,
      "loss": 1.0375,
      "num_input_tokens_seen": 12206632,
      "step": 22030
    },
    {
      "epoch": 386.5840707964602,
      "grad_norm": 9.9375,
      "learning_rate": 2.1023215464373965e-05,
      "loss": 1.2041,
      "num_input_tokens_seen": 12209688,
      "step": 22035
    },
    {
      "epoch": 386.6725663716814,
      "grad_norm": 8.0625,
      "learning_rate": 2.1013523299323908e-05,
      "loss": 0.8335,
      "num_input_tokens_seen": 12212184,
      "step": 22040
    },
    {
      "epoch": 386.76106194690266,
      "grad_norm": 8.8125,
      "learning_rate": 2.1003831749038654e-05,
      "loss": 1.3586,
      "num_input_tokens_seen": 12215128,
      "step": 22045
    },
    {
      "epoch": 386.8495575221239,
      "grad_norm": 9.8125,
      "learning_rate": 2.099414081501277e-05,
      "loss": 1.009,
      "num_input_tokens_seen": 12218136,
      "step": 22050
    },
    {
      "epoch": 386.9380530973451,
      "grad_norm": 9.125,
      "learning_rate": 2.09844504987407e-05,
      "loss": 0.9316,
      "num_input_tokens_seen": 12220968,
      "step": 22055
    },
    {
      "epoch": 387.01769911504425,
      "grad_norm": 6.6875,
      "learning_rate": 2.097476080171683e-05,
      "loss": 1.124,
      "num_input_tokens_seen": 12223416,
      "step": 22060
    },
    {
      "epoch": 387.1061946902655,
      "grad_norm": 9.625,
      "learning_rate": 2.0965071725435436e-05,
      "loss": 0.8768,
      "num_input_tokens_seen": 12226344,
      "step": 22065
    },
    {
      "epoch": 387.1946902654867,
      "grad_norm": 7.59375,
      "learning_rate": 2.0955383271390684e-05,
      "loss": 0.9439,
      "num_input_tokens_seen": 12229256,
      "step": 22070
    },
    {
      "epoch": 387.283185840708,
      "grad_norm": 8.25,
      "learning_rate": 2.094569544107666e-05,
      "loss": 0.9876,
      "num_input_tokens_seen": 12231912,
      "step": 22075
    },
    {
      "epoch": 387.37168141592923,
      "grad_norm": 7.75,
      "learning_rate": 2.093600823598735e-05,
      "loss": 1.2418,
      "num_input_tokens_seen": 12234312,
      "step": 22080
    },
    {
      "epoch": 387.46017699115043,
      "grad_norm": 7.65625,
      "learning_rate": 2.092632165761663e-05,
      "loss": 1.1513,
      "num_input_tokens_seen": 12237032,
      "step": 22085
    },
    {
      "epoch": 387.5486725663717,
      "grad_norm": 8.625,
      "learning_rate": 2.091663570745832e-05,
      "loss": 1.0764,
      "num_input_tokens_seen": 12239992,
      "step": 22090
    },
    {
      "epoch": 387.6371681415929,
      "grad_norm": 9.25,
      "learning_rate": 2.0906950387006086e-05,
      "loss": 1.0482,
      "num_input_tokens_seen": 12242808,
      "step": 22095
    },
    {
      "epoch": 387.72566371681415,
      "grad_norm": 8.9375,
      "learning_rate": 2.0897265697753543e-05,
      "loss": 1.1754,
      "num_input_tokens_seen": 12245672,
      "step": 22100
    },
    {
      "epoch": 387.8141592920354,
      "grad_norm": 8.0,
      "learning_rate": 2.088758164119419e-05,
      "loss": 1.0182,
      "num_input_tokens_seen": 12248744,
      "step": 22105
    },
    {
      "epoch": 387.9026548672566,
      "grad_norm": 9.1875,
      "learning_rate": 2.0877898218821428e-05,
      "loss": 0.9506,
      "num_input_tokens_seen": 12251656,
      "step": 22110
    },
    {
      "epoch": 387.9911504424779,
      "grad_norm": 7.78125,
      "learning_rate": 2.0868215432128565e-05,
      "loss": 1.1092,
      "num_input_tokens_seen": 12254696,
      "step": 22115
    },
    {
      "epoch": 388.070796460177,
      "grad_norm": 8.25,
      "learning_rate": 2.0858533282608796e-05,
      "loss": 1.1599,
      "num_input_tokens_seen": 12257480,
      "step": 22120
    },
    {
      "epoch": 388.1592920353982,
      "grad_norm": 7.21875,
      "learning_rate": 2.084885177175524e-05,
      "loss": 1.0668,
      "num_input_tokens_seen": 12260600,
      "step": 22125
    },
    {
      "epoch": 388.24778761061947,
      "grad_norm": 9.0625,
      "learning_rate": 2.0839170901060917e-05,
      "loss": 1.0033,
      "num_input_tokens_seen": 12263096,
      "step": 22130
    },
    {
      "epoch": 388.3362831858407,
      "grad_norm": 8.0625,
      "learning_rate": 2.082949067201872e-05,
      "loss": 0.873,
      "num_input_tokens_seen": 12265688,
      "step": 22135
    },
    {
      "epoch": 388.42477876106193,
      "grad_norm": 8.5625,
      "learning_rate": 2.0819811086121475e-05,
      "loss": 1.2635,
      "num_input_tokens_seen": 12268360,
      "step": 22140
    },
    {
      "epoch": 388.5132743362832,
      "grad_norm": 5.875,
      "learning_rate": 2.08101321448619e-05,
      "loss": 0.9708,
      "num_input_tokens_seen": 12271176,
      "step": 22145
    },
    {
      "epoch": 388.60176991150445,
      "grad_norm": 7.75,
      "learning_rate": 2.080045384973259e-05,
      "loss": 1.0115,
      "num_input_tokens_seen": 12273976,
      "step": 22150
    },
    {
      "epoch": 388.69026548672565,
      "grad_norm": 7.78125,
      "learning_rate": 2.0790776202226082e-05,
      "loss": 1.0355,
      "num_input_tokens_seen": 12276536,
      "step": 22155
    },
    {
      "epoch": 388.7787610619469,
      "grad_norm": 7.40625,
      "learning_rate": 2.078109920383477e-05,
      "loss": 0.9624,
      "num_input_tokens_seen": 12279336,
      "step": 22160
    },
    {
      "epoch": 388.86725663716817,
      "grad_norm": 11.5625,
      "learning_rate": 2.0771422856050978e-05,
      "loss": 1.0097,
      "num_input_tokens_seen": 12282024,
      "step": 22165
    },
    {
      "epoch": 388.95575221238937,
      "grad_norm": 10.875,
      "learning_rate": 2.076174716036693e-05,
      "loss": 1.2306,
      "num_input_tokens_seen": 12284984,
      "step": 22170
    },
    {
      "epoch": 389.0353982300885,
      "grad_norm": 11.5,
      "learning_rate": 2.075207211827472e-05,
      "loss": 1.0969,
      "num_input_tokens_seen": 12287296,
      "step": 22175
    },
    {
      "epoch": 389.12389380530976,
      "grad_norm": 8.5,
      "learning_rate": 2.074239773126638e-05,
      "loss": 1.0004,
      "num_input_tokens_seen": 12290160,
      "step": 22180
    },
    {
      "epoch": 389.21238938053096,
      "grad_norm": 9.25,
      "learning_rate": 2.073272400083382e-05,
      "loss": 1.2541,
      "num_input_tokens_seen": 12293120,
      "step": 22185
    },
    {
      "epoch": 389.3008849557522,
      "grad_norm": 6.78125,
      "learning_rate": 2.072305092846883e-05,
      "loss": 0.82,
      "num_input_tokens_seen": 12295568,
      "step": 22190
    },
    {
      "epoch": 389.3893805309734,
      "grad_norm": 8.5,
      "learning_rate": 2.0713378515663152e-05,
      "loss": 1.0907,
      "num_input_tokens_seen": 12298656,
      "step": 22195
    },
    {
      "epoch": 389.4778761061947,
      "grad_norm": 9.5,
      "learning_rate": 2.070370676390836e-05,
      "loss": 1.0961,
      "num_input_tokens_seen": 12301648,
      "step": 22200
    },
    {
      "epoch": 389.4778761061947,
      "eval_loss": 1.0216825008392334,
      "eval_runtime": 0.936,
      "eval_samples_per_second": 26.711,
      "eval_steps_per_second": 13.889,
      "num_input_tokens_seen": 12301648,
      "step": 22200
    },
    {
      "epoch": 389.56637168141594,
      "grad_norm": 8.6875,
      "learning_rate": 2.0694035674695974e-05,
      "loss": 1.1358,
      "num_input_tokens_seen": 12304560,
      "step": 22205
    },
    {
      "epoch": 389.65486725663715,
      "grad_norm": 7.90625,
      "learning_rate": 2.0684365249517416e-05,
      "loss": 1.1901,
      "num_input_tokens_seen": 12306912,
      "step": 22210
    },
    {
      "epoch": 389.7433628318584,
      "grad_norm": 6.59375,
      "learning_rate": 2.067469548986396e-05,
      "loss": 1.0174,
      "num_input_tokens_seen": 12309440,
      "step": 22215
    },
    {
      "epoch": 389.83185840707966,
      "grad_norm": 10.0,
      "learning_rate": 2.066502639722681e-05,
      "loss": 0.9044,
      "num_input_tokens_seen": 12312464,
      "step": 22220
    },
    {
      "epoch": 389.92035398230087,
      "grad_norm": 7.46875,
      "learning_rate": 2.065535797309708e-05,
      "loss": 0.9559,
      "num_input_tokens_seen": 12315392,
      "step": 22225
    },
    {
      "epoch": 390.0,
      "grad_norm": 8.1875,
      "learning_rate": 2.0645690218965736e-05,
      "loss": 1.1326,
      "num_input_tokens_seen": 12317784,
      "step": 22230
    },
    {
      "epoch": 390.08849557522126,
      "grad_norm": 8.0625,
      "learning_rate": 2.063602313632369e-05,
      "loss": 1.1351,
      "num_input_tokens_seen": 12321048,
      "step": 22235
    },
    {
      "epoch": 390.17699115044246,
      "grad_norm": 8.5625,
      "learning_rate": 2.0626356726661704e-05,
      "loss": 1.0616,
      "num_input_tokens_seen": 12324072,
      "step": 22240
    },
    {
      "epoch": 390.2654867256637,
      "grad_norm": 7.75,
      "learning_rate": 2.0616690991470477e-05,
      "loss": 1.097,
      "num_input_tokens_seen": 12326552,
      "step": 22245
    },
    {
      "epoch": 390.353982300885,
      "grad_norm": 9.125,
      "learning_rate": 2.0607025932240595e-05,
      "loss": 1.0231,
      "num_input_tokens_seen": 12329496,
      "step": 22250
    },
    {
      "epoch": 390.4424778761062,
      "grad_norm": 8.25,
      "learning_rate": 2.059736155046251e-05,
      "loss": 0.9646,
      "num_input_tokens_seen": 12332120,
      "step": 22255
    },
    {
      "epoch": 390.53097345132744,
      "grad_norm": 9.125,
      "learning_rate": 2.0587697847626603e-05,
      "loss": 1.058,
      "num_input_tokens_seen": 12335048,
      "step": 22260
    },
    {
      "epoch": 390.6194690265487,
      "grad_norm": 13.9375,
      "learning_rate": 2.057803482522314e-05,
      "loss": 1.1535,
      "num_input_tokens_seen": 12337608,
      "step": 22265
    },
    {
      "epoch": 390.7079646017699,
      "grad_norm": 6.6875,
      "learning_rate": 2.056837248474227e-05,
      "loss": 0.9977,
      "num_input_tokens_seen": 12340056,
      "step": 22270
    },
    {
      "epoch": 390.79646017699116,
      "grad_norm": 9.125,
      "learning_rate": 2.0558710827674064e-05,
      "loss": 1.1573,
      "num_input_tokens_seen": 12342632,
      "step": 22275
    },
    {
      "epoch": 390.88495575221236,
      "grad_norm": 5.0,
      "learning_rate": 2.054904985550845e-05,
      "loss": 0.945,
      "num_input_tokens_seen": 12345288,
      "step": 22280
    },
    {
      "epoch": 390.9734513274336,
      "grad_norm": 6.875,
      "learning_rate": 2.0539389569735287e-05,
      "loss": 0.9461,
      "num_input_tokens_seen": 12348792,
      "step": 22285
    },
    {
      "epoch": 391.05309734513276,
      "grad_norm": 7.125,
      "learning_rate": 2.052972997184431e-05,
      "loss": 1.6287,
      "num_input_tokens_seen": 12351176,
      "step": 22290
    },
    {
      "epoch": 391.14159292035396,
      "grad_norm": 6.625,
      "learning_rate": 2.0520071063325146e-05,
      "loss": 0.8949,
      "num_input_tokens_seen": 12353784,
      "step": 22295
    },
    {
      "epoch": 391.2300884955752,
      "grad_norm": 8.3125,
      "learning_rate": 2.051041284566732e-05,
      "loss": 1.0581,
      "num_input_tokens_seen": 12357128,
      "step": 22300
    },
    {
      "epoch": 391.3185840707965,
      "grad_norm": 8.25,
      "learning_rate": 2.050075532036026e-05,
      "loss": 1.1056,
      "num_input_tokens_seen": 12359976,
      "step": 22305
    },
    {
      "epoch": 391.4070796460177,
      "grad_norm": 8.0625,
      "learning_rate": 2.0491098488893264e-05,
      "loss": 0.9702,
      "num_input_tokens_seen": 12363224,
      "step": 22310
    },
    {
      "epoch": 391.49557522123894,
      "grad_norm": 10.6875,
      "learning_rate": 2.0481442352755546e-05,
      "loss": 0.9708,
      "num_input_tokens_seen": 12365688,
      "step": 22315
    },
    {
      "epoch": 391.5840707964602,
      "grad_norm": 10.25,
      "learning_rate": 2.0471786913436198e-05,
      "loss": 1.2156,
      "num_input_tokens_seen": 12368504,
      "step": 22320
    },
    {
      "epoch": 391.6725663716814,
      "grad_norm": 7.40625,
      "learning_rate": 2.0462132172424218e-05,
      "loss": 1.02,
      "num_input_tokens_seen": 12371032,
      "step": 22325
    },
    {
      "epoch": 391.76106194690266,
      "grad_norm": 9.75,
      "learning_rate": 2.0452478131208484e-05,
      "loss": 1.0133,
      "num_input_tokens_seen": 12373464,
      "step": 22330
    },
    {
      "epoch": 391.8495575221239,
      "grad_norm": 10.3125,
      "learning_rate": 2.0442824791277765e-05,
      "loss": 1.1514,
      "num_input_tokens_seen": 12376088,
      "step": 22335
    },
    {
      "epoch": 391.9380530973451,
      "grad_norm": 8.875,
      "learning_rate": 2.0433172154120727e-05,
      "loss": 1.0261,
      "num_input_tokens_seen": 12378920,
      "step": 22340
    },
    {
      "epoch": 392.01769911504425,
      "grad_norm": 6.75,
      "learning_rate": 2.0423520221225947e-05,
      "loss": 1.0857,
      "num_input_tokens_seen": 12381328,
      "step": 22345
    },
    {
      "epoch": 392.1061946902655,
      "grad_norm": 8.3125,
      "learning_rate": 2.0413868994081848e-05,
      "loss": 1.0564,
      "num_input_tokens_seen": 12383968,
      "step": 22350
    },
    {
      "epoch": 392.1946902654867,
      "grad_norm": 6.59375,
      "learning_rate": 2.0404218474176795e-05,
      "loss": 1.0809,
      "num_input_tokens_seen": 12387248,
      "step": 22355
    },
    {
      "epoch": 392.283185840708,
      "grad_norm": 11.4375,
      "learning_rate": 2.0394568662999002e-05,
      "loss": 1.1679,
      "num_input_tokens_seen": 12389952,
      "step": 22360
    },
    {
      "epoch": 392.37168141592923,
      "grad_norm": 7.4375,
      "learning_rate": 2.0384919562036593e-05,
      "loss": 0.9922,
      "num_input_tokens_seen": 12392688,
      "step": 22365
    },
    {
      "epoch": 392.46017699115043,
      "grad_norm": 10.0,
      "learning_rate": 2.0375271172777593e-05,
      "loss": 0.9307,
      "num_input_tokens_seen": 12395376,
      "step": 22370
    },
    {
      "epoch": 392.5486725663717,
      "grad_norm": 8.9375,
      "learning_rate": 2.0365623496709885e-05,
      "loss": 1.0906,
      "num_input_tokens_seen": 12398736,
      "step": 22375
    },
    {
      "epoch": 392.6371681415929,
      "grad_norm": 9.375,
      "learning_rate": 2.0355976535321283e-05,
      "loss": 1.1483,
      "num_input_tokens_seen": 12401536,
      "step": 22380
    },
    {
      "epoch": 392.72566371681415,
      "grad_norm": 8.875,
      "learning_rate": 2.034633029009945e-05,
      "loss": 1.0982,
      "num_input_tokens_seen": 12404176,
      "step": 22385
    },
    {
      "epoch": 392.8141592920354,
      "grad_norm": 9.125,
      "learning_rate": 2.0336684762531972e-05,
      "loss": 0.9385,
      "num_input_tokens_seen": 12407008,
      "step": 22390
    },
    {
      "epoch": 392.9026548672566,
      "grad_norm": 8.4375,
      "learning_rate": 2.032703995410631e-05,
      "loss": 1.0721,
      "num_input_tokens_seen": 12409680,
      "step": 22395
    },
    {
      "epoch": 392.9911504424779,
      "grad_norm": 7.03125,
      "learning_rate": 2.031739586630981e-05,
      "loss": 1.0122,
      "num_input_tokens_seen": 12412384,
      "step": 22400
    },
    {
      "epoch": 392.9911504424779,
      "eval_loss": 1.0289069414138794,
      "eval_runtime": 0.9365,
      "eval_samples_per_second": 26.694,
      "eval_steps_per_second": 13.881,
      "num_input_tokens_seen": 12412384,
      "step": 22400
    },
    {
      "epoch": 393.070796460177,
      "grad_norm": 5.875,
      "learning_rate": 2.0307752500629707e-05,
      "loss": 1.0497,
      "num_input_tokens_seen": 12414912,
      "step": 22405
    },
    {
      "epoch": 393.1592920353982,
      "grad_norm": 10.0625,
      "learning_rate": 2.0298109858553144e-05,
      "loss": 1.2096,
      "num_input_tokens_seen": 12417792,
      "step": 22410
    },
    {
      "epoch": 393.24778761061947,
      "grad_norm": 10.125,
      "learning_rate": 2.028846794156712e-05,
      "loss": 0.9876,
      "num_input_tokens_seen": 12420784,
      "step": 22415
    },
    {
      "epoch": 393.3362831858407,
      "grad_norm": 8.9375,
      "learning_rate": 2.027882675115856e-05,
      "loss": 1.0005,
      "num_input_tokens_seen": 12423600,
      "step": 22420
    },
    {
      "epoch": 393.42477876106193,
      "grad_norm": 8.6875,
      "learning_rate": 2.026918628881423e-05,
      "loss": 1.0643,
      "num_input_tokens_seen": 12426192,
      "step": 22425
    },
    {
      "epoch": 393.5132743362832,
      "grad_norm": 7.28125,
      "learning_rate": 2.0259546556020833e-05,
      "loss": 1.0656,
      "num_input_tokens_seen": 12428704,
      "step": 22430
    },
    {
      "epoch": 393.60176991150445,
      "grad_norm": 7.84375,
      "learning_rate": 2.024990755426493e-05,
      "loss": 0.9682,
      "num_input_tokens_seen": 12430960,
      "step": 22435
    },
    {
      "epoch": 393.69026548672565,
      "grad_norm": 9.0625,
      "learning_rate": 2.0240269285032975e-05,
      "loss": 1.0675,
      "num_input_tokens_seen": 12434192,
      "step": 22440
    },
    {
      "epoch": 393.7787610619469,
      "grad_norm": 8.6875,
      "learning_rate": 2.0230631749811306e-05,
      "loss": 1.1457,
      "num_input_tokens_seen": 12436752,
      "step": 22445
    },
    {
      "epoch": 393.86725663716817,
      "grad_norm": 6.71875,
      "learning_rate": 2.0220994950086162e-05,
      "loss": 1.006,
      "num_input_tokens_seen": 12439792,
      "step": 22450
    },
    {
      "epoch": 393.95575221238937,
      "grad_norm": 12.125,
      "learning_rate": 2.021135888734365e-05,
      "loss": 0.954,
      "num_input_tokens_seen": 12443136,
      "step": 22455
    },
    {
      "epoch": 394.0353982300885,
      "grad_norm": 8.0625,
      "learning_rate": 2.0201723563069783e-05,
      "loss": 1.2422,
      "num_input_tokens_seen": 12445288,
      "step": 22460
    },
    {
      "epoch": 394.12389380530976,
      "grad_norm": 7.1875,
      "learning_rate": 2.0192088978750433e-05,
      "loss": 1.2643,
      "num_input_tokens_seen": 12447944,
      "step": 22465
    },
    {
      "epoch": 394.21238938053096,
      "grad_norm": 9.75,
      "learning_rate": 2.0182455135871385e-05,
      "loss": 0.9564,
      "num_input_tokens_seen": 12450792,
      "step": 22470
    },
    {
      "epoch": 394.3008849557522,
      "grad_norm": 8.125,
      "learning_rate": 2.0172822035918305e-05,
      "loss": 1.0786,
      "num_input_tokens_seen": 12453512,
      "step": 22475
    },
    {
      "epoch": 394.3893805309734,
      "grad_norm": 7.28125,
      "learning_rate": 2.016318968037671e-05,
      "loss": 0.9551,
      "num_input_tokens_seen": 12456472,
      "step": 22480
    },
    {
      "epoch": 394.4778761061947,
      "grad_norm": 8.5,
      "learning_rate": 2.015355807073206e-05,
      "loss": 1.082,
      "num_input_tokens_seen": 12459080,
      "step": 22485
    },
    {
      "epoch": 394.56637168141594,
      "grad_norm": 10.8125,
      "learning_rate": 2.0143927208469664e-05,
      "loss": 1.1676,
      "num_input_tokens_seen": 12462232,
      "step": 22490
    },
    {
      "epoch": 394.65486725663715,
      "grad_norm": 9.5,
      "learning_rate": 2.0134297095074708e-05,
      "loss": 0.8998,
      "num_input_tokens_seen": 12464872,
      "step": 22495
    },
    {
      "epoch": 394.7433628318584,
      "grad_norm": 8.4375,
      "learning_rate": 2.0124667732032297e-05,
      "loss": 1.0311,
      "num_input_tokens_seen": 12467720,
      "step": 22500
    },
    {
      "epoch": 394.83185840707966,
      "grad_norm": 12.875,
      "learning_rate": 2.011503912082738e-05,
      "loss": 1.1629,
      "num_input_tokens_seen": 12470616,
      "step": 22505
    },
    {
      "epoch": 394.92035398230087,
      "grad_norm": 5.84375,
      "learning_rate": 2.0105411262944823e-05,
      "loss": 0.8838,
      "num_input_tokens_seen": 12473288,
      "step": 22510
    },
    {
      "epoch": 395.0,
      "grad_norm": 9.3125,
      "learning_rate": 2.0095784159869366e-05,
      "loss": 1.1109,
      "num_input_tokens_seen": 12475984,
      "step": 22515
    },
    {
      "epoch": 395.08849557522126,
      "grad_norm": 7.34375,
      "learning_rate": 2.0086157813085608e-05,
      "loss": 1.1386,
      "num_input_tokens_seen": 12478560,
      "step": 22520
    },
    {
      "epoch": 395.17699115044246,
      "grad_norm": 7.96875,
      "learning_rate": 2.0076532224078068e-05,
      "loss": 1.1379,
      "num_input_tokens_seen": 12481248,
      "step": 22525
    },
    {
      "epoch": 395.2654867256637,
      "grad_norm": 11.5625,
      "learning_rate": 2.0066907394331142e-05,
      "loss": 1.0545,
      "num_input_tokens_seen": 12484736,
      "step": 22530
    },
    {
      "epoch": 395.353982300885,
      "grad_norm": 7.3125,
      "learning_rate": 2.0057283325329077e-05,
      "loss": 1.068,
      "num_input_tokens_seen": 12487712,
      "step": 22535
    },
    {
      "epoch": 395.4424778761062,
      "grad_norm": 10.5,
      "learning_rate": 2.0047660018556047e-05,
      "loss": 0.8826,
      "num_input_tokens_seen": 12490192,
      "step": 22540
    },
    {
      "epoch": 395.53097345132744,
      "grad_norm": 8.75,
      "learning_rate": 2.0038037475496075e-05,
      "loss": 1.063,
      "num_input_tokens_seen": 12492800,
      "step": 22545
    },
    {
      "epoch": 395.6194690265487,
      "grad_norm": 8.5625,
      "learning_rate": 2.0028415697633073e-05,
      "loss": 0.9106,
      "num_input_tokens_seen": 12495408,
      "step": 22550
    },
    {
      "epoch": 395.7079646017699,
      "grad_norm": 7.78125,
      "learning_rate": 2.0018794686450858e-05,
      "loss": 0.9919,
      "num_input_tokens_seen": 12498144,
      "step": 22555
    },
    {
      "epoch": 395.79646017699116,
      "grad_norm": 8.5,
      "learning_rate": 2.0009174443433088e-05,
      "loss": 1.2908,
      "num_input_tokens_seen": 12501040,
      "step": 22560
    },
    {
      "epoch": 395.88495575221236,
      "grad_norm": 7.34375,
      "learning_rate": 1.999955497006334e-05,
      "loss": 1.146,
      "num_input_tokens_seen": 12504016,
      "step": 22565
    },
    {
      "epoch": 395.9734513274336,
      "grad_norm": 6.5,
      "learning_rate": 1.9989936267825067e-05,
      "loss": 0.9876,
      "num_input_tokens_seen": 12506736,
      "step": 22570
    },
    {
      "epoch": 396.05309734513276,
      "grad_norm": 5.875,
      "learning_rate": 1.9980318338201572e-05,
      "loss": 0.9684,
      "num_input_tokens_seen": 12509008,
      "step": 22575
    },
    {
      "epoch": 396.14159292035396,
      "grad_norm": 8.5625,
      "learning_rate": 1.997070118267607e-05,
      "loss": 1.1307,
      "num_input_tokens_seen": 12511872,
      "step": 22580
    },
    {
      "epoch": 396.2300884955752,
      "grad_norm": 7.25,
      "learning_rate": 1.9961084802731654e-05,
      "loss": 1.0925,
      "num_input_tokens_seen": 12514384,
      "step": 22585
    },
    {
      "epoch": 396.3185840707965,
      "grad_norm": 12.625,
      "learning_rate": 1.9951469199851273e-05,
      "loss": 1.1046,
      "num_input_tokens_seen": 12517328,
      "step": 22590
    },
    {
      "epoch": 396.4070796460177,
      "grad_norm": 8.4375,
      "learning_rate": 1.99418543755178e-05,
      "loss": 1.1654,
      "num_input_tokens_seen": 12519952,
      "step": 22595
    },
    {
      "epoch": 396.49557522123894,
      "grad_norm": 7.5625,
      "learning_rate": 1.9932240331213936e-05,
      "loss": 1.0517,
      "num_input_tokens_seen": 12523264,
      "step": 22600
    },
    {
      "epoch": 396.49557522123894,
      "eval_loss": 1.0287283658981323,
      "eval_runtime": 0.9378,
      "eval_samples_per_second": 26.657,
      "eval_steps_per_second": 13.862,
      "num_input_tokens_seen": 12523264,
      "step": 22600
    },
    {
      "epoch": 396.5840707964602,
      "grad_norm": 7.125,
      "learning_rate": 1.9922627068422297e-05,
      "loss": 0.962,
      "num_input_tokens_seen": 12526864,
      "step": 22605
    },
    {
      "epoch": 396.6725663716814,
      "grad_norm": 7.4375,
      "learning_rate": 1.991301458862538e-05,
      "loss": 0.8776,
      "num_input_tokens_seen": 12529712,
      "step": 22610
    },
    {
      "epoch": 396.76106194690266,
      "grad_norm": 6.8125,
      "learning_rate": 1.9903402893305536e-05,
      "loss": 1.1618,
      "num_input_tokens_seen": 12532288,
      "step": 22615
    },
    {
      "epoch": 396.8495575221239,
      "grad_norm": 9.375,
      "learning_rate": 1.9893791983945016e-05,
      "loss": 1.1024,
      "num_input_tokens_seen": 12534896,
      "step": 22620
    },
    {
      "epoch": 396.9380530973451,
      "grad_norm": 7.0,
      "learning_rate": 1.988418186202594e-05,
      "loss": 1.1149,
      "num_input_tokens_seen": 12537360,
      "step": 22625
    },
    {
      "epoch": 397.01769911504425,
      "grad_norm": 8.1875,
      "learning_rate": 1.98745725290303e-05,
      "loss": 0.9912,
      "num_input_tokens_seen": 12539384,
      "step": 22630
    },
    {
      "epoch": 397.1061946902655,
      "grad_norm": 10.8125,
      "learning_rate": 1.986496398644e-05,
      "loss": 1.1499,
      "num_input_tokens_seen": 12542040,
      "step": 22635
    },
    {
      "epoch": 397.1946902654867,
      "grad_norm": 6.875,
      "learning_rate": 1.9855356235736777e-05,
      "loss": 0.7756,
      "num_input_tokens_seen": 12544600,
      "step": 22640
    },
    {
      "epoch": 397.283185840708,
      "grad_norm": 7.5625,
      "learning_rate": 1.9845749278402277e-05,
      "loss": 0.9735,
      "num_input_tokens_seen": 12547432,
      "step": 22645
    },
    {
      "epoch": 397.37168141592923,
      "grad_norm": 8.4375,
      "learning_rate": 1.9836143115918006e-05,
      "loss": 0.9059,
      "num_input_tokens_seen": 12550152,
      "step": 22650
    },
    {
      "epoch": 397.46017699115043,
      "grad_norm": 8.5625,
      "learning_rate": 1.9826537749765367e-05,
      "loss": 0.9978,
      "num_input_tokens_seen": 12552792,
      "step": 22655
    },
    {
      "epoch": 397.5486725663717,
      "grad_norm": 7.90625,
      "learning_rate": 1.9816933181425625e-05,
      "loss": 1.0502,
      "num_input_tokens_seen": 12555384,
      "step": 22660
    },
    {
      "epoch": 397.6371681415929,
      "grad_norm": 8.375,
      "learning_rate": 1.9807329412379903e-05,
      "loss": 1.1106,
      "num_input_tokens_seen": 12558328,
      "step": 22665
    },
    {
      "epoch": 397.72566371681415,
      "grad_norm": 13.8125,
      "learning_rate": 1.9797726444109247e-05,
      "loss": 1.4616,
      "num_input_tokens_seen": 12561176,
      "step": 22670
    },
    {
      "epoch": 397.8141592920354,
      "grad_norm": 9.875,
      "learning_rate": 1.9788124278094557e-05,
      "loss": 1.1601,
      "num_input_tokens_seen": 12564216,
      "step": 22675
    },
    {
      "epoch": 397.9026548672566,
      "grad_norm": 7.8125,
      "learning_rate": 1.9778522915816594e-05,
      "loss": 1.0609,
      "num_input_tokens_seen": 12566968,
      "step": 22680
    },
    {
      "epoch": 397.9911504424779,
      "grad_norm": 7.1875,
      "learning_rate": 1.9768922358756014e-05,
      "loss": 0.9881,
      "num_input_tokens_seen": 12570584,
      "step": 22685
    },
    {
      "epoch": 398.070796460177,
      "grad_norm": 9.0,
      "learning_rate": 1.9759322608393353e-05,
      "loss": 1.0918,
      "num_input_tokens_seen": 12573200,
      "step": 22690
    },
    {
      "epoch": 398.1592920353982,
      "grad_norm": 8.75,
      "learning_rate": 1.9749723666208992e-05,
      "loss": 1.1854,
      "num_input_tokens_seen": 12575664,
      "step": 22695
    },
    {
      "epoch": 398.24778761061947,
      "grad_norm": 5.65625,
      "learning_rate": 1.9740125533683235e-05,
      "loss": 0.9191,
      "num_input_tokens_seen": 12578512,
      "step": 22700
    },
    {
      "epoch": 398.3362831858407,
      "grad_norm": 8.3125,
      "learning_rate": 1.9730528212296208e-05,
      "loss": 1.1078,
      "num_input_tokens_seen": 12581440,
      "step": 22705
    },
    {
      "epoch": 398.42477876106193,
      "grad_norm": 8.4375,
      "learning_rate": 1.9720931703527945e-05,
      "loss": 1.0719,
      "num_input_tokens_seen": 12584240,
      "step": 22710
    },
    {
      "epoch": 398.5132743362832,
      "grad_norm": 6.21875,
      "learning_rate": 1.9711336008858373e-05,
      "loss": 0.948,
      "num_input_tokens_seen": 12586800,
      "step": 22715
    },
    {
      "epoch": 398.60176991150445,
      "grad_norm": 6.8125,
      "learning_rate": 1.9701741129767233e-05,
      "loss": 1.1122,
      "num_input_tokens_seen": 12589664,
      "step": 22720
    },
    {
      "epoch": 398.69026548672565,
      "grad_norm": 7.09375,
      "learning_rate": 1.9692147067734202e-05,
      "loss": 0.8723,
      "num_input_tokens_seen": 12592608,
      "step": 22725
    },
    {
      "epoch": 398.7787610619469,
      "grad_norm": 8.0625,
      "learning_rate": 1.96825538242388e-05,
      "loss": 1.0909,
      "num_input_tokens_seen": 12595600,
      "step": 22730
    },
    {
      "epoch": 398.86725663716817,
      "grad_norm": 8.5625,
      "learning_rate": 1.967296140076041e-05,
      "loss": 0.9648,
      "num_input_tokens_seen": 12598496,
      "step": 22735
    },
    {
      "epoch": 398.95575221238937,
      "grad_norm": 10.25,
      "learning_rate": 1.966336979877833e-05,
      "loss": 1.2193,
      "num_input_tokens_seen": 12600928,
      "step": 22740
    },
    {
      "epoch": 399.0353982300885,
      "grad_norm": 7.0625,
      "learning_rate": 1.9653779019771678e-05,
      "loss": 0.912,
      "num_input_tokens_seen": 12603184,
      "step": 22745
    },
    {
      "epoch": 399.12389380530976,
      "grad_norm": 13.75,
      "learning_rate": 1.9644189065219488e-05,
      "loss": 1.0921,
      "num_input_tokens_seen": 12605744,
      "step": 22750
    },
    {
      "epoch": 399.21238938053096,
      "grad_norm": 8.4375,
      "learning_rate": 1.9634599936600655e-05,
      "loss": 1.1426,
      "num_input_tokens_seen": 12608784,
      "step": 22755
    },
    {
      "epoch": 399.3008849557522,
      "grad_norm": 9.5,
      "learning_rate": 1.9625011635393935e-05,
      "loss": 1.169,
      "num_input_tokens_seen": 12611504,
      "step": 22760
    },
    {
      "epoch": 399.3893805309734,
      "grad_norm": 10.0625,
      "learning_rate": 1.9615424163077963e-05,
      "loss": 1.0737,
      "num_input_tokens_seen": 12614336,
      "step": 22765
    },
    {
      "epoch": 399.4778761061947,
      "grad_norm": 10.6875,
      "learning_rate": 1.9605837521131263e-05,
      "loss": 0.9371,
      "num_input_tokens_seen": 12617408,
      "step": 22770
    },
    {
      "epoch": 399.56637168141594,
      "grad_norm": 10.5625,
      "learning_rate": 1.9596251711032192e-05,
      "loss": 1.054,
      "num_input_tokens_seen": 12619872,
      "step": 22775
    },
    {
      "epoch": 399.65486725663715,
      "grad_norm": 10.75,
      "learning_rate": 1.958666673425903e-05,
      "loss": 1.0225,
      "num_input_tokens_seen": 12623312,
      "step": 22780
    },
    {
      "epoch": 399.7433628318584,
      "grad_norm": 8.4375,
      "learning_rate": 1.957708259228987e-05,
      "loss": 0.9264,
      "num_input_tokens_seen": 12626032,
      "step": 22785
    },
    {
      "epoch": 399.83185840707966,
      "grad_norm": 7.71875,
      "learning_rate": 1.956749928660273e-05,
      "loss": 1.0661,
      "num_input_tokens_seen": 12628864,
      "step": 22790
    },
    {
      "epoch": 399.92035398230087,
      "grad_norm": 9.625,
      "learning_rate": 1.955791681867547e-05,
      "loss": 1.1598,
      "num_input_tokens_seen": 12631728,
      "step": 22795
    },
    {
      "epoch": 400.0,
      "grad_norm": 7.3125,
      "learning_rate": 1.9548335189985824e-05,
      "loss": 0.9128,
      "num_input_tokens_seen": 12633656,
      "step": 22800
    },
    {
      "epoch": 400.0,
      "eval_loss": 1.0313019752502441,
      "eval_runtime": 0.9341,
      "eval_samples_per_second": 26.764,
      "eval_steps_per_second": 13.917,
      "num_input_tokens_seen": 12633656,
      "step": 22800
    },
    {
      "epoch": 400.08849557522126,
      "grad_norm": 7.71875,
      "learning_rate": 1.9538754402011396e-05,
      "loss": 0.9512,
      "num_input_tokens_seen": 12636680,
      "step": 22805
    },
    {
      "epoch": 400.17699115044246,
      "grad_norm": 6.65625,
      "learning_rate": 1.952917445622968e-05,
      "loss": 0.9389,
      "num_input_tokens_seen": 12639176,
      "step": 22810
    },
    {
      "epoch": 400.2654867256637,
      "grad_norm": 7.21875,
      "learning_rate": 1.9519595354118005e-05,
      "loss": 1.1646,
      "num_input_tokens_seen": 12642072,
      "step": 22815
    },
    {
      "epoch": 400.353982300885,
      "grad_norm": 10.375,
      "learning_rate": 1.951001709715361e-05,
      "loss": 1.2072,
      "num_input_tokens_seen": 12644616,
      "step": 22820
    },
    {
      "epoch": 400.4424778761062,
      "grad_norm": 7.0625,
      "learning_rate": 1.9500439686813556e-05,
      "loss": 0.9432,
      "num_input_tokens_seen": 12647240,
      "step": 22825
    },
    {
      "epoch": 400.53097345132744,
      "grad_norm": 7.75,
      "learning_rate": 1.949086312457482e-05,
      "loss": 0.9541,
      "num_input_tokens_seen": 12650216,
      "step": 22830
    },
    {
      "epoch": 400.6194690265487,
      "grad_norm": 6.71875,
      "learning_rate": 1.9481287411914223e-05,
      "loss": 1.0048,
      "num_input_tokens_seen": 12653160,
      "step": 22835
    },
    {
      "epoch": 400.7079646017699,
      "grad_norm": 8.3125,
      "learning_rate": 1.9471712550308457e-05,
      "loss": 1.1852,
      "num_input_tokens_seen": 12656024,
      "step": 22840
    },
    {
      "epoch": 400.79646017699116,
      "grad_norm": 9.0625,
      "learning_rate": 1.946213854123409e-05,
      "loss": 1.0892,
      "num_input_tokens_seen": 12658840,
      "step": 22845
    },
    {
      "epoch": 400.88495575221236,
      "grad_norm": 7.1875,
      "learning_rate": 1.9452565386167554e-05,
      "loss": 1.3122,
      "num_input_tokens_seen": 12661400,
      "step": 22850
    },
    {
      "epoch": 400.9734513274336,
      "grad_norm": 7.65625,
      "learning_rate": 1.9442993086585142e-05,
      "loss": 0.8485,
      "num_input_tokens_seen": 12664168,
      "step": 22855
    },
    {
      "epoch": 401.05309734513276,
      "grad_norm": 6.25,
      "learning_rate": 1.9433421643963043e-05,
      "loss": 0.7583,
      "num_input_tokens_seen": 12666592,
      "step": 22860
    },
    {
      "epoch": 401.14159292035396,
      "grad_norm": 7.3125,
      "learning_rate": 1.942385105977727e-05,
      "loss": 1.0847,
      "num_input_tokens_seen": 12669488,
      "step": 22865
    },
    {
      "epoch": 401.2300884955752,
      "grad_norm": 9.8125,
      "learning_rate": 1.9414281335503743e-05,
      "loss": 1.0008,
      "num_input_tokens_seen": 12672240,
      "step": 22870
    },
    {
      "epoch": 401.3185840707965,
      "grad_norm": 10.1875,
      "learning_rate": 1.9404712472618232e-05,
      "loss": 1.0539,
      "num_input_tokens_seen": 12674624,
      "step": 22875
    },
    {
      "epoch": 401.4070796460177,
      "grad_norm": 6.75,
      "learning_rate": 1.939514447259636e-05,
      "loss": 0.9836,
      "num_input_tokens_seen": 12677424,
      "step": 22880
    },
    {
      "epoch": 401.49557522123894,
      "grad_norm": 10.1875,
      "learning_rate": 1.938557733691365e-05,
      "loss": 1.0793,
      "num_input_tokens_seen": 12680288,
      "step": 22885
    },
    {
      "epoch": 401.5840707964602,
      "grad_norm": 7.4375,
      "learning_rate": 1.9376011067045476e-05,
      "loss": 0.9877,
      "num_input_tokens_seen": 12683248,
      "step": 22890
    },
    {
      "epoch": 401.6725663716814,
      "grad_norm": 8.25,
      "learning_rate": 1.9366445664467065e-05,
      "loss": 0.9523,
      "num_input_tokens_seen": 12686576,
      "step": 22895
    },
    {
      "epoch": 401.76106194690266,
      "grad_norm": 6.9375,
      "learning_rate": 1.9356881130653533e-05,
      "loss": 0.9432,
      "num_input_tokens_seen": 12689760,
      "step": 22900
    },
    {
      "epoch": 401.8495575221239,
      "grad_norm": 8.625,
      "learning_rate": 1.9347317467079846e-05,
      "loss": 1.2858,
      "num_input_tokens_seen": 12692160,
      "step": 22905
    },
    {
      "epoch": 401.9380530973451,
      "grad_norm": 9.5625,
      "learning_rate": 1.9337754675220836e-05,
      "loss": 1.2106,
      "num_input_tokens_seen": 12694720,
      "step": 22910
    },
    {
      "epoch": 402.01769911504425,
      "grad_norm": 7.0,
      "learning_rate": 1.9328192756551218e-05,
      "loss": 1.0559,
      "num_input_tokens_seen": 12697104,
      "step": 22915
    },
    {
      "epoch": 402.1061946902655,
      "grad_norm": 7.84375,
      "learning_rate": 1.931863171254555e-05,
      "loss": 0.917,
      "num_input_tokens_seen": 12699904,
      "step": 22920
    },
    {
      "epoch": 402.1946902654867,
      "grad_norm": 8.125,
      "learning_rate": 1.930907154467826e-05,
      "loss": 0.9398,
      "num_input_tokens_seen": 12702416,
      "step": 22925
    },
    {
      "epoch": 402.283185840708,
      "grad_norm": 7.84375,
      "learning_rate": 1.9299512254423673e-05,
      "loss": 1.0525,
      "num_input_tokens_seen": 12705584,
      "step": 22930
    },
    {
      "epoch": 402.37168141592923,
      "grad_norm": 6.71875,
      "learning_rate": 1.9289953843255914e-05,
      "loss": 1.259,
      "num_input_tokens_seen": 12708208,
      "step": 22935
    },
    {
      "epoch": 402.46017699115043,
      "grad_norm": 9.75,
      "learning_rate": 1.9280396312649048e-05,
      "loss": 0.9463,
      "num_input_tokens_seen": 12711040,
      "step": 22940
    },
    {
      "epoch": 402.5486725663717,
      "grad_norm": 9.125,
      "learning_rate": 1.9270839664076936e-05,
      "loss": 1.1198,
      "num_input_tokens_seen": 12713808,
      "step": 22945
    },
    {
      "epoch": 402.6371681415929,
      "grad_norm": 9.0625,
      "learning_rate": 1.9261283899013345e-05,
      "loss": 0.9513,
      "num_input_tokens_seen": 12716992,
      "step": 22950
    },
    {
      "epoch": 402.72566371681415,
      "grad_norm": 7.4375,
      "learning_rate": 1.92517290189319e-05,
      "loss": 1.1228,
      "num_input_tokens_seen": 12719808,
      "step": 22955
    },
    {
      "epoch": 402.8141592920354,
      "grad_norm": 8.5,
      "learning_rate": 1.924217502530607e-05,
      "loss": 1.1123,
      "num_input_tokens_seen": 12722688,
      "step": 22960
    },
    {
      "epoch": 402.9026548672566,
      "grad_norm": 8.625,
      "learning_rate": 1.9232621919609207e-05,
      "loss": 1.1588,
      "num_input_tokens_seen": 12725296,
      "step": 22965
    },
    {
      "epoch": 402.9911504424779,
      "grad_norm": 7.84375,
      "learning_rate": 1.9223069703314534e-05,
      "loss": 1.0557,
      "num_input_tokens_seen": 12728080,
      "step": 22970
    },
    {
      "epoch": 403.070796460177,
      "grad_norm": 8.3125,
      "learning_rate": 1.92135183778951e-05,
      "loss": 1.075,
      "num_input_tokens_seen": 12730408,
      "step": 22975
    },
    {
      "epoch": 403.1592920353982,
      "grad_norm": 7.78125,
      "learning_rate": 1.9203967944823857e-05,
      "loss": 1.0399,
      "num_input_tokens_seen": 12732648,
      "step": 22980
    },
    {
      "epoch": 403.24778761061947,
      "grad_norm": 10.1875,
      "learning_rate": 1.9194418405573588e-05,
      "loss": 1.0781,
      "num_input_tokens_seen": 12735576,
      "step": 22985
    },
    {
      "epoch": 403.3362831858407,
      "grad_norm": 7.34375,
      "learning_rate": 1.9184869761616954e-05,
      "loss": 0.8924,
      "num_input_tokens_seen": 12738168,
      "step": 22990
    },
    {
      "epoch": 403.42477876106193,
      "grad_norm": 7.46875,
      "learning_rate": 1.9175322014426495e-05,
      "loss": 0.9953,
      "num_input_tokens_seen": 12741256,
      "step": 22995
    },
    {
      "epoch": 403.5132743362832,
      "grad_norm": 8.875,
      "learning_rate": 1.9165775165474565e-05,
      "loss": 1.134,
      "num_input_tokens_seen": 12743928,
      "step": 23000
    },
    {
      "epoch": 403.5132743362832,
      "eval_loss": 1.0303553342819214,
      "eval_runtime": 0.9376,
      "eval_samples_per_second": 26.663,
      "eval_steps_per_second": 13.865,
      "num_input_tokens_seen": 12743928,
      "step": 23000
    },
    {
      "epoch": 403.60176991150445,
      "grad_norm": 7.71875,
      "learning_rate": 1.9156229216233434e-05,
      "loss": 1.0771,
      "num_input_tokens_seen": 12746888,
      "step": 23005
    },
    {
      "epoch": 403.69026548672565,
      "grad_norm": 10.0625,
      "learning_rate": 1.9146684168175184e-05,
      "loss": 1.0778,
      "num_input_tokens_seen": 12749976,
      "step": 23010
    },
    {
      "epoch": 403.7787610619469,
      "grad_norm": 13.5625,
      "learning_rate": 1.9137140022771796e-05,
      "loss": 1.1744,
      "num_input_tokens_seen": 12752728,
      "step": 23015
    },
    {
      "epoch": 403.86725663716817,
      "grad_norm": 8.9375,
      "learning_rate": 1.9127596781495103e-05,
      "loss": 0.9909,
      "num_input_tokens_seen": 12755432,
      "step": 23020
    },
    {
      "epoch": 403.95575221238937,
      "grad_norm": 10.0625,
      "learning_rate": 1.9118054445816767e-05,
      "loss": 1.0646,
      "num_input_tokens_seen": 12758728,
      "step": 23025
    },
    {
      "epoch": 404.0353982300885,
      "grad_norm": 5.625,
      "learning_rate": 1.9108513017208356e-05,
      "loss": 0.9953,
      "num_input_tokens_seen": 12761264,
      "step": 23030
    },
    {
      "epoch": 404.12389380530976,
      "grad_norm": 8.25,
      "learning_rate": 1.9098972497141287e-05,
      "loss": 1.0603,
      "num_input_tokens_seen": 12763920,
      "step": 23035
    },
    {
      "epoch": 404.21238938053096,
      "grad_norm": 9.25,
      "learning_rate": 1.9089432887086806e-05,
      "loss": 1.0345,
      "num_input_tokens_seen": 12766528,
      "step": 23040
    },
    {
      "epoch": 404.3008849557522,
      "grad_norm": 11.6875,
      "learning_rate": 1.9079894188516056e-05,
      "loss": 1.0873,
      "num_input_tokens_seen": 12769648,
      "step": 23045
    },
    {
      "epoch": 404.3893805309734,
      "grad_norm": 7.46875,
      "learning_rate": 1.907035640290002e-05,
      "loss": 1.024,
      "num_input_tokens_seen": 12772560,
      "step": 23050
    },
    {
      "epoch": 404.4778761061947,
      "grad_norm": 8.125,
      "learning_rate": 1.9060819531709534e-05,
      "loss": 1.1561,
      "num_input_tokens_seen": 12775312,
      "step": 23055
    },
    {
      "epoch": 404.56637168141594,
      "grad_norm": 8.0,
      "learning_rate": 1.9051283576415325e-05,
      "loss": 1.0254,
      "num_input_tokens_seen": 12778128,
      "step": 23060
    },
    {
      "epoch": 404.65486725663715,
      "grad_norm": 6.53125,
      "learning_rate": 1.904174853848793e-05,
      "loss": 1.0663,
      "num_input_tokens_seen": 12781264,
      "step": 23065
    },
    {
      "epoch": 404.7433628318584,
      "grad_norm": 8.0625,
      "learning_rate": 1.903221441939779e-05,
      "loss": 1.1128,
      "num_input_tokens_seen": 12784096,
      "step": 23070
    },
    {
      "epoch": 404.83185840707966,
      "grad_norm": 9.5625,
      "learning_rate": 1.9022681220615194e-05,
      "loss": 1.1143,
      "num_input_tokens_seen": 12786688,
      "step": 23075
    },
    {
      "epoch": 404.92035398230087,
      "grad_norm": 8.0,
      "learning_rate": 1.9013148943610255e-05,
      "loss": 1.0949,
      "num_input_tokens_seen": 12789216,
      "step": 23080
    },
    {
      "epoch": 405.0,
      "grad_norm": 15.4375,
      "learning_rate": 1.9003617589852998e-05,
      "loss": 0.9148,
      "num_input_tokens_seen": 12791376,
      "step": 23085
    },
    {
      "epoch": 405.08849557522126,
      "grad_norm": 8.125,
      "learning_rate": 1.899408716081326e-05,
      "loss": 1.2798,
      "num_input_tokens_seen": 12794112,
      "step": 23090
    },
    {
      "epoch": 405.17699115044246,
      "grad_norm": 8.6875,
      "learning_rate": 1.898455765796075e-05,
      "loss": 1.2414,
      "num_input_tokens_seen": 12796912,
      "step": 23095
    },
    {
      "epoch": 405.2654867256637,
      "grad_norm": 7.71875,
      "learning_rate": 1.8975029082765053e-05,
      "loss": 1.1915,
      "num_input_tokens_seen": 12799536,
      "step": 23100
    },
    {
      "epoch": 405.353982300885,
      "grad_norm": 7.5625,
      "learning_rate": 1.8965501436695577e-05,
      "loss": 0.7965,
      "num_input_tokens_seen": 12802480,
      "step": 23105
    },
    {
      "epoch": 405.4424778761062,
      "grad_norm": 8.1875,
      "learning_rate": 1.895597472122161e-05,
      "loss": 1.0322,
      "num_input_tokens_seen": 12804976,
      "step": 23110
    },
    {
      "epoch": 405.53097345132744,
      "grad_norm": 8.6875,
      "learning_rate": 1.894644893781231e-05,
      "loss": 1.0021,
      "num_input_tokens_seen": 12807968,
      "step": 23115
    },
    {
      "epoch": 405.6194690265487,
      "grad_norm": 8.0,
      "learning_rate": 1.893692408793665e-05,
      "loss": 1.124,
      "num_input_tokens_seen": 12810960,
      "step": 23120
    },
    {
      "epoch": 405.7079646017699,
      "grad_norm": 7.03125,
      "learning_rate": 1.8927400173063493e-05,
      "loss": 0.9966,
      "num_input_tokens_seen": 12814304,
      "step": 23125
    },
    {
      "epoch": 405.79646017699116,
      "grad_norm": 8.25,
      "learning_rate": 1.891787719466154e-05,
      "loss": 1.1332,
      "num_input_tokens_seen": 12817120,
      "step": 23130
    },
    {
      "epoch": 405.88495575221236,
      "grad_norm": 7.3125,
      "learning_rate": 1.8908355154199346e-05,
      "loss": 0.9876,
      "num_input_tokens_seen": 12819984,
      "step": 23135
    },
    {
      "epoch": 405.9734513274336,
      "grad_norm": 10.125,
      "learning_rate": 1.8898834053145357e-05,
      "loss": 0.9176,
      "num_input_tokens_seen": 12822624,
      "step": 23140
    },
    {
      "epoch": 406.05309734513276,
      "grad_norm": 9.6875,
      "learning_rate": 1.8889313892967813e-05,
      "loss": 0.8023,
      "num_input_tokens_seen": 12825384,
      "step": 23145
    },
    {
      "epoch": 406.14159292035396,
      "grad_norm": 9.8125,
      "learning_rate": 1.8879794675134863e-05,
      "loss": 1.1916,
      "num_input_tokens_seen": 12828360,
      "step": 23150
    },
    {
      "epoch": 406.2300884955752,
      "grad_norm": 7.75,
      "learning_rate": 1.8870276401114494e-05,
      "loss": 1.0611,
      "num_input_tokens_seen": 12831256,
      "step": 23155
    },
    {
      "epoch": 406.3185840707965,
      "grad_norm": 10.0,
      "learning_rate": 1.886075907237453e-05,
      "loss": 1.2284,
      "num_input_tokens_seen": 12834312,
      "step": 23160
    },
    {
      "epoch": 406.4070796460177,
      "grad_norm": 9.625,
      "learning_rate": 1.8851242690382672e-05,
      "loss": 1.2107,
      "num_input_tokens_seen": 12836856,
      "step": 23165
    },
    {
      "epoch": 406.49557522123894,
      "grad_norm": 8.0,
      "learning_rate": 1.884172725660645e-05,
      "loss": 1.0471,
      "num_input_tokens_seen": 12839608,
      "step": 23170
    },
    {
      "epoch": 406.5840707964602,
      "grad_norm": 6.59375,
      "learning_rate": 1.8832212772513277e-05,
      "loss": 0.9567,
      "num_input_tokens_seen": 12842248,
      "step": 23175
    },
    {
      "epoch": 406.6725663716814,
      "grad_norm": 6.3125,
      "learning_rate": 1.8822699239570414e-05,
      "loss": 1.0737,
      "num_input_tokens_seen": 12845384,
      "step": 23180
    },
    {
      "epoch": 406.76106194690266,
      "grad_norm": 8.6875,
      "learning_rate": 1.8813186659244943e-05,
      "loss": 1.0156,
      "num_input_tokens_seen": 12847704,
      "step": 23185
    },
    {
      "epoch": 406.8495575221239,
      "grad_norm": 7.8125,
      "learning_rate": 1.880367503300385e-05,
      "loss": 0.992,
      "num_input_tokens_seen": 12850472,
      "step": 23190
    },
    {
      "epoch": 406.9380530973451,
      "grad_norm": 6.65625,
      "learning_rate": 1.8794164362313927e-05,
      "loss": 0.9936,
      "num_input_tokens_seen": 12853240,
      "step": 23195
    },
    {
      "epoch": 407.01769911504425,
      "grad_norm": 7.875,
      "learning_rate": 1.878465464864185e-05,
      "loss": 0.6801,
      "num_input_tokens_seen": 12855568,
      "step": 23200
    },
    {
      "epoch": 407.01769911504425,
      "eval_loss": 1.0216825008392334,
      "eval_runtime": 0.9404,
      "eval_samples_per_second": 26.584,
      "eval_steps_per_second": 13.823,
      "num_input_tokens_seen": 12855568,
      "step": 23200
    },
    {
      "epoch": 407.1061946902655,
      "grad_norm": 7.5625,
      "learning_rate": 1.877514589345414e-05,
      "loss": 1.0588,
      "num_input_tokens_seen": 12858496,
      "step": 23205
    },
    {
      "epoch": 407.1946902654867,
      "grad_norm": 10.0625,
      "learning_rate": 1.876563809821715e-05,
      "loss": 1.0627,
      "num_input_tokens_seen": 12861200,
      "step": 23210
    },
    {
      "epoch": 407.283185840708,
      "grad_norm": 8.0625,
      "learning_rate": 1.8756131264397106e-05,
      "loss": 1.0303,
      "num_input_tokens_seen": 12864144,
      "step": 23215
    },
    {
      "epoch": 407.37168141592923,
      "grad_norm": 6.84375,
      "learning_rate": 1.87466253934601e-05,
      "loss": 0.8663,
      "num_input_tokens_seen": 12867088,
      "step": 23220
    },
    {
      "epoch": 407.46017699115043,
      "grad_norm": 8.5,
      "learning_rate": 1.8737120486872033e-05,
      "loss": 1.1712,
      "num_input_tokens_seen": 12869968,
      "step": 23225
    },
    {
      "epoch": 407.5486725663717,
      "grad_norm": 6.4375,
      "learning_rate": 1.8727616546098696e-05,
      "loss": 0.8897,
      "num_input_tokens_seen": 12872944,
      "step": 23230
    },
    {
      "epoch": 407.6371681415929,
      "grad_norm": 6.96875,
      "learning_rate": 1.8718113572605716e-05,
      "loss": 0.9099,
      "num_input_tokens_seen": 12875808,
      "step": 23235
    },
    {
      "epoch": 407.72566371681415,
      "grad_norm": 12.9375,
      "learning_rate": 1.8708611567858554e-05,
      "loss": 1.1422,
      "num_input_tokens_seen": 12878400,
      "step": 23240
    },
    {
      "epoch": 407.8141592920354,
      "grad_norm": 9.0,
      "learning_rate": 1.8699110533322565e-05,
      "loss": 1.1814,
      "num_input_tokens_seen": 12881200,
      "step": 23245
    },
    {
      "epoch": 407.9026548672566,
      "grad_norm": 8.5625,
      "learning_rate": 1.8689610470462897e-05,
      "loss": 1.0979,
      "num_input_tokens_seen": 12884112,
      "step": 23250
    },
    {
      "epoch": 407.9911504424779,
      "grad_norm": 7.71875,
      "learning_rate": 1.8680111380744604e-05,
      "loss": 1.1649,
      "num_input_tokens_seen": 12886784,
      "step": 23255
    },
    {
      "epoch": 408.070796460177,
      "grad_norm": 8.125,
      "learning_rate": 1.8670613265632564e-05,
      "loss": 1.377,
      "num_input_tokens_seen": 12888952,
      "step": 23260
    },
    {
      "epoch": 408.1592920353982,
      "grad_norm": 7.1875,
      "learning_rate": 1.866111612659149e-05,
      "loss": 1.079,
      "num_input_tokens_seen": 12892008,
      "step": 23265
    },
    {
      "epoch": 408.24778761061947,
      "grad_norm": 6.4375,
      "learning_rate": 1.8651619965085967e-05,
      "loss": 0.9348,
      "num_input_tokens_seen": 12894504,
      "step": 23270
    },
    {
      "epoch": 408.3362831858407,
      "grad_norm": 6.8125,
      "learning_rate": 1.8642124782580433e-05,
      "loss": 0.9228,
      "num_input_tokens_seen": 12897656,
      "step": 23275
    },
    {
      "epoch": 408.42477876106193,
      "grad_norm": 8.5625,
      "learning_rate": 1.8632630580539144e-05,
      "loss": 1.1241,
      "num_input_tokens_seen": 12900472,
      "step": 23280
    },
    {
      "epoch": 408.5132743362832,
      "grad_norm": 12.4375,
      "learning_rate": 1.862313736042625e-05,
      "loss": 1.1952,
      "num_input_tokens_seen": 12903288,
      "step": 23285
    },
    {
      "epoch": 408.60176991150445,
      "grad_norm": 6.65625,
      "learning_rate": 1.8613645123705703e-05,
      "loss": 0.9328,
      "num_input_tokens_seen": 12906072,
      "step": 23290
    },
    {
      "epoch": 408.69026548672565,
      "grad_norm": 8.125,
      "learning_rate": 1.8604153871841328e-05,
      "loss": 1.0674,
      "num_input_tokens_seen": 12908856,
      "step": 23295
    },
    {
      "epoch": 408.7787610619469,
      "grad_norm": 8.125,
      "learning_rate": 1.859466360629682e-05,
      "loss": 0.9223,
      "num_input_tokens_seen": 12911400,
      "step": 23300
    },
    {
      "epoch": 408.86725663716817,
      "grad_norm": 8.5625,
      "learning_rate": 1.8585174328535666e-05,
      "loss": 0.8375,
      "num_input_tokens_seen": 12914376,
      "step": 23305
    },
    {
      "epoch": 408.95575221238937,
      "grad_norm": 8.875,
      "learning_rate": 1.857568604002124e-05,
      "loss": 1.3411,
      "num_input_tokens_seen": 12917160,
      "step": 23310
    },
    {
      "epoch": 409.0353982300885,
      "grad_norm": 7.625,
      "learning_rate": 1.8566198742216774e-05,
      "loss": 1.2803,
      "num_input_tokens_seen": 12919384,
      "step": 23315
    },
    {
      "epoch": 409.12389380530976,
      "grad_norm": 7.6875,
      "learning_rate": 1.85567124365853e-05,
      "loss": 0.894,
      "num_input_tokens_seen": 12922232,
      "step": 23320
    },
    {
      "epoch": 409.21238938053096,
      "grad_norm": 9.25,
      "learning_rate": 1.854722712458975e-05,
      "loss": 0.9688,
      "num_input_tokens_seen": 12925160,
      "step": 23325
    },
    {
      "epoch": 409.3008849557522,
      "grad_norm": 8.3125,
      "learning_rate": 1.853774280769286e-05,
      "loss": 1.0523,
      "num_input_tokens_seen": 12927864,
      "step": 23330
    },
    {
      "epoch": 409.3893805309734,
      "grad_norm": 10.125,
      "learning_rate": 1.852825948735724e-05,
      "loss": 1.0018,
      "num_input_tokens_seen": 12930456,
      "step": 23335
    },
    {
      "epoch": 409.4778761061947,
      "grad_norm": 9.4375,
      "learning_rate": 1.851877716504534e-05,
      "loss": 1.1586,
      "num_input_tokens_seen": 12933240,
      "step": 23340
    },
    {
      "epoch": 409.56637168141594,
      "grad_norm": 8.0625,
      "learning_rate": 1.8509295842219448e-05,
      "loss": 1.0387,
      "num_input_tokens_seen": 12936584,
      "step": 23345
    },
    {
      "epoch": 409.65486725663715,
      "grad_norm": 8.9375,
      "learning_rate": 1.8499815520341697e-05,
      "loss": 1.1277,
      "num_input_tokens_seen": 12939784,
      "step": 23350
    },
    {
      "epoch": 409.7433628318584,
      "grad_norm": 9.25,
      "learning_rate": 1.8490336200874094e-05,
      "loss": 1.2097,
      "num_input_tokens_seen": 12942424,
      "step": 23355
    },
    {
      "epoch": 409.83185840707966,
      "grad_norm": 7.65625,
      "learning_rate": 1.848085788527844e-05,
      "loss": 1.0245,
      "num_input_tokens_seen": 12944760,
      "step": 23360
    },
    {
      "epoch": 409.92035398230087,
      "grad_norm": 8.875,
      "learning_rate": 1.847138057501644e-05,
      "loss": 1.2438,
      "num_input_tokens_seen": 12947576,
      "step": 23365
    },
    {
      "epoch": 410.0,
      "grad_norm": 4.15625,
      "learning_rate": 1.8461904271549582e-05,
      "loss": 0.7008,
      "num_input_tokens_seen": 12950032,
      "step": 23370
    },
    {
      "epoch": 410.08849557522126,
      "grad_norm": 7.96875,
      "learning_rate": 1.845242897633926e-05,
      "loss": 1.1688,
      "num_input_tokens_seen": 12952896,
      "step": 23375
    },
    {
      "epoch": 410.17699115044246,
      "grad_norm": 7.25,
      "learning_rate": 1.844295469084667e-05,
      "loss": 1.035,
      "num_input_tokens_seen": 12955808,
      "step": 23380
    },
    {
      "epoch": 410.2654867256637,
      "grad_norm": 7.625,
      "learning_rate": 1.843348141653286e-05,
      "loss": 1.1288,
      "num_input_tokens_seen": 12958400,
      "step": 23385
    },
    {
      "epoch": 410.353982300885,
      "grad_norm": 11.375,
      "learning_rate": 1.842400915485874e-05,
      "loss": 0.9646,
      "num_input_tokens_seen": 12961040,
      "step": 23390
    },
    {
      "epoch": 410.4424778761062,
      "grad_norm": 7.96875,
      "learning_rate": 1.8414537907285053e-05,
      "loss": 0.825,
      "num_input_tokens_seen": 12963904,
      "step": 23395
    },
    {
      "epoch": 410.53097345132744,
      "grad_norm": 8.3125,
      "learning_rate": 1.840506767527237e-05,
      "loss": 1.2031,
      "num_input_tokens_seen": 12966544,
      "step": 23400
    },
    {
      "epoch": 410.53097345132744,
      "eval_loss": 1.0216825008392334,
      "eval_runtime": 0.9365,
      "eval_samples_per_second": 26.694,
      "eval_steps_per_second": 13.881,
      "num_input_tokens_seen": 12966544,
      "step": 23400
    },
    {
      "epoch": 410.6194690265487,
      "grad_norm": 10.0,
      "learning_rate": 1.8395598460281137e-05,
      "loss": 1.008,
      "num_input_tokens_seen": 12969328,
      "step": 23405
    },
    {
      "epoch": 410.7079646017699,
      "grad_norm": 9.4375,
      "learning_rate": 1.838613026377161e-05,
      "loss": 0.9995,
      "num_input_tokens_seen": 12972368,
      "step": 23410
    },
    {
      "epoch": 410.79646017699116,
      "grad_norm": 8.0625,
      "learning_rate": 1.8376663087203917e-05,
      "loss": 1.0,
      "num_input_tokens_seen": 12975200,
      "step": 23415
    },
    {
      "epoch": 410.88495575221236,
      "grad_norm": 8.625,
      "learning_rate": 1.8367196932038014e-05,
      "loss": 1.0666,
      "num_input_tokens_seen": 12978464,
      "step": 23420
    },
    {
      "epoch": 410.9734513274336,
      "grad_norm": 7.875,
      "learning_rate": 1.8357731799733686e-05,
      "loss": 1.129,
      "num_input_tokens_seen": 12980976,
      "step": 23425
    },
    {
      "epoch": 411.05309734513276,
      "grad_norm": 11.625,
      "learning_rate": 1.8348267691750586e-05,
      "loss": 1.0929,
      "num_input_tokens_seen": 12983512,
      "step": 23430
    },
    {
      "epoch": 411.14159292035396,
      "grad_norm": 7.09375,
      "learning_rate": 1.833880460954821e-05,
      "loss": 1.056,
      "num_input_tokens_seen": 12986504,
      "step": 23435
    },
    {
      "epoch": 411.2300884955752,
      "grad_norm": 8.3125,
      "learning_rate": 1.8329342554585866e-05,
      "loss": 1.1567,
      "num_input_tokens_seen": 12989160,
      "step": 23440
    },
    {
      "epoch": 411.3185840707965,
      "grad_norm": 11.5625,
      "learning_rate": 1.8319881528322735e-05,
      "loss": 1.1351,
      "num_input_tokens_seen": 12992360,
      "step": 23445
    },
    {
      "epoch": 411.4070796460177,
      "grad_norm": 8.5625,
      "learning_rate": 1.8310421532217815e-05,
      "loss": 1.0906,
      "num_input_tokens_seen": 12995448,
      "step": 23450
    },
    {
      "epoch": 411.49557522123894,
      "grad_norm": 7.9375,
      "learning_rate": 1.8300962567729958e-05,
      "loss": 1.0709,
      "num_input_tokens_seen": 12997960,
      "step": 23455
    },
    {
      "epoch": 411.5840707964602,
      "grad_norm": 7.78125,
      "learning_rate": 1.8291504636317866e-05,
      "loss": 0.8292,
      "num_input_tokens_seen": 13000232,
      "step": 23460
    },
    {
      "epoch": 411.6725663716814,
      "grad_norm": 7.875,
      "learning_rate": 1.8282047739440055e-05,
      "loss": 0.9875,
      "num_input_tokens_seen": 13002936,
      "step": 23465
    },
    {
      "epoch": 411.76106194690266,
      "grad_norm": 8.6875,
      "learning_rate": 1.8272591878554903e-05,
      "loss": 1.085,
      "num_input_tokens_seen": 13005640,
      "step": 23470
    },
    {
      "epoch": 411.8495575221239,
      "grad_norm": 8.3125,
      "learning_rate": 1.8263137055120638e-05,
      "loss": 1.0105,
      "num_input_tokens_seen": 13008392,
      "step": 23475
    },
    {
      "epoch": 411.9380530973451,
      "grad_norm": 6.78125,
      "learning_rate": 1.8253683270595295e-05,
      "loss": 0.8713,
      "num_input_tokens_seen": 13011384,
      "step": 23480
    },
    {
      "epoch": 412.01769911504425,
      "grad_norm": 8.875,
      "learning_rate": 1.824423052643677e-05,
      "loss": 1.2259,
      "num_input_tokens_seen": 13013784,
      "step": 23485
    },
    {
      "epoch": 412.1061946902655,
      "grad_norm": 8.25,
      "learning_rate": 1.82347788241028e-05,
      "loss": 1.0081,
      "num_input_tokens_seen": 13016744,
      "step": 23490
    },
    {
      "epoch": 412.1946902654867,
      "grad_norm": 7.40625,
      "learning_rate": 1.8225328165050942e-05,
      "loss": 1.1681,
      "num_input_tokens_seen": 13019512,
      "step": 23495
    },
    {
      "epoch": 412.283185840708,
      "grad_norm": 8.5,
      "learning_rate": 1.821587855073863e-05,
      "loss": 0.9774,
      "num_input_tokens_seen": 13021800,
      "step": 23500
    },
    {
      "epoch": 412.37168141592923,
      "grad_norm": 8.6875,
      "learning_rate": 1.8206429982623086e-05,
      "loss": 1.0373,
      "num_input_tokens_seen": 13025288,
      "step": 23505
    },
    {
      "epoch": 412.46017699115043,
      "grad_norm": 9.8125,
      "learning_rate": 1.8196982462161416e-05,
      "loss": 1.0505,
      "num_input_tokens_seen": 13028008,
      "step": 23510
    },
    {
      "epoch": 412.5486725663717,
      "grad_norm": 6.625,
      "learning_rate": 1.818753599081055e-05,
      "loss": 1.1928,
      "num_input_tokens_seen": 13030344,
      "step": 23515
    },
    {
      "epoch": 412.6371681415929,
      "grad_norm": 7.0625,
      "learning_rate": 1.817809057002724e-05,
      "loss": 0.9688,
      "num_input_tokens_seen": 13033192,
      "step": 23520
    },
    {
      "epoch": 412.72566371681415,
      "grad_norm": 8.875,
      "learning_rate": 1.8168646201268096e-05,
      "loss": 1.1069,
      "num_input_tokens_seen": 13035576,
      "step": 23525
    },
    {
      "epoch": 412.8141592920354,
      "grad_norm": 10.3125,
      "learning_rate": 1.8159202885989557e-05,
      "loss": 1.0582,
      "num_input_tokens_seen": 13038824,
      "step": 23530
    },
    {
      "epoch": 412.9026548672566,
      "grad_norm": 11.0,
      "learning_rate": 1.814976062564789e-05,
      "loss": 1.0909,
      "num_input_tokens_seen": 13041448,
      "step": 23535
    },
    {
      "epoch": 412.9911504424779,
      "grad_norm": 8.6875,
      "learning_rate": 1.8140319421699234e-05,
      "loss": 0.9412,
      "num_input_tokens_seen": 13044376,
      "step": 23540
    },
    {
      "epoch": 413.070796460177,
      "grad_norm": 13.0625,
      "learning_rate": 1.8130879275599515e-05,
      "loss": 1.0274,
      "num_input_tokens_seen": 13046664,
      "step": 23545
    },
    {
      "epoch": 413.1592920353982,
      "grad_norm": 6.5625,
      "learning_rate": 1.8121440188804544e-05,
      "loss": 0.9343,
      "num_input_tokens_seen": 13049544,
      "step": 23550
    },
    {
      "epoch": 413.24778761061947,
      "grad_norm": 8.5625,
      "learning_rate": 1.811200216276993e-05,
      "loss": 1.0819,
      "num_input_tokens_seen": 13052456,
      "step": 23555
    },
    {
      "epoch": 413.3362831858407,
      "grad_norm": 7.78125,
      "learning_rate": 1.810256519895115e-05,
      "loss": 0.9427,
      "num_input_tokens_seen": 13054968,
      "step": 23560
    },
    {
      "epoch": 413.42477876106193,
      "grad_norm": 12.0625,
      "learning_rate": 1.8093129298803494e-05,
      "loss": 1.3116,
      "num_input_tokens_seen": 13057608,
      "step": 23565
    },
    {
      "epoch": 413.5132743362832,
      "grad_norm": 6.40625,
      "learning_rate": 1.808369446378209e-05,
      "loss": 1.0593,
      "num_input_tokens_seen": 13060504,
      "step": 23570
    },
    {
      "epoch": 413.60176991150445,
      "grad_norm": 14.4375,
      "learning_rate": 1.8074260695341914e-05,
      "loss": 1.0611,
      "num_input_tokens_seen": 13063464,
      "step": 23575
    },
    {
      "epoch": 413.69026548672565,
      "grad_norm": 8.75,
      "learning_rate": 1.8064827994937782e-05,
      "loss": 1.2958,
      "num_input_tokens_seen": 13066472,
      "step": 23580
    },
    {
      "epoch": 413.7787610619469,
      "grad_norm": 9.125,
      "learning_rate": 1.8055396364024317e-05,
      "loss": 0.9589,
      "num_input_tokens_seen": 13069208,
      "step": 23585
    },
    {
      "epoch": 413.86725663716817,
      "grad_norm": 8.1875,
      "learning_rate": 1.804596580405601e-05,
      "loss": 0.9995,
      "num_input_tokens_seen": 13072072,
      "step": 23590
    },
    {
      "epoch": 413.95575221238937,
      "grad_norm": 5.78125,
      "learning_rate": 1.8036536316487174e-05,
      "loss": 1.0102,
      "num_input_tokens_seen": 13075048,
      "step": 23595
    },
    {
      "epoch": 414.0353982300885,
      "grad_norm": 8.3125,
      "learning_rate": 1.802710790277193e-05,
      "loss": 0.9253,
      "num_input_tokens_seen": 13077752,
      "step": 23600
    },
    {
      "epoch": 414.0353982300885,
      "eval_loss": 1.0303553342819214,
      "eval_runtime": 0.9334,
      "eval_samples_per_second": 26.783,
      "eval_steps_per_second": 13.927,
      "num_input_tokens_seen": 13077752,
      "step": 23600
    },
    {
      "epoch": 414.12389380530976,
      "grad_norm": 7.1875,
      "learning_rate": 1.801768056436429e-05,
      "loss": 0.9414,
      "num_input_tokens_seen": 13080504,
      "step": 23605
    },
    {
      "epoch": 414.21238938053096,
      "grad_norm": 7.46875,
      "learning_rate": 1.8008254302718035e-05,
      "loss": 1.0197,
      "num_input_tokens_seen": 13082904,
      "step": 23610
    },
    {
      "epoch": 414.3008849557522,
      "grad_norm": 6.65625,
      "learning_rate": 1.7998829119286837e-05,
      "loss": 0.9185,
      "num_input_tokens_seen": 13085448,
      "step": 23615
    },
    {
      "epoch": 414.3893805309734,
      "grad_norm": 8.3125,
      "learning_rate": 1.798940501552418e-05,
      "loss": 1.1933,
      "num_input_tokens_seen": 13087592,
      "step": 23620
    },
    {
      "epoch": 414.4778761061947,
      "grad_norm": 8.8125,
      "learning_rate": 1.797998199288336e-05,
      "loss": 1.1475,
      "num_input_tokens_seen": 13090616,
      "step": 23625
    },
    {
      "epoch": 414.56637168141594,
      "grad_norm": 12.5625,
      "learning_rate": 1.7970560052817543e-05,
      "loss": 1.1703,
      "num_input_tokens_seen": 13093448,
      "step": 23630
    },
    {
      "epoch": 414.65486725663715,
      "grad_norm": 8.9375,
      "learning_rate": 1.7961139196779702e-05,
      "loss": 1.0124,
      "num_input_tokens_seen": 13096424,
      "step": 23635
    },
    {
      "epoch": 414.7433628318584,
      "grad_norm": 13.5625,
      "learning_rate": 1.7951719426222647e-05,
      "loss": 1.0618,
      "num_input_tokens_seen": 13099848,
      "step": 23640
    },
    {
      "epoch": 414.83185840707966,
      "grad_norm": 7.5625,
      "learning_rate": 1.794230074259904e-05,
      "loss": 1.1282,
      "num_input_tokens_seen": 13102648,
      "step": 23645
    },
    {
      "epoch": 414.92035398230087,
      "grad_norm": 9.0625,
      "learning_rate": 1.7932883147361336e-05,
      "loss": 1.0789,
      "num_input_tokens_seen": 13105784,
      "step": 23650
    },
    {
      "epoch": 415.0,
      "grad_norm": 4.96875,
      "learning_rate": 1.7923466641961865e-05,
      "loss": 0.8707,
      "num_input_tokens_seen": 13108280,
      "step": 23655
    },
    {
      "epoch": 415.08849557522126,
      "grad_norm": 8.0625,
      "learning_rate": 1.791405122785278e-05,
      "loss": 1.0394,
      "num_input_tokens_seen": 13110856,
      "step": 23660
    },
    {
      "epoch": 415.17699115044246,
      "grad_norm": 9.0,
      "learning_rate": 1.7904636906486037e-05,
      "loss": 1.125,
      "num_input_tokens_seen": 13113608,
      "step": 23665
    },
    {
      "epoch": 415.2654867256637,
      "grad_norm": 14.75,
      "learning_rate": 1.7895223679313448e-05,
      "loss": 1.092,
      "num_input_tokens_seen": 13115992,
      "step": 23670
    },
    {
      "epoch": 415.353982300885,
      "grad_norm": 8.1875,
      "learning_rate": 1.7885811547786653e-05,
      "loss": 0.9849,
      "num_input_tokens_seen": 13118840,
      "step": 23675
    },
    {
      "epoch": 415.4424778761062,
      "grad_norm": 7.28125,
      "learning_rate": 1.7876400513357115e-05,
      "loss": 1.0221,
      "num_input_tokens_seen": 13122296,
      "step": 23680
    },
    {
      "epoch": 415.53097345132744,
      "grad_norm": 7.5625,
      "learning_rate": 1.7866990577476146e-05,
      "loss": 1.032,
      "num_input_tokens_seen": 13125368,
      "step": 23685
    },
    {
      "epoch": 415.6194690265487,
      "grad_norm": 7.96875,
      "learning_rate": 1.7857581741594863e-05,
      "loss": 0.9649,
      "num_input_tokens_seen": 13128456,
      "step": 23690
    },
    {
      "epoch": 415.7079646017699,
      "grad_norm": 9.875,
      "learning_rate": 1.7848174007164237e-05,
      "loss": 1.1567,
      "num_input_tokens_seen": 13131608,
      "step": 23695
    },
    {
      "epoch": 415.79646017699116,
      "grad_norm": 8.1875,
      "learning_rate": 1.7838767375635052e-05,
      "loss": 1.1944,
      "num_input_tokens_seen": 13134264,
      "step": 23700
    },
    {
      "epoch": 415.88495575221236,
      "grad_norm": 6.875,
      "learning_rate": 1.782936184845793e-05,
      "loss": 0.8093,
      "num_input_tokens_seen": 13136856,
      "step": 23705
    },
    {
      "epoch": 415.9734513274336,
      "grad_norm": 8.875,
      "learning_rate": 1.7819957427083334e-05,
      "loss": 1.1125,
      "num_input_tokens_seen": 13139448,
      "step": 23710
    },
    {
      "epoch": 416.05309734513276,
      "grad_norm": 9.0,
      "learning_rate": 1.7810554112961516e-05,
      "loss": 1.1934,
      "num_input_tokens_seen": 13142240,
      "step": 23715
    },
    {
      "epoch": 416.14159292035396,
      "grad_norm": 11.875,
      "learning_rate": 1.7801151907542607e-05,
      "loss": 1.1143,
      "num_input_tokens_seen": 13144720,
      "step": 23720
    },
    {
      "epoch": 416.2300884955752,
      "grad_norm": 9.5625,
      "learning_rate": 1.7791750812276547e-05,
      "loss": 1.0774,
      "num_input_tokens_seen": 13147408,
      "step": 23725
    },
    {
      "epoch": 416.3185840707965,
      "grad_norm": 8.3125,
      "learning_rate": 1.778235082861309e-05,
      "loss": 0.9696,
      "num_input_tokens_seen": 13150336,
      "step": 23730
    },
    {
      "epoch": 416.4070796460177,
      "grad_norm": 8.1875,
      "learning_rate": 1.777295195800184e-05,
      "loss": 0.948,
      "num_input_tokens_seen": 13153280,
      "step": 23735
    },
    {
      "epoch": 416.49557522123894,
      "grad_norm": 7.53125,
      "learning_rate": 1.7763554201892215e-05,
      "loss": 0.9833,
      "num_input_tokens_seen": 13156288,
      "step": 23740
    },
    {
      "epoch": 416.5840707964602,
      "grad_norm": 6.9375,
      "learning_rate": 1.7754157561733476e-05,
      "loss": 1.0197,
      "num_input_tokens_seen": 13159568,
      "step": 23745
    },
    {
      "epoch": 416.6725663716814,
      "grad_norm": 6.96875,
      "learning_rate": 1.7744762038974702e-05,
      "loss": 1.0842,
      "num_input_tokens_seen": 13162384,
      "step": 23750
    },
    {
      "epoch": 416.76106194690266,
      "grad_norm": 8.4375,
      "learning_rate": 1.7735367635064788e-05,
      "loss": 1.2747,
      "num_input_tokens_seen": 13164816,
      "step": 23755
    },
    {
      "epoch": 416.8495575221239,
      "grad_norm": 9.875,
      "learning_rate": 1.7725974351452474e-05,
      "loss": 1.1133,
      "num_input_tokens_seen": 13167488,
      "step": 23760
    },
    {
      "epoch": 416.9380530973451,
      "grad_norm": 7.21875,
      "learning_rate": 1.771658218958634e-05,
      "loss": 0.8672,
      "num_input_tokens_seen": 13170336,
      "step": 23765
    },
    {
      "epoch": 417.01769911504425,
      "grad_norm": 10.0625,
      "learning_rate": 1.770719115091475e-05,
      "loss": 1.1486,
      "num_input_tokens_seen": 13172232,
      "step": 23770
    },
    {
      "epoch": 417.1061946902655,
      "grad_norm": 10.25,
      "learning_rate": 1.7697801236885935e-05,
      "loss": 0.9662,
      "num_input_tokens_seen": 13175336,
      "step": 23775
    },
    {
      "epoch": 417.1946902654867,
      "grad_norm": 6.71875,
      "learning_rate": 1.7688412448947944e-05,
      "loss": 1.1228,
      "num_input_tokens_seen": 13178552,
      "step": 23780
    },
    {
      "epoch": 417.283185840708,
      "grad_norm": 7.8125,
      "learning_rate": 1.767902478854862e-05,
      "loss": 1.088,
      "num_input_tokens_seen": 13181032,
      "step": 23785
    },
    {
      "epoch": 417.37168141592923,
      "grad_norm": 7.78125,
      "learning_rate": 1.766963825713569e-05,
      "loss": 0.9185,
      "num_input_tokens_seen": 13183656,
      "step": 23790
    },
    {
      "epoch": 417.46017699115043,
      "grad_norm": 10.8125,
      "learning_rate": 1.766025285615665e-05,
      "loss": 1.1763,
      "num_input_tokens_seen": 13187064,
      "step": 23795
    },
    {
      "epoch": 417.5486725663717,
      "grad_norm": 7.8125,
      "learning_rate": 1.7650868587058854e-05,
      "loss": 1.0901,
      "num_input_tokens_seen": 13189592,
      "step": 23800
    },
    {
      "epoch": 417.5486725663717,
      "eval_loss": 1.0249234437942505,
      "eval_runtime": 0.9398,
      "eval_samples_per_second": 26.601,
      "eval_steps_per_second": 13.833,
      "num_input_tokens_seen": 13189592,
      "step": 23800
    },
    {
      "epoch": 417.6371681415929,
      "grad_norm": 10.25,
      "learning_rate": 1.7641485451289484e-05,
      "loss": 1.1686,
      "num_input_tokens_seen": 13192296,
      "step": 23805
    },
    {
      "epoch": 417.72566371681415,
      "grad_norm": 8.4375,
      "learning_rate": 1.7632103450295534e-05,
      "loss": 1.0008,
      "num_input_tokens_seen": 13194696,
      "step": 23810
    },
    {
      "epoch": 417.8141592920354,
      "grad_norm": 8.625,
      "learning_rate": 1.762272258552381e-05,
      "loss": 1.03,
      "num_input_tokens_seen": 13197544,
      "step": 23815
    },
    {
      "epoch": 417.9026548672566,
      "grad_norm": 8.1875,
      "learning_rate": 1.7613342858420988e-05,
      "loss": 1.026,
      "num_input_tokens_seen": 13200712,
      "step": 23820
    },
    {
      "epoch": 417.9911504424779,
      "grad_norm": 6.6875,
      "learning_rate": 1.760396427043351e-05,
      "loss": 0.9232,
      "num_input_tokens_seen": 13203048,
      "step": 23825
    },
    {
      "epoch": 418.070796460177,
      "grad_norm": 7.34375,
      "learning_rate": 1.7594586823007696e-05,
      "loss": 0.9111,
      "num_input_tokens_seen": 13205808,
      "step": 23830
    },
    {
      "epoch": 418.1592920353982,
      "grad_norm": 5.9375,
      "learning_rate": 1.7585210517589646e-05,
      "loss": 0.7975,
      "num_input_tokens_seen": 13208672,
      "step": 23835
    },
    {
      "epoch": 418.24778761061947,
      "grad_norm": 9.875,
      "learning_rate": 1.7575835355625314e-05,
      "loss": 1.0207,
      "num_input_tokens_seen": 13211392,
      "step": 23840
    },
    {
      "epoch": 418.3362831858407,
      "grad_norm": 8.0625,
      "learning_rate": 1.756646133856048e-05,
      "loss": 1.3275,
      "num_input_tokens_seen": 13214240,
      "step": 23845
    },
    {
      "epoch": 418.42477876106193,
      "grad_norm": 6.9375,
      "learning_rate": 1.7557088467840714e-05,
      "loss": 1.0552,
      "num_input_tokens_seen": 13216896,
      "step": 23850
    },
    {
      "epoch": 418.5132743362832,
      "grad_norm": 8.25,
      "learning_rate": 1.7547716744911438e-05,
      "loss": 1.0631,
      "num_input_tokens_seen": 13219488,
      "step": 23855
    },
    {
      "epoch": 418.60176991150445,
      "grad_norm": 6.875,
      "learning_rate": 1.7538346171217902e-05,
      "loss": 1.0125,
      "num_input_tokens_seen": 13221872,
      "step": 23860
    },
    {
      "epoch": 418.69026548672565,
      "grad_norm": 7.03125,
      "learning_rate": 1.7528976748205146e-05,
      "loss": 1.1375,
      "num_input_tokens_seen": 13224528,
      "step": 23865
    },
    {
      "epoch": 418.7787610619469,
      "grad_norm": 9.875,
      "learning_rate": 1.751960847731807e-05,
      "loss": 1.0829,
      "num_input_tokens_seen": 13227296,
      "step": 23870
    },
    {
      "epoch": 418.86725663716817,
      "grad_norm": 7.5,
      "learning_rate": 1.7510241360001362e-05,
      "loss": 1.0866,
      "num_input_tokens_seen": 13231008,
      "step": 23875
    },
    {
      "epoch": 418.95575221238937,
      "grad_norm": 10.75,
      "learning_rate": 1.7500875397699562e-05,
      "loss": 1.1327,
      "num_input_tokens_seen": 13233840,
      "step": 23880
    },
    {
      "epoch": 419.0353982300885,
      "grad_norm": 9.4375,
      "learning_rate": 1.7491510591857015e-05,
      "loss": 1.0257,
      "num_input_tokens_seen": 13236488,
      "step": 23885
    },
    {
      "epoch": 419.12389380530976,
      "grad_norm": 11.75,
      "learning_rate": 1.7482146943917896e-05,
      "loss": 0.9493,
      "num_input_tokens_seen": 13239096,
      "step": 23890
    },
    {
      "epoch": 419.21238938053096,
      "grad_norm": 7.1875,
      "learning_rate": 1.7472784455326185e-05,
      "loss": 0.9143,
      "num_input_tokens_seen": 13241944,
      "step": 23895
    },
    {
      "epoch": 419.3008849557522,
      "grad_norm": 7.15625,
      "learning_rate": 1.746342312752572e-05,
      "loss": 1.173,
      "num_input_tokens_seen": 13244680,
      "step": 23900
    },
    {
      "epoch": 419.3893805309734,
      "grad_norm": 9.0,
      "learning_rate": 1.74540629619601e-05,
      "loss": 1.1488,
      "num_input_tokens_seen": 13247624,
      "step": 23905
    },
    {
      "epoch": 419.4778761061947,
      "grad_norm": 7.8125,
      "learning_rate": 1.7444703960072815e-05,
      "loss": 0.9799,
      "num_input_tokens_seen": 13250536,
      "step": 23910
    },
    {
      "epoch": 419.56637168141594,
      "grad_norm": 7.8125,
      "learning_rate": 1.7435346123307118e-05,
      "loss": 1.1401,
      "num_input_tokens_seen": 13253512,
      "step": 23915
    },
    {
      "epoch": 419.65486725663715,
      "grad_norm": 8.4375,
      "learning_rate": 1.742598945310611e-05,
      "loss": 1.0041,
      "num_input_tokens_seen": 13256152,
      "step": 23920
    },
    {
      "epoch": 419.7433628318584,
      "grad_norm": 7.28125,
      "learning_rate": 1.741663395091272e-05,
      "loss": 1.0759,
      "num_input_tokens_seen": 13258744,
      "step": 23925
    },
    {
      "epoch": 419.83185840707966,
      "grad_norm": 7.90625,
      "learning_rate": 1.7407279618169657e-05,
      "loss": 1.1379,
      "num_input_tokens_seen": 13261464,
      "step": 23930
    },
    {
      "epoch": 419.92035398230087,
      "grad_norm": 9.5,
      "learning_rate": 1.73979264563195e-05,
      "loss": 1.0383,
      "num_input_tokens_seen": 13264296,
      "step": 23935
    },
    {
      "epoch": 420.0,
      "grad_norm": 18.75,
      "learning_rate": 1.7388574466804625e-05,
      "loss": 1.1391,
      "num_input_tokens_seen": 13266712,
      "step": 23940
    },
    {
      "epoch": 420.08849557522126,
      "grad_norm": 9.0625,
      "learning_rate": 1.7379223651067207e-05,
      "loss": 1.1454,
      "num_input_tokens_seen": 13269272,
      "step": 23945
    },
    {
      "epoch": 420.17699115044246,
      "grad_norm": 7.3125,
      "learning_rate": 1.736987401054928e-05,
      "loss": 1.0262,
      "num_input_tokens_seen": 13272104,
      "step": 23950
    },
    {
      "epoch": 420.2654867256637,
      "grad_norm": 13.5625,
      "learning_rate": 1.736052554669266e-05,
      "loss": 1.0601,
      "num_input_tokens_seen": 13275208,
      "step": 23955
    },
    {
      "epoch": 420.353982300885,
      "grad_norm": 8.3125,
      "learning_rate": 1.7351178260939007e-05,
      "loss": 0.9875,
      "num_input_tokens_seen": 13278344,
      "step": 23960
    },
    {
      "epoch": 420.4424778761062,
      "grad_norm": 7.03125,
      "learning_rate": 1.7341832154729794e-05,
      "loss": 1.079,
      "num_input_tokens_seen": 13280728,
      "step": 23965
    },
    {
      "epoch": 420.53097345132744,
      "grad_norm": 7.875,
      "learning_rate": 1.7332487229506286e-05,
      "loss": 1.0964,
      "num_input_tokens_seen": 13283736,
      "step": 23970
    },
    {
      "epoch": 420.6194690265487,
      "grad_norm": 10.6875,
      "learning_rate": 1.732314348670961e-05,
      "loss": 1.0464,
      "num_input_tokens_seen": 13286584,
      "step": 23975
    },
    {
      "epoch": 420.7079646017699,
      "grad_norm": 7.40625,
      "learning_rate": 1.7313800927780686e-05,
      "loss": 1.0375,
      "num_input_tokens_seen": 13289608,
      "step": 23980
    },
    {
      "epoch": 420.79646017699116,
      "grad_norm": 9.5,
      "learning_rate": 1.7304459554160245e-05,
      "loss": 1.1584,
      "num_input_tokens_seen": 13292552,
      "step": 23985
    },
    {
      "epoch": 420.88495575221236,
      "grad_norm": 7.71875,
      "learning_rate": 1.7295119367288853e-05,
      "loss": 0.9817,
      "num_input_tokens_seen": 13294936,
      "step": 23990
    },
    {
      "epoch": 420.9734513274336,
      "grad_norm": 7.5625,
      "learning_rate": 1.728578036860688e-05,
      "loss": 1.0323,
      "num_input_tokens_seen": 13297544,
      "step": 23995
    },
    {
      "epoch": 421.05309734513276,
      "grad_norm": 8.1875,
      "learning_rate": 1.7276442559554513e-05,
      "loss": 0.8616,
      "num_input_tokens_seen": 13299920,
      "step": 24000
    },
    {
      "epoch": 421.05309734513276,
      "eval_loss": 1.0303720235824585,
      "eval_runtime": 0.9422,
      "eval_samples_per_second": 26.534,
      "eval_steps_per_second": 13.798,
      "num_input_tokens_seen": 13299920,
      "step": 24000
    },
    {
      "epoch": 421.14159292035396,
      "grad_norm": 7.75,
      "learning_rate": 1.726710594157177e-05,
      "loss": 1.1632,
      "num_input_tokens_seen": 13302912,
      "step": 24005
    },
    {
      "epoch": 421.2300884955752,
      "grad_norm": 9.5625,
      "learning_rate": 1.725777051609846e-05,
      "loss": 1.2107,
      "num_input_tokens_seen": 13305488,
      "step": 24010
    },
    {
      "epoch": 421.3185840707965,
      "grad_norm": 7.28125,
      "learning_rate": 1.7248436284574228e-05,
      "loss": 0.9073,
      "num_input_tokens_seen": 13308496,
      "step": 24015
    },
    {
      "epoch": 421.4070796460177,
      "grad_norm": 10.5625,
      "learning_rate": 1.723910324843855e-05,
      "loss": 1.2742,
      "num_input_tokens_seen": 13310960,
      "step": 24020
    },
    {
      "epoch": 421.49557522123894,
      "grad_norm": 7.59375,
      "learning_rate": 1.722977140913067e-05,
      "loss": 1.0419,
      "num_input_tokens_seen": 13314112,
      "step": 24025
    },
    {
      "epoch": 421.5840707964602,
      "grad_norm": 7.125,
      "learning_rate": 1.7220440768089688e-05,
      "loss": 0.902,
      "num_input_tokens_seen": 13316864,
      "step": 24030
    },
    {
      "epoch": 421.6725663716814,
      "grad_norm": 8.4375,
      "learning_rate": 1.7211111326754505e-05,
      "loss": 1.131,
      "num_input_tokens_seen": 13319376,
      "step": 24035
    },
    {
      "epoch": 421.76106194690266,
      "grad_norm": 9.5,
      "learning_rate": 1.720178308656383e-05,
      "loss": 0.9306,
      "num_input_tokens_seen": 13322416,
      "step": 24040
    },
    {
      "epoch": 421.8495575221239,
      "grad_norm": 9.125,
      "learning_rate": 1.719245604895621e-05,
      "loss": 1.0298,
      "num_input_tokens_seen": 13325184,
      "step": 24045
    },
    {
      "epoch": 421.9380530973451,
      "grad_norm": 8.125,
      "learning_rate": 1.7183130215369972e-05,
      "loss": 0.9842,
      "num_input_tokens_seen": 13328048,
      "step": 24050
    },
    {
      "epoch": 422.01769911504425,
      "grad_norm": 13.25,
      "learning_rate": 1.7173805587243292e-05,
      "loss": 1.1132,
      "num_input_tokens_seen": 13330216,
      "step": 24055
    },
    {
      "epoch": 422.1061946902655,
      "grad_norm": 5.4375,
      "learning_rate": 1.7164482166014147e-05,
      "loss": 1.0768,
      "num_input_tokens_seen": 13332984,
      "step": 24060
    },
    {
      "epoch": 422.1946902654867,
      "grad_norm": 8.0625,
      "learning_rate": 1.7155159953120313e-05,
      "loss": 1.1827,
      "num_input_tokens_seen": 13335368,
      "step": 24065
    },
    {
      "epoch": 422.283185840708,
      "grad_norm": 7.78125,
      "learning_rate": 1.714583894999941e-05,
      "loss": 0.7841,
      "num_input_tokens_seen": 13338104,
      "step": 24070
    },
    {
      "epoch": 422.37168141592923,
      "grad_norm": 8.0,
      "learning_rate": 1.7136519158088826e-05,
      "loss": 0.9744,
      "num_input_tokens_seen": 13341800,
      "step": 24075
    },
    {
      "epoch": 422.46017699115043,
      "grad_norm": 8.375,
      "learning_rate": 1.712720057882581e-05,
      "loss": 1.2691,
      "num_input_tokens_seen": 13344440,
      "step": 24080
    },
    {
      "epoch": 422.5486725663717,
      "grad_norm": 8.375,
      "learning_rate": 1.7117883213647413e-05,
      "loss": 0.8478,
      "num_input_tokens_seen": 13347352,
      "step": 24085
    },
    {
      "epoch": 422.6371681415929,
      "grad_norm": 7.9375,
      "learning_rate": 1.710856706399046e-05,
      "loss": 1.0736,
      "num_input_tokens_seen": 13350376,
      "step": 24090
    },
    {
      "epoch": 422.72566371681415,
      "grad_norm": 7.21875,
      "learning_rate": 1.7099252131291648e-05,
      "loss": 0.9679,
      "num_input_tokens_seen": 13353064,
      "step": 24095
    },
    {
      "epoch": 422.8141592920354,
      "grad_norm": 8.8125,
      "learning_rate": 1.708993841698744e-05,
      "loss": 1.066,
      "num_input_tokens_seen": 13355800,
      "step": 24100
    },
    {
      "epoch": 422.9026548672566,
      "grad_norm": 7.53125,
      "learning_rate": 1.7080625922514132e-05,
      "loss": 1.1144,
      "num_input_tokens_seen": 13358680,
      "step": 24105
    },
    {
      "epoch": 422.9911504424779,
      "grad_norm": 7.28125,
      "learning_rate": 1.7071314649307836e-05,
      "loss": 1.2394,
      "num_input_tokens_seen": 13361224,
      "step": 24110
    },
    {
      "epoch": 423.070796460177,
      "grad_norm": 10.375,
      "learning_rate": 1.7062004598804448e-05,
      "loss": 1.0622,
      "num_input_tokens_seen": 13363368,
      "step": 24115
    },
    {
      "epoch": 423.1592920353982,
      "grad_norm": 10.0,
      "learning_rate": 1.7052695772439702e-05,
      "loss": 1.0996,
      "num_input_tokens_seen": 13366088,
      "step": 24120
    },
    {
      "epoch": 423.24778761061947,
      "grad_norm": 6.84375,
      "learning_rate": 1.7043388171649154e-05,
      "loss": 1.0812,
      "num_input_tokens_seen": 13368824,
      "step": 24125
    },
    {
      "epoch": 423.3362831858407,
      "grad_norm": 7.84375,
      "learning_rate": 1.7034081797868127e-05,
      "loss": 0.9162,
      "num_input_tokens_seen": 13371768,
      "step": 24130
    },
    {
      "epoch": 423.42477876106193,
      "grad_norm": 12.25,
      "learning_rate": 1.70247766525318e-05,
      "loss": 1.0771,
      "num_input_tokens_seen": 13374360,
      "step": 24135
    },
    {
      "epoch": 423.5132743362832,
      "grad_norm": 7.71875,
      "learning_rate": 1.701547273707514e-05,
      "loss": 1.0836,
      "num_input_tokens_seen": 13377304,
      "step": 24140
    },
    {
      "epoch": 423.60176991150445,
      "grad_norm": 7.6875,
      "learning_rate": 1.7006170052932916e-05,
      "loss": 0.9219,
      "num_input_tokens_seen": 13380024,
      "step": 24145
    },
    {
      "epoch": 423.69026548672565,
      "grad_norm": 10.5,
      "learning_rate": 1.6996868601539735e-05,
      "loss": 1.1756,
      "num_input_tokens_seen": 13383176,
      "step": 24150
    },
    {
      "epoch": 423.7787610619469,
      "grad_norm": 7.78125,
      "learning_rate": 1.6987568384329977e-05,
      "loss": 1.0343,
      "num_input_tokens_seen": 13385896,
      "step": 24155
    },
    {
      "epoch": 423.86725663716817,
      "grad_norm": 8.0625,
      "learning_rate": 1.6978269402737866e-05,
      "loss": 1.0671,
      "num_input_tokens_seen": 13388728,
      "step": 24160
    },
    {
      "epoch": 423.95575221238937,
      "grad_norm": 7.75,
      "learning_rate": 1.696897165819743e-05,
      "loss": 1.1673,
      "num_input_tokens_seen": 13391560,
      "step": 24165
    },
    {
      "epoch": 424.0353982300885,
      "grad_norm": 6.5625,
      "learning_rate": 1.6959675152142487e-05,
      "loss": 0.9798,
      "num_input_tokens_seen": 13394232,
      "step": 24170
    },
    {
      "epoch": 424.12389380530976,
      "grad_norm": 7.75,
      "learning_rate": 1.6950379886006667e-05,
      "loss": 0.925,
      "num_input_tokens_seen": 13397400,
      "step": 24175
    },
    {
      "epoch": 424.21238938053096,
      "grad_norm": 8.875,
      "learning_rate": 1.6941085861223438e-05,
      "loss": 1.1775,
      "num_input_tokens_seen": 13400056,
      "step": 24180
    },
    {
      "epoch": 424.3008849557522,
      "grad_norm": 10.125,
      "learning_rate": 1.6931793079226034e-05,
      "loss": 0.934,
      "num_input_tokens_seen": 13402696,
      "step": 24185
    },
    {
      "epoch": 424.3893805309734,
      "grad_norm": 6.90625,
      "learning_rate": 1.692250154144754e-05,
      "loss": 1.2689,
      "num_input_tokens_seen": 13405048,
      "step": 24190
    },
    {
      "epoch": 424.4778761061947,
      "grad_norm": 8.3125,
      "learning_rate": 1.6913211249320807e-05,
      "loss": 1.0661,
      "num_input_tokens_seen": 13407848,
      "step": 24195
    },
    {
      "epoch": 424.56637168141594,
      "grad_norm": 8.3125,
      "learning_rate": 1.6903922204278522e-05,
      "loss": 0.9668,
      "num_input_tokens_seen": 13410872,
      "step": 24200
    },
    {
      "epoch": 424.56637168141594,
      "eval_loss": 1.0240559577941895,
      "eval_runtime": 0.9351,
      "eval_samples_per_second": 26.736,
      "eval_steps_per_second": 13.903,
      "num_input_tokens_seen": 13410872,
      "step": 24200
    },
    {
      "epoch": 424.65486725663715,
      "grad_norm": 9.0,
      "learning_rate": 1.6894634407753186e-05,
      "loss": 0.9931,
      "num_input_tokens_seen": 13413704,
      "step": 24205
    },
    {
      "epoch": 424.7433628318584,
      "grad_norm": 9.5625,
      "learning_rate": 1.6885347861177077e-05,
      "loss": 1.0909,
      "num_input_tokens_seen": 13417000,
      "step": 24210
    },
    {
      "epoch": 424.83185840707966,
      "grad_norm": 8.0,
      "learning_rate": 1.6876062565982298e-05,
      "loss": 1.0084,
      "num_input_tokens_seen": 13419800,
      "step": 24215
    },
    {
      "epoch": 424.92035398230087,
      "grad_norm": 8.9375,
      "learning_rate": 1.6866778523600774e-05,
      "loss": 1.1255,
      "num_input_tokens_seen": 13422536,
      "step": 24220
    },
    {
      "epoch": 425.0,
      "grad_norm": 44.5,
      "learning_rate": 1.6857495735464195e-05,
      "loss": 1.6413,
      "num_input_tokens_seen": 13424656,
      "step": 24225
    },
    {
      "epoch": 425.08849557522126,
      "grad_norm": 7.34375,
      "learning_rate": 1.6848214203004115e-05,
      "loss": 1.1841,
      "num_input_tokens_seen": 13427536,
      "step": 24230
    },
    {
      "epoch": 425.17699115044246,
      "grad_norm": 8.25,
      "learning_rate": 1.6838933927651835e-05,
      "loss": 1.0326,
      "num_input_tokens_seen": 13430496,
      "step": 24235
    },
    {
      "epoch": 425.2654867256637,
      "grad_norm": 7.0,
      "learning_rate": 1.6829654910838506e-05,
      "loss": 0.9754,
      "num_input_tokens_seen": 13433312,
      "step": 24240
    },
    {
      "epoch": 425.353982300885,
      "grad_norm": 8.125,
      "learning_rate": 1.6820377153995065e-05,
      "loss": 1.0618,
      "num_input_tokens_seen": 13435904,
      "step": 24245
    },
    {
      "epoch": 425.4424778761062,
      "grad_norm": 7.1875,
      "learning_rate": 1.681110065855226e-05,
      "loss": 0.9963,
      "num_input_tokens_seen": 13438352,
      "step": 24250
    },
    {
      "epoch": 425.53097345132744,
      "grad_norm": 8.9375,
      "learning_rate": 1.6801825425940642e-05,
      "loss": 1.0171,
      "num_input_tokens_seen": 13441280,
      "step": 24255
    },
    {
      "epoch": 425.6194690265487,
      "grad_norm": 7.03125,
      "learning_rate": 1.679255145759056e-05,
      "loss": 0.9709,
      "num_input_tokens_seen": 13444096,
      "step": 24260
    },
    {
      "epoch": 425.7079646017699,
      "grad_norm": 8.375,
      "learning_rate": 1.6783278754932187e-05,
      "loss": 1.0425,
      "num_input_tokens_seen": 13446880,
      "step": 24265
    },
    {
      "epoch": 425.79646017699116,
      "grad_norm": 11.375,
      "learning_rate": 1.6774007319395496e-05,
      "loss": 1.0287,
      "num_input_tokens_seen": 13449952,
      "step": 24270
    },
    {
      "epoch": 425.88495575221236,
      "grad_norm": 6.75,
      "learning_rate": 1.6764737152410243e-05,
      "loss": 1.1795,
      "num_input_tokens_seen": 13452656,
      "step": 24275
    },
    {
      "epoch": 425.9734513274336,
      "grad_norm": 9.25,
      "learning_rate": 1.6755468255406016e-05,
      "loss": 1.1396,
      "num_input_tokens_seen": 13455504,
      "step": 24280
    },
    {
      "epoch": 426.05309734513276,
      "grad_norm": 8.625,
      "learning_rate": 1.674620062981219e-05,
      "loss": 1.0778,
      "num_input_tokens_seen": 13457984,
      "step": 24285
    },
    {
      "epoch": 426.14159292035396,
      "grad_norm": 8.1875,
      "learning_rate": 1.6736934277057947e-05,
      "loss": 1.0072,
      "num_input_tokens_seen": 13460928,
      "step": 24290
    },
    {
      "epoch": 426.2300884955752,
      "grad_norm": 7.125,
      "learning_rate": 1.6727669198572286e-05,
      "loss": 1.066,
      "num_input_tokens_seen": 13463456,
      "step": 24295
    },
    {
      "epoch": 426.3185840707965,
      "grad_norm": 8.125,
      "learning_rate": 1.6718405395783984e-05,
      "loss": 0.9164,
      "num_input_tokens_seen": 13466688,
      "step": 24300
    },
    {
      "epoch": 426.4070796460177,
      "grad_norm": 8.1875,
      "learning_rate": 1.6709142870121643e-05,
      "loss": 1.1629,
      "num_input_tokens_seen": 13469136,
      "step": 24305
    },
    {
      "epoch": 426.49557522123894,
      "grad_norm": 8.875,
      "learning_rate": 1.669988162301367e-05,
      "loss": 1.1163,
      "num_input_tokens_seen": 13472144,
      "step": 24310
    },
    {
      "epoch": 426.5840707964602,
      "grad_norm": 11.5,
      "learning_rate": 1.6690621655888243e-05,
      "loss": 1.2157,
      "num_input_tokens_seen": 13475216,
      "step": 24315
    },
    {
      "epoch": 426.6725663716814,
      "grad_norm": 8.125,
      "learning_rate": 1.6681362970173386e-05,
      "loss": 0.8895,
      "num_input_tokens_seen": 13478432,
      "step": 24320
    },
    {
      "epoch": 426.76106194690266,
      "grad_norm": 7.15625,
      "learning_rate": 1.6672105567296904e-05,
      "loss": 0.9775,
      "num_input_tokens_seen": 13481440,
      "step": 24325
    },
    {
      "epoch": 426.8495575221239,
      "grad_norm": 7.53125,
      "learning_rate": 1.666284944868639e-05,
      "loss": 1.0308,
      "num_input_tokens_seen": 13483792,
      "step": 24330
    },
    {
      "epoch": 426.9380530973451,
      "grad_norm": 7.75,
      "learning_rate": 1.665359461576927e-05,
      "loss": 1.1167,
      "num_input_tokens_seen": 13486432,
      "step": 24335
    },
    {
      "epoch": 427.01769911504425,
      "grad_norm": 8.75,
      "learning_rate": 1.6644341069972736e-05,
      "loss": 1.1611,
      "num_input_tokens_seen": 13488896,
      "step": 24340
    },
    {
      "epoch": 427.1061946902655,
      "grad_norm": 6.6875,
      "learning_rate": 1.6635088812723813e-05,
      "loss": 1.0998,
      "num_input_tokens_seen": 13491264,
      "step": 24345
    },
    {
      "epoch": 427.1946902654867,
      "grad_norm": 8.3125,
      "learning_rate": 1.6625837845449328e-05,
      "loss": 1.0544,
      "num_input_tokens_seen": 13494336,
      "step": 24350
    },
    {
      "epoch": 427.283185840708,
      "grad_norm": 7.5,
      "learning_rate": 1.6616588169575874e-05,
      "loss": 0.9119,
      "num_input_tokens_seen": 13497680,
      "step": 24355
    },
    {
      "epoch": 427.37168141592923,
      "grad_norm": 12.5,
      "learning_rate": 1.6607339786529878e-05,
      "loss": 1.242,
      "num_input_tokens_seen": 13500672,
      "step": 24360
    },
    {
      "epoch": 427.46017699115043,
      "grad_norm": 8.5625,
      "learning_rate": 1.659809269773756e-05,
      "loss": 0.8234,
      "num_input_tokens_seen": 13503440,
      "step": 24365
    },
    {
      "epoch": 427.5486725663717,
      "grad_norm": 8.5,
      "learning_rate": 1.658884690462493e-05,
      "loss": 1.2665,
      "num_input_tokens_seen": 13506016,
      "step": 24370
    },
    {
      "epoch": 427.6371681415929,
      "grad_norm": 8.375,
      "learning_rate": 1.6579602408617813e-05,
      "loss": 1.004,
      "num_input_tokens_seen": 13508800,
      "step": 24375
    },
    {
      "epoch": 427.72566371681415,
      "grad_norm": 8.5,
      "learning_rate": 1.657035921114181e-05,
      "loss": 0.991,
      "num_input_tokens_seen": 13511712,
      "step": 24380
    },
    {
      "epoch": 427.8141592920354,
      "grad_norm": 8.4375,
      "learning_rate": 1.656111731362236e-05,
      "loss": 1.1212,
      "num_input_tokens_seen": 13514528,
      "step": 24385
    },
    {
      "epoch": 427.9026548672566,
      "grad_norm": 8.6875,
      "learning_rate": 1.6551876717484666e-05,
      "loss": 1.0075,
      "num_input_tokens_seen": 13517728,
      "step": 24390
    },
    {
      "epoch": 427.9911504424779,
      "grad_norm": 11.75,
      "learning_rate": 1.6542637424153752e-05,
      "loss": 1.1864,
      "num_input_tokens_seen": 13520208,
      "step": 24395
    },
    {
      "epoch": 428.070796460177,
      "grad_norm": 11.8125,
      "learning_rate": 1.6533399435054418e-05,
      "loss": 1.0,
      "num_input_tokens_seen": 13522656,
      "step": 24400
    },
    {
      "epoch": 428.070796460177,
      "eval_loss": 1.0349030494689941,
      "eval_runtime": 0.9361,
      "eval_samples_per_second": 26.706,
      "eval_steps_per_second": 13.887,
      "num_input_tokens_seen": 13522656,
      "step": 24400
    },
    {
      "epoch": 428.1592920353982,
      "grad_norm": 7.0625,
      "learning_rate": 1.6524162751611304e-05,
      "loss": 0.9878,
      "num_input_tokens_seen": 13526464,
      "step": 24405
    },
    {
      "epoch": 428.24778761061947,
      "grad_norm": 9.5625,
      "learning_rate": 1.6514927375248796e-05,
      "loss": 1.0192,
      "num_input_tokens_seen": 13529376,
      "step": 24410
    },
    {
      "epoch": 428.3362831858407,
      "grad_norm": 7.78125,
      "learning_rate": 1.6505693307391127e-05,
      "loss": 1.062,
      "num_input_tokens_seen": 13532256,
      "step": 24415
    },
    {
      "epoch": 428.42477876106193,
      "grad_norm": 8.8125,
      "learning_rate": 1.6496460549462288e-05,
      "loss": 0.9513,
      "num_input_tokens_seen": 13535072,
      "step": 24420
    },
    {
      "epoch": 428.5132743362832,
      "grad_norm": 9.1875,
      "learning_rate": 1.6487229102886097e-05,
      "loss": 1.1435,
      "num_input_tokens_seen": 13537552,
      "step": 24425
    },
    {
      "epoch": 428.60176991150445,
      "grad_norm": 10.3125,
      "learning_rate": 1.6477998969086155e-05,
      "loss": 1.1715,
      "num_input_tokens_seen": 13540112,
      "step": 24430
    },
    {
      "epoch": 428.69026548672565,
      "grad_norm": 8.4375,
      "learning_rate": 1.646877014948587e-05,
      "loss": 1.1173,
      "num_input_tokens_seen": 13542480,
      "step": 24435
    },
    {
      "epoch": 428.7787610619469,
      "grad_norm": 9.25,
      "learning_rate": 1.6459542645508433e-05,
      "loss": 0.835,
      "num_input_tokens_seen": 13545456,
      "step": 24440
    },
    {
      "epoch": 428.86725663716817,
      "grad_norm": 6.71875,
      "learning_rate": 1.6450316458576852e-05,
      "loss": 1.1558,
      "num_input_tokens_seen": 13547968,
      "step": 24445
    },
    {
      "epoch": 428.95575221238937,
      "grad_norm": 7.03125,
      "learning_rate": 1.6441091590113912e-05,
      "loss": 1.0004,
      "num_input_tokens_seen": 13550976,
      "step": 24450
    },
    {
      "epoch": 429.0353982300885,
      "grad_norm": 6.9375,
      "learning_rate": 1.6431868041542213e-05,
      "loss": 0.9782,
      "num_input_tokens_seen": 13553344,
      "step": 24455
    },
    {
      "epoch": 429.12389380530976,
      "grad_norm": 7.40625,
      "learning_rate": 1.6422645814284123e-05,
      "loss": 0.9381,
      "num_input_tokens_seen": 13556240,
      "step": 24460
    },
    {
      "epoch": 429.21238938053096,
      "grad_norm": 8.4375,
      "learning_rate": 1.6413424909761846e-05,
      "loss": 1.161,
      "num_input_tokens_seen": 13558768,
      "step": 24465
    },
    {
      "epoch": 429.3008849557522,
      "grad_norm": 8.625,
      "learning_rate": 1.640420532939736e-05,
      "loss": 1.1344,
      "num_input_tokens_seen": 13561520,
      "step": 24470
    },
    {
      "epoch": 429.3893805309734,
      "grad_norm": 8.8125,
      "learning_rate": 1.639498707461242e-05,
      "loss": 1.1636,
      "num_input_tokens_seen": 13564512,
      "step": 24475
    },
    {
      "epoch": 429.4778761061947,
      "grad_norm": 7.15625,
      "learning_rate": 1.6385770146828614e-05,
      "loss": 0.9373,
      "num_input_tokens_seen": 13567104,
      "step": 24480
    },
    {
      "epoch": 429.56637168141594,
      "grad_norm": 7.96875,
      "learning_rate": 1.637655454746731e-05,
      "loss": 0.9639,
      "num_input_tokens_seen": 13570288,
      "step": 24485
    },
    {
      "epoch": 429.65486725663715,
      "grad_norm": 7.90625,
      "learning_rate": 1.6367340277949658e-05,
      "loss": 1.1284,
      "num_input_tokens_seen": 13573056,
      "step": 24490
    },
    {
      "epoch": 429.7433628318584,
      "grad_norm": 11.1875,
      "learning_rate": 1.635812733969663e-05,
      "loss": 0.9278,
      "num_input_tokens_seen": 13576080,
      "step": 24495
    },
    {
      "epoch": 429.83185840707966,
      "grad_norm": 9.4375,
      "learning_rate": 1.634891573412896e-05,
      "loss": 1.1248,
      "num_input_tokens_seen": 13578928,
      "step": 24500
    },
    {
      "epoch": 429.92035398230087,
      "grad_norm": 9.3125,
      "learning_rate": 1.6339705462667196e-05,
      "loss": 1.245,
      "num_input_tokens_seen": 13581536,
      "step": 24505
    },
    {
      "epoch": 430.0,
      "grad_norm": 7.9375,
      "learning_rate": 1.633049652673169e-05,
      "loss": 0.9518,
      "num_input_tokens_seen": 13583664,
      "step": 24510
    },
    {
      "epoch": 430.08849557522126,
      "grad_norm": 8.0625,
      "learning_rate": 1.632128892774256e-05,
      "loss": 0.894,
      "num_input_tokens_seen": 13586384,
      "step": 24515
    },
    {
      "epoch": 430.17699115044246,
      "grad_norm": 6.40625,
      "learning_rate": 1.6312082667119737e-05,
      "loss": 1.0406,
      "num_input_tokens_seen": 13589568,
      "step": 24520
    },
    {
      "epoch": 430.2654867256637,
      "grad_norm": 9.3125,
      "learning_rate": 1.630287774628296e-05,
      "loss": 1.0237,
      "num_input_tokens_seen": 13591904,
      "step": 24525
    },
    {
      "epoch": 430.353982300885,
      "grad_norm": 8.3125,
      "learning_rate": 1.6293674166651718e-05,
      "loss": 0.84,
      "num_input_tokens_seen": 13594720,
      "step": 24530
    },
    {
      "epoch": 430.4424778761062,
      "grad_norm": 9.8125,
      "learning_rate": 1.6284471929645338e-05,
      "loss": 1.2396,
      "num_input_tokens_seen": 13597520,
      "step": 24535
    },
    {
      "epoch": 430.53097345132744,
      "grad_norm": 7.03125,
      "learning_rate": 1.627527103668291e-05,
      "loss": 0.9911,
      "num_input_tokens_seen": 13600688,
      "step": 24540
    },
    {
      "epoch": 430.6194690265487,
      "grad_norm": 10.875,
      "learning_rate": 1.6266071489183327e-05,
      "loss": 1.1021,
      "num_input_tokens_seen": 13603472,
      "step": 24545
    },
    {
      "epoch": 430.7079646017699,
      "grad_norm": 9.875,
      "learning_rate": 1.6256873288565283e-05,
      "loss": 1.1675,
      "num_input_tokens_seen": 13606112,
      "step": 24550
    },
    {
      "epoch": 430.79646017699116,
      "grad_norm": 10.0625,
      "learning_rate": 1.6247676436247245e-05,
      "loss": 0.9144,
      "num_input_tokens_seen": 13608784,
      "step": 24555
    },
    {
      "epoch": 430.88495575221236,
      "grad_norm": 8.6875,
      "learning_rate": 1.6238480933647486e-05,
      "loss": 1.3121,
      "num_input_tokens_seen": 13611536,
      "step": 24560
    },
    {
      "epoch": 430.9734513274336,
      "grad_norm": 8.625,
      "learning_rate": 1.6229286782184083e-05,
      "loss": 0.9682,
      "num_input_tokens_seen": 13614256,
      "step": 24565
    },
    {
      "epoch": 431.05309734513276,
      "grad_norm": 9.5,
      "learning_rate": 1.622009398327487e-05,
      "loss": 1.2879,
      "num_input_tokens_seen": 13616632,
      "step": 24570
    },
    {
      "epoch": 431.14159292035396,
      "grad_norm": 7.1875,
      "learning_rate": 1.6210902538337502e-05,
      "loss": 1.0726,
      "num_input_tokens_seen": 13619256,
      "step": 24575
    },
    {
      "epoch": 431.2300884955752,
      "grad_norm": 6.875,
      "learning_rate": 1.6201712448789413e-05,
      "loss": 0.9616,
      "num_input_tokens_seen": 13622136,
      "step": 24580
    },
    {
      "epoch": 431.3185840707965,
      "grad_norm": 7.25,
      "learning_rate": 1.6192523716047827e-05,
      "loss": 1.077,
      "num_input_tokens_seen": 13624872,
      "step": 24585
    },
    {
      "epoch": 431.4070796460177,
      "grad_norm": 8.375,
      "learning_rate": 1.6183336341529776e-05,
      "loss": 0.9868,
      "num_input_tokens_seen": 13627256,
      "step": 24590
    },
    {
      "epoch": 431.49557522123894,
      "grad_norm": 8.0,
      "learning_rate": 1.6174150326652047e-05,
      "loss": 1.138,
      "num_input_tokens_seen": 13629608,
      "step": 24595
    },
    {
      "epoch": 431.5840707964602,
      "grad_norm": 8.4375,
      "learning_rate": 1.6164965672831256e-05,
      "loss": 1.1078,
      "num_input_tokens_seen": 13632696,
      "step": 24600
    },
    {
      "epoch": 431.5840707964602,
      "eval_loss": 1.0262783765792847,
      "eval_runtime": 0.9346,
      "eval_samples_per_second": 26.751,
      "eval_steps_per_second": 13.91,
      "num_input_tokens_seen": 13632696,
      "step": 24600
    },
    {
      "epoch": 431.6725663716814,
      "grad_norm": 9.4375,
      "learning_rate": 1.6155782381483784e-05,
      "loss": 0.9567,
      "num_input_tokens_seen": 13635816,
      "step": 24605
    },
    {
      "epoch": 431.76106194690266,
      "grad_norm": 7.65625,
      "learning_rate": 1.6146600454025813e-05,
      "loss": 1.0497,
      "num_input_tokens_seen": 13639000,
      "step": 24610
    },
    {
      "epoch": 431.8495575221239,
      "grad_norm": 6.96875,
      "learning_rate": 1.6137419891873317e-05,
      "loss": 1.0802,
      "num_input_tokens_seen": 13641784,
      "step": 24615
    },
    {
      "epoch": 431.9380530973451,
      "grad_norm": 10.0,
      "learning_rate": 1.6128240696442038e-05,
      "loss": 1.1165,
      "num_input_tokens_seen": 13644296,
      "step": 24620
    },
    {
      "epoch": 432.01769911504425,
      "grad_norm": 9.3125,
      "learning_rate": 1.611906286914753e-05,
      "loss": 0.9573,
      "num_input_tokens_seen": 13646576,
      "step": 24625
    },
    {
      "epoch": 432.1061946902655,
      "grad_norm": 9.4375,
      "learning_rate": 1.6109886411405144e-05,
      "loss": 1.3303,
      "num_input_tokens_seen": 13649040,
      "step": 24630
    },
    {
      "epoch": 432.1946902654867,
      "grad_norm": 8.375,
      "learning_rate": 1.6100711324629985e-05,
      "loss": 1.1535,
      "num_input_tokens_seen": 13651888,
      "step": 24635
    },
    {
      "epoch": 432.283185840708,
      "grad_norm": 9.0,
      "learning_rate": 1.609153761023698e-05,
      "loss": 1.13,
      "num_input_tokens_seen": 13654848,
      "step": 24640
    },
    {
      "epoch": 432.37168141592923,
      "grad_norm": 7.71875,
      "learning_rate": 1.608236526964083e-05,
      "loss": 0.9546,
      "num_input_tokens_seen": 13657504,
      "step": 24645
    },
    {
      "epoch": 432.46017699115043,
      "grad_norm": 7.71875,
      "learning_rate": 1.607319430425601e-05,
      "loss": 1.076,
      "num_input_tokens_seen": 13660224,
      "step": 24650
    },
    {
      "epoch": 432.5486725663717,
      "grad_norm": 6.59375,
      "learning_rate": 1.606402471549682e-05,
      "loss": 1.0343,
      "num_input_tokens_seen": 13662736,
      "step": 24655
    },
    {
      "epoch": 432.6371681415929,
      "grad_norm": 12.3125,
      "learning_rate": 1.6054856504777312e-05,
      "loss": 1.1375,
      "num_input_tokens_seen": 13666288,
      "step": 24660
    },
    {
      "epoch": 432.72566371681415,
      "grad_norm": 8.875,
      "learning_rate": 1.6045689673511334e-05,
      "loss": 0.8954,
      "num_input_tokens_seen": 13669136,
      "step": 24665
    },
    {
      "epoch": 432.8141592920354,
      "grad_norm": 8.375,
      "learning_rate": 1.6036524223112548e-05,
      "loss": 1.0206,
      "num_input_tokens_seen": 13671952,
      "step": 24670
    },
    {
      "epoch": 432.9026548672566,
      "grad_norm": 9.125,
      "learning_rate": 1.602736015499436e-05,
      "loss": 0.8794,
      "num_input_tokens_seen": 13674448,
      "step": 24675
    },
    {
      "epoch": 432.9911504424779,
      "grad_norm": 7.15625,
      "learning_rate": 1.601819747057e-05,
      "loss": 0.9881,
      "num_input_tokens_seen": 13677840,
      "step": 24680
    },
    {
      "epoch": 433.070796460177,
      "grad_norm": 7.96875,
      "learning_rate": 1.6009036171252465e-05,
      "loss": 0.8432,
      "num_input_tokens_seen": 13680552,
      "step": 24685
    },
    {
      "epoch": 433.1592920353982,
      "grad_norm": 14.8125,
      "learning_rate": 1.599987625845453e-05,
      "loss": 1.0886,
      "num_input_tokens_seen": 13683784,
      "step": 24690
    },
    {
      "epoch": 433.24778761061947,
      "grad_norm": 8.0625,
      "learning_rate": 1.599071773358879e-05,
      "loss": 0.9673,
      "num_input_tokens_seen": 13686728,
      "step": 24695
    },
    {
      "epoch": 433.3362831858407,
      "grad_norm": 9.0,
      "learning_rate": 1.598156059806758e-05,
      "loss": 1.0765,
      "num_input_tokens_seen": 13689496,
      "step": 24700
    },
    {
      "epoch": 433.42477876106193,
      "grad_norm": 9.4375,
      "learning_rate": 1.5972404853303062e-05,
      "loss": 1.0991,
      "num_input_tokens_seen": 13692008,
      "step": 24705
    },
    {
      "epoch": 433.5132743362832,
      "grad_norm": 7.25,
      "learning_rate": 1.5963250500707172e-05,
      "loss": 1.0691,
      "num_input_tokens_seen": 13695080,
      "step": 24710
    },
    {
      "epoch": 433.60176991150445,
      "grad_norm": 8.4375,
      "learning_rate": 1.5954097541691612e-05,
      "loss": 1.3034,
      "num_input_tokens_seen": 13697480,
      "step": 24715
    },
    {
      "epoch": 433.69026548672565,
      "grad_norm": 6.90625,
      "learning_rate": 1.5944945977667884e-05,
      "loss": 0.9011,
      "num_input_tokens_seen": 13700328,
      "step": 24720
    },
    {
      "epoch": 433.7787610619469,
      "grad_norm": 8.25,
      "learning_rate": 1.593579581004729e-05,
      "loss": 1.0828,
      "num_input_tokens_seen": 13703272,
      "step": 24725
    },
    {
      "epoch": 433.86725663716817,
      "grad_norm": 8.375,
      "learning_rate": 1.592664704024088e-05,
      "loss": 1.0232,
      "num_input_tokens_seen": 13705544,
      "step": 24730
    },
    {
      "epoch": 433.95575221238937,
      "grad_norm": 7.625,
      "learning_rate": 1.591749966965953e-05,
      "loss": 0.9831,
      "num_input_tokens_seen": 13708248,
      "step": 24735
    },
    {
      "epoch": 434.0353982300885,
      "grad_norm": 7.5625,
      "learning_rate": 1.5908353699713856e-05,
      "loss": 0.9697,
      "num_input_tokens_seen": 13710856,
      "step": 24740
    },
    {
      "epoch": 434.12389380530976,
      "grad_norm": 6.40625,
      "learning_rate": 1.5899209131814298e-05,
      "loss": 0.951,
      "num_input_tokens_seen": 13713800,
      "step": 24745
    },
    {
      "epoch": 434.21238938053096,
      "grad_norm": 7.96875,
      "learning_rate": 1.5890065967371067e-05,
      "loss": 0.9546,
      "num_input_tokens_seen": 13716648,
      "step": 24750
    },
    {
      "epoch": 434.3008849557522,
      "grad_norm": 6.46875,
      "learning_rate": 1.5880924207794144e-05,
      "loss": 0.9824,
      "num_input_tokens_seen": 13719064,
      "step": 24755
    },
    {
      "epoch": 434.3893805309734,
      "grad_norm": 8.625,
      "learning_rate": 1.5871783854493298e-05,
      "loss": 0.9545,
      "num_input_tokens_seen": 13721928,
      "step": 24760
    },
    {
      "epoch": 434.4778761061947,
      "grad_norm": 9.875,
      "learning_rate": 1.5862644908878106e-05,
      "loss": 1.2922,
      "num_input_tokens_seen": 13724872,
      "step": 24765
    },
    {
      "epoch": 434.56637168141594,
      "grad_norm": 8.0,
      "learning_rate": 1.5853507372357885e-05,
      "loss": 1.1426,
      "num_input_tokens_seen": 13727960,
      "step": 24770
    },
    {
      "epoch": 434.65486725663715,
      "grad_norm": 7.4375,
      "learning_rate": 1.5844371246341776e-05,
      "loss": 1.2035,
      "num_input_tokens_seen": 13730344,
      "step": 24775
    },
    {
      "epoch": 434.7433628318584,
      "grad_norm": 8.375,
      "learning_rate": 1.5835236532238674e-05,
      "loss": 0.9857,
      "num_input_tokens_seen": 13732984,
      "step": 24780
    },
    {
      "epoch": 434.83185840707966,
      "grad_norm": 8.0625,
      "learning_rate": 1.582610323145727e-05,
      "loss": 1.1584,
      "num_input_tokens_seen": 13736168,
      "step": 24785
    },
    {
      "epoch": 434.92035398230087,
      "grad_norm": 9.5,
      "learning_rate": 1.5816971345406035e-05,
      "loss": 1.0916,
      "num_input_tokens_seen": 13738904,
      "step": 24790
    },
    {
      "epoch": 435.0,
      "grad_norm": 12.0625,
      "learning_rate": 1.5807840875493225e-05,
      "loss": 0.9003,
      "num_input_tokens_seen": 13741160,
      "step": 24795
    },
    {
      "epoch": 435.08849557522126,
      "grad_norm": 10.125,
      "learning_rate": 1.5798711823126854e-05,
      "loss": 1.2902,
      "num_input_tokens_seen": 13743576,
      "step": 24800
    },
    {
      "epoch": 435.08849557522126,
      "eval_loss": 1.0262783765792847,
      "eval_runtime": 0.9877,
      "eval_samples_per_second": 25.311,
      "eval_steps_per_second": 13.162,
      "num_input_tokens_seen": 13743576,
      "step": 24800
    },
    {
      "epoch": 435.17699115044246,
      "grad_norm": 7.9375,
      "learning_rate": 1.578958418971477e-05,
      "loss": 1.0472,
      "num_input_tokens_seen": 13746552,
      "step": 24805
    },
    {
      "epoch": 435.2654867256637,
      "grad_norm": 15.125,
      "learning_rate": 1.578045797666453e-05,
      "loss": 1.0646,
      "num_input_tokens_seen": 13749656,
      "step": 24810
    },
    {
      "epoch": 435.353982300885,
      "grad_norm": 8.8125,
      "learning_rate": 1.5771333185383548e-05,
      "loss": 0.9732,
      "num_input_tokens_seen": 13752264,
      "step": 24815
    },
    {
      "epoch": 435.4424778761062,
      "grad_norm": 8.375,
      "learning_rate": 1.576220981727895e-05,
      "loss": 0.9562,
      "num_input_tokens_seen": 13755176,
      "step": 24820
    },
    {
      "epoch": 435.53097345132744,
      "grad_norm": 7.21875,
      "learning_rate": 1.575308787375769e-05,
      "loss": 1.0298,
      "num_input_tokens_seen": 13758392,
      "step": 24825
    },
    {
      "epoch": 435.6194690265487,
      "grad_norm": 9.125,
      "learning_rate": 1.5743967356226492e-05,
      "loss": 0.8948,
      "num_input_tokens_seen": 13760808,
      "step": 24830
    },
    {
      "epoch": 435.7079646017699,
      "grad_norm": 8.625,
      "learning_rate": 1.5734848266091835e-05,
      "loss": 1.0506,
      "num_input_tokens_seen": 13763800,
      "step": 24835
    },
    {
      "epoch": 435.79646017699116,
      "grad_norm": 7.59375,
      "learning_rate": 1.572573060476001e-05,
      "loss": 0.9959,
      "num_input_tokens_seen": 13766680,
      "step": 24840
    },
    {
      "epoch": 435.88495575221236,
      "grad_norm": 8.1875,
      "learning_rate": 1.5716614373637085e-05,
      "loss": 1.3173,
      "num_input_tokens_seen": 13769592,
      "step": 24845
    },
    {
      "epoch": 435.9734513274336,
      "grad_norm": 11.625,
      "learning_rate": 1.570749957412887e-05,
      "loss": 0.9995,
      "num_input_tokens_seen": 13772312,
      "step": 24850
    },
    {
      "epoch": 436.05309734513276,
      "grad_norm": 7.6875,
      "learning_rate": 1.5698386207641013e-05,
      "loss": 1.1124,
      "num_input_tokens_seen": 13774608,
      "step": 24855
    },
    {
      "epoch": 436.14159292035396,
      "grad_norm": 8.0,
      "learning_rate": 1.5689274275578884e-05,
      "loss": 1.0184,
      "num_input_tokens_seen": 13777488,
      "step": 24860
    },
    {
      "epoch": 436.2300884955752,
      "grad_norm": 8.625,
      "learning_rate": 1.5680163779347667e-05,
      "loss": 0.9515,
      "num_input_tokens_seen": 13780688,
      "step": 24865
    },
    {
      "epoch": 436.3185840707965,
      "grad_norm": 15.4375,
      "learning_rate": 1.5671054720352327e-05,
      "loss": 1.0094,
      "num_input_tokens_seen": 13783312,
      "step": 24870
    },
    {
      "epoch": 436.4070796460177,
      "grad_norm": 8.4375,
      "learning_rate": 1.566194709999757e-05,
      "loss": 1.0375,
      "num_input_tokens_seen": 13785968,
      "step": 24875
    },
    {
      "epoch": 436.49557522123894,
      "grad_norm": 8.3125,
      "learning_rate": 1.5652840919687933e-05,
      "loss": 1.1774,
      "num_input_tokens_seen": 13788528,
      "step": 24880
    },
    {
      "epoch": 436.5840707964602,
      "grad_norm": 8.125,
      "learning_rate": 1.5643736180827676e-05,
      "loss": 1.175,
      "num_input_tokens_seen": 13791392,
      "step": 24885
    },
    {
      "epoch": 436.6725663716814,
      "grad_norm": 7.4375,
      "learning_rate": 1.5634632884820878e-05,
      "loss": 0.9853,
      "num_input_tokens_seen": 13793808,
      "step": 24890
    },
    {
      "epoch": 436.76106194690266,
      "grad_norm": 8.125,
      "learning_rate": 1.5625531033071395e-05,
      "loss": 0.9299,
      "num_input_tokens_seen": 13796704,
      "step": 24895
    },
    {
      "epoch": 436.8495575221239,
      "grad_norm": 8.0625,
      "learning_rate": 1.5616430626982828e-05,
      "loss": 1.1269,
      "num_input_tokens_seen": 13799568,
      "step": 24900
    },
    {
      "epoch": 436.9380530973451,
      "grad_norm": 8.0625,
      "learning_rate": 1.5607331667958575e-05,
      "loss": 1.2114,
      "num_input_tokens_seen": 13802240,
      "step": 24905
    },
    {
      "epoch": 437.01769911504425,
      "grad_norm": 7.875,
      "learning_rate": 1.5598234157401824e-05,
      "loss": 1.117,
      "num_input_tokens_seen": 13804920,
      "step": 24910
    },
    {
      "epoch": 437.1061946902655,
      "grad_norm": 11.5625,
      "learning_rate": 1.5589138096715503e-05,
      "loss": 1.1378,
      "num_input_tokens_seen": 13807704,
      "step": 24915
    },
    {
      "epoch": 437.1946902654867,
      "grad_norm": 7.75,
      "learning_rate": 1.5580043487302365e-05,
      "loss": 1.1555,
      "num_input_tokens_seen": 13810488,
      "step": 24920
    },
    {
      "epoch": 437.283185840708,
      "grad_norm": 7.15625,
      "learning_rate": 1.5570950330564888e-05,
      "loss": 1.2048,
      "num_input_tokens_seen": 13813544,
      "step": 24925
    },
    {
      "epoch": 437.37168141592923,
      "grad_norm": 8.1875,
      "learning_rate": 1.5561858627905367e-05,
      "loss": 1.0115,
      "num_input_tokens_seen": 13816616,
      "step": 24930
    },
    {
      "epoch": 437.46017699115043,
      "grad_norm": 8.0,
      "learning_rate": 1.5552768380725857e-05,
      "loss": 0.942,
      "num_input_tokens_seen": 13819352,
      "step": 24935
    },
    {
      "epoch": 437.5486725663717,
      "grad_norm": 8.4375,
      "learning_rate": 1.5543679590428183e-05,
      "loss": 1.0411,
      "num_input_tokens_seen": 13822536,
      "step": 24940
    },
    {
      "epoch": 437.6371681415929,
      "grad_norm": 7.5,
      "learning_rate": 1.5534592258413943e-05,
      "loss": 0.9714,
      "num_input_tokens_seen": 13825160,
      "step": 24945
    },
    {
      "epoch": 437.72566371681415,
      "grad_norm": 11.625,
      "learning_rate": 1.5525506386084538e-05,
      "loss": 0.8937,
      "num_input_tokens_seen": 13828152,
      "step": 24950
    },
    {
      "epoch": 437.8141592920354,
      "grad_norm": 10.625,
      "learning_rate": 1.55164219748411e-05,
      "loss": 1.2085,
      "num_input_tokens_seen": 13830648,
      "step": 24955
    },
    {
      "epoch": 437.9026548672566,
      "grad_norm": 6.78125,
      "learning_rate": 1.550733902608459e-05,
      "loss": 1.0239,
      "num_input_tokens_seen": 13833128,
      "step": 24960
    },
    {
      "epoch": 437.9911504424779,
      "grad_norm": 10.3125,
      "learning_rate": 1.549825754121568e-05,
      "loss": 1.0293,
      "num_input_tokens_seen": 13836024,
      "step": 24965
    },
    {
      "epoch": 438.070796460177,
      "grad_norm": 9.75,
      "learning_rate": 1.5489177521634864e-05,
      "loss": 0.9987,
      "num_input_tokens_seen": 13838080,
      "step": 24970
    },
    {
      "epoch": 438.1592920353982,
      "grad_norm": 8.6875,
      "learning_rate": 1.5480098968742402e-05,
      "loss": 1.0204,
      "num_input_tokens_seen": 13840784,
      "step": 24975
    },
    {
      "epoch": 438.24778761061947,
      "grad_norm": 7.78125,
      "learning_rate": 1.5471021883938304e-05,
      "loss": 1.0936,
      "num_input_tokens_seen": 13843840,
      "step": 24980
    },
    {
      "epoch": 438.3362831858407,
      "grad_norm": 10.3125,
      "learning_rate": 1.546194626862238e-05,
      "loss": 1.0211,
      "num_input_tokens_seen": 13847104,
      "step": 24985
    },
    {
      "epoch": 438.42477876106193,
      "grad_norm": 11.5625,
      "learning_rate": 1.5452872124194216e-05,
      "loss": 0.9443,
      "num_input_tokens_seen": 13850128,
      "step": 24990
    },
    {
      "epoch": 438.5132743362832,
      "grad_norm": 8.3125,
      "learning_rate": 1.5443799452053136e-05,
      "loss": 1.2089,
      "num_input_tokens_seen": 13853248,
      "step": 24995
    },
    {
      "epoch": 438.60176991150445,
      "grad_norm": 7.59375,
      "learning_rate": 1.543472825359828e-05,
      "loss": 1.107,
      "num_input_tokens_seen": 13856080,
      "step": 25000
    },
    {
      "epoch": 438.60176991150445,
      "eval_loss": 1.0240559577941895,
      "eval_runtime": 0.9351,
      "eval_samples_per_second": 26.736,
      "eval_steps_per_second": 13.903,
      "num_input_tokens_seen": 13856080,
      "step": 25000
    },
    {
      "epoch": 438.69026548672565,
      "grad_norm": 8.9375,
      "learning_rate": 1.5425658530228522e-05,
      "loss": 0.9951,
      "num_input_tokens_seen": 13858768,
      "step": 25005
    },
    {
      "epoch": 438.7787610619469,
      "grad_norm": 8.6875,
      "learning_rate": 1.5416590283342546e-05,
      "loss": 0.9977,
      "num_input_tokens_seen": 13861456,
      "step": 25010
    },
    {
      "epoch": 438.86725663716817,
      "grad_norm": 6.40625,
      "learning_rate": 1.5407523514338783e-05,
      "loss": 0.9834,
      "num_input_tokens_seen": 13863984,
      "step": 25015
    },
    {
      "epoch": 438.95575221238937,
      "grad_norm": 10.625,
      "learning_rate": 1.539845822461543e-05,
      "loss": 1.0665,
      "num_input_tokens_seen": 13866688,
      "step": 25020
    },
    {
      "epoch": 439.0353982300885,
      "grad_norm": 5.53125,
      "learning_rate": 1.538939441557048e-05,
      "loss": 1.0918,
      "num_input_tokens_seen": 13869120,
      "step": 25025
    },
    {
      "epoch": 439.12389380530976,
      "grad_norm": 8.6875,
      "learning_rate": 1.5380332088601696e-05,
      "loss": 1.1368,
      "num_input_tokens_seen": 13871920,
      "step": 25030
    },
    {
      "epoch": 439.21238938053096,
      "grad_norm": 7.96875,
      "learning_rate": 1.537127124510658e-05,
      "loss": 1.1225,
      "num_input_tokens_seen": 13874416,
      "step": 25035
    },
    {
      "epoch": 439.3008849557522,
      "grad_norm": 7.28125,
      "learning_rate": 1.5362211886482457e-05,
      "loss": 1.1192,
      "num_input_tokens_seen": 13876896,
      "step": 25040
    },
    {
      "epoch": 439.3893805309734,
      "grad_norm": 10.875,
      "learning_rate": 1.5353154014126363e-05,
      "loss": 0.9617,
      "num_input_tokens_seen": 13879344,
      "step": 25045
    },
    {
      "epoch": 439.4778761061947,
      "grad_norm": 8.9375,
      "learning_rate": 1.534409762943515e-05,
      "loss": 1.0321,
      "num_input_tokens_seen": 13882016,
      "step": 25050
    },
    {
      "epoch": 439.56637168141594,
      "grad_norm": 6.6875,
      "learning_rate": 1.5335042733805438e-05,
      "loss": 1.0137,
      "num_input_tokens_seen": 13884928,
      "step": 25055
    },
    {
      "epoch": 439.65486725663715,
      "grad_norm": 8.375,
      "learning_rate": 1.532598932863358e-05,
      "loss": 0.9506,
      "num_input_tokens_seen": 13887984,
      "step": 25060
    },
    {
      "epoch": 439.7433628318584,
      "grad_norm": 8.625,
      "learning_rate": 1.531693741531574e-05,
      "loss": 0.9454,
      "num_input_tokens_seen": 13890736,
      "step": 25065
    },
    {
      "epoch": 439.83185840707966,
      "grad_norm": 7.625,
      "learning_rate": 1.5307886995247844e-05,
      "loss": 1.162,
      "num_input_tokens_seen": 13893632,
      "step": 25070
    },
    {
      "epoch": 439.92035398230087,
      "grad_norm": 7.4375,
      "learning_rate": 1.529883806982557e-05,
      "loss": 1.1661,
      "num_input_tokens_seen": 13896992,
      "step": 25075
    },
    {
      "epoch": 440.0,
      "grad_norm": 9.5625,
      "learning_rate": 1.5289790640444376e-05,
      "loss": 1.0134,
      "num_input_tokens_seen": 13899512,
      "step": 25080
    },
    {
      "epoch": 440.08849557522126,
      "grad_norm": 8.625,
      "learning_rate": 1.5280744708499494e-05,
      "loss": 1.1122,
      "num_input_tokens_seen": 13902552,
      "step": 25085
    },
    {
      "epoch": 440.17699115044246,
      "grad_norm": 9.625,
      "learning_rate": 1.527170027538591e-05,
      "loss": 1.0759,
      "num_input_tokens_seen": 13905304,
      "step": 25090
    },
    {
      "epoch": 440.2654867256637,
      "grad_norm": 8.1875,
      "learning_rate": 1.5262657342498407e-05,
      "loss": 1.1186,
      "num_input_tokens_seen": 13908408,
      "step": 25095
    },
    {
      "epoch": 440.353982300885,
      "grad_norm": 7.0625,
      "learning_rate": 1.52536159112315e-05,
      "loss": 1.067,
      "num_input_tokens_seen": 13911064,
      "step": 25100
    },
    {
      "epoch": 440.4424778761062,
      "grad_norm": 9.0625,
      "learning_rate": 1.5244575982979497e-05,
      "loss": 1.0055,
      "num_input_tokens_seen": 13913560,
      "step": 25105
    },
    {
      "epoch": 440.53097345132744,
      "grad_norm": 7.5,
      "learning_rate": 1.5235537559136487e-05,
      "loss": 1.0587,
      "num_input_tokens_seen": 13916680,
      "step": 25110
    },
    {
      "epoch": 440.6194690265487,
      "grad_norm": 7.21875,
      "learning_rate": 1.5226500641096286e-05,
      "loss": 0.9643,
      "num_input_tokens_seen": 13919224,
      "step": 25115
    },
    {
      "epoch": 440.7079646017699,
      "grad_norm": 8.3125,
      "learning_rate": 1.5217465230252509e-05,
      "loss": 1.2083,
      "num_input_tokens_seen": 13921704,
      "step": 25120
    },
    {
      "epoch": 440.79646017699116,
      "grad_norm": 10.6875,
      "learning_rate": 1.5208431327998523e-05,
      "loss": 0.947,
      "num_input_tokens_seen": 13924776,
      "step": 25125
    },
    {
      "epoch": 440.88495575221236,
      "grad_norm": 11.6875,
      "learning_rate": 1.5199398935727477e-05,
      "loss": 1.12,
      "num_input_tokens_seen": 13927336,
      "step": 25130
    },
    {
      "epoch": 440.9734513274336,
      "grad_norm": 7.4375,
      "learning_rate": 1.5190368054832282e-05,
      "loss": 0.9646,
      "num_input_tokens_seen": 13930184,
      "step": 25135
    },
    {
      "epoch": 441.05309734513276,
      "grad_norm": 7.75,
      "learning_rate": 1.5181338686705601e-05,
      "loss": 0.9882,
      "num_input_tokens_seen": 13932752,
      "step": 25140
    },
    {
      "epoch": 441.14159292035396,
      "grad_norm": 9.125,
      "learning_rate": 1.5172310832739889e-05,
      "loss": 0.9224,
      "num_input_tokens_seen": 13935760,
      "step": 25145
    },
    {
      "epoch": 441.2300884955752,
      "grad_norm": 7.90625,
      "learning_rate": 1.5163284494327346e-05,
      "loss": 0.9553,
      "num_input_tokens_seen": 13938336,
      "step": 25150
    },
    {
      "epoch": 441.3185840707965,
      "grad_norm": 7.625,
      "learning_rate": 1.5154259672859952e-05,
      "loss": 1.0525,
      "num_input_tokens_seen": 13941136,
      "step": 25155
    },
    {
      "epoch": 441.4070796460177,
      "grad_norm": 9.0,
      "learning_rate": 1.5145236369729452e-05,
      "loss": 1.0737,
      "num_input_tokens_seen": 13943824,
      "step": 25160
    },
    {
      "epoch": 441.49557522123894,
      "grad_norm": 7.65625,
      "learning_rate": 1.5136214586327335e-05,
      "loss": 1.1984,
      "num_input_tokens_seen": 13946768,
      "step": 25165
    },
    {
      "epoch": 441.5840707964602,
      "grad_norm": 8.8125,
      "learning_rate": 1.5127194324044885e-05,
      "loss": 0.9524,
      "num_input_tokens_seen": 13949296,
      "step": 25170
    },
    {
      "epoch": 441.6725663716814,
      "grad_norm": 12.4375,
      "learning_rate": 1.5118175584273148e-05,
      "loss": 1.2641,
      "num_input_tokens_seen": 13951840,
      "step": 25175
    },
    {
      "epoch": 441.76106194690266,
      "grad_norm": 7.90625,
      "learning_rate": 1.5109158368402909e-05,
      "loss": 1.2071,
      "num_input_tokens_seen": 13954656,
      "step": 25180
    },
    {
      "epoch": 441.8495575221239,
      "grad_norm": 7.375,
      "learning_rate": 1.5100142677824753e-05,
      "loss": 1.0428,
      "num_input_tokens_seen": 13957440,
      "step": 25185
    },
    {
      "epoch": 441.9380530973451,
      "grad_norm": 9.6875,
      "learning_rate": 1.509112851392901e-05,
      "loss": 1.0781,
      "num_input_tokens_seen": 13960592,
      "step": 25190
    },
    {
      "epoch": 442.01769911504425,
      "grad_norm": 7.03125,
      "learning_rate": 1.5082115878105763e-05,
      "loss": 0.7866,
      "num_input_tokens_seen": 13963688,
      "step": 25195
    },
    {
      "epoch": 442.1061946902655,
      "grad_norm": 9.3125,
      "learning_rate": 1.5073104771744892e-05,
      "loss": 1.0065,
      "num_input_tokens_seen": 13966552,
      "step": 25200
    },
    {
      "epoch": 442.1061946902655,
      "eval_loss": 1.0303720235824585,
      "eval_runtime": 0.934,
      "eval_samples_per_second": 26.768,
      "eval_steps_per_second": 13.919,
      "num_input_tokens_seen": 13966552,
      "step": 25200
    },
    {
      "epoch": 442.1946902654867,
      "grad_norm": 7.0,
      "learning_rate": 1.5064095196236006e-05,
      "loss": 1.016,
      "num_input_tokens_seen": 13968968,
      "step": 25205
    },
    {
      "epoch": 442.283185840708,
      "grad_norm": 16.125,
      "learning_rate": 1.50550871529685e-05,
      "loss": 0.9831,
      "num_input_tokens_seen": 13971544,
      "step": 25210
    },
    {
      "epoch": 442.37168141592923,
      "grad_norm": 8.1875,
      "learning_rate": 1.5046080643331546e-05,
      "loss": 1.0849,
      "num_input_tokens_seen": 13975176,
      "step": 25215
    },
    {
      "epoch": 442.46017699115043,
      "grad_norm": 8.625,
      "learning_rate": 1.5037075668714028e-05,
      "loss": 1.3976,
      "num_input_tokens_seen": 13977752,
      "step": 25220
    },
    {
      "epoch": 442.5486725663717,
      "grad_norm": 9.0,
      "learning_rate": 1.5028072230504656e-05,
      "loss": 0.9064,
      "num_input_tokens_seen": 13981224,
      "step": 25225
    },
    {
      "epoch": 442.6371681415929,
      "grad_norm": 8.4375,
      "learning_rate": 1.5019070330091861e-05,
      "loss": 1.2169,
      "num_input_tokens_seen": 13984056,
      "step": 25230
    },
    {
      "epoch": 442.72566371681415,
      "grad_norm": 10.125,
      "learning_rate": 1.5010069968863843e-05,
      "loss": 0.9579,
      "num_input_tokens_seen": 13986440,
      "step": 25235
    },
    {
      "epoch": 442.8141592920354,
      "grad_norm": 8.0,
      "learning_rate": 1.5001071148208584e-05,
      "loss": 0.992,
      "num_input_tokens_seen": 13989112,
      "step": 25240
    },
    {
      "epoch": 442.9026548672566,
      "grad_norm": 7.09375,
      "learning_rate": 1.49920738695138e-05,
      "loss": 0.9839,
      "num_input_tokens_seen": 13991368,
      "step": 25245
    },
    {
      "epoch": 442.9911504424779,
      "grad_norm": 7.0625,
      "learning_rate": 1.4983078134166995e-05,
      "loss": 1.0516,
      "num_input_tokens_seen": 13994200,
      "step": 25250
    },
    {
      "epoch": 443.070796460177,
      "grad_norm": 7.4375,
      "learning_rate": 1.4974083943555428e-05,
      "loss": 1.1527,
      "num_input_tokens_seen": 13996384,
      "step": 25255
    },
    {
      "epoch": 443.1592920353982,
      "grad_norm": 8.25,
      "learning_rate": 1.496509129906611e-05,
      "loss": 1.0266,
      "num_input_tokens_seen": 13998976,
      "step": 25260
    },
    {
      "epoch": 443.24778761061947,
      "grad_norm": 10.3125,
      "learning_rate": 1.4956100202085809e-05,
      "loss": 0.9894,
      "num_input_tokens_seen": 14001872,
      "step": 25265
    },
    {
      "epoch": 443.3362831858407,
      "grad_norm": 8.4375,
      "learning_rate": 1.4947110654001093e-05,
      "loss": 1.0193,
      "num_input_tokens_seen": 14005072,
      "step": 25270
    },
    {
      "epoch": 443.42477876106193,
      "grad_norm": 7.46875,
      "learning_rate": 1.4938122656198234e-05,
      "loss": 1.1811,
      "num_input_tokens_seen": 14007712,
      "step": 25275
    },
    {
      "epoch": 443.5132743362832,
      "grad_norm": 8.5625,
      "learning_rate": 1.4929136210063316e-05,
      "loss": 1.0802,
      "num_input_tokens_seen": 14010624,
      "step": 25280
    },
    {
      "epoch": 443.60176991150445,
      "grad_norm": 12.0,
      "learning_rate": 1.4920151316982146e-05,
      "loss": 1.1162,
      "num_input_tokens_seen": 14013584,
      "step": 25285
    },
    {
      "epoch": 443.69026548672565,
      "grad_norm": 7.46875,
      "learning_rate": 1.4911167978340312e-05,
      "loss": 1.0103,
      "num_input_tokens_seen": 14016224,
      "step": 25290
    },
    {
      "epoch": 443.7787610619469,
      "grad_norm": 8.0,
      "learning_rate": 1.4902186195523166e-05,
      "loss": 1.1586,
      "num_input_tokens_seen": 14019248,
      "step": 25295
    },
    {
      "epoch": 443.86725663716817,
      "grad_norm": 9.0625,
      "learning_rate": 1.4893205969915805e-05,
      "loss": 1.0128,
      "num_input_tokens_seen": 14021840,
      "step": 25300
    },
    {
      "epoch": 443.95575221238937,
      "grad_norm": 9.0,
      "learning_rate": 1.4884227302903086e-05,
      "loss": 1.0343,
      "num_input_tokens_seen": 14025168,
      "step": 25305
    },
    {
      "epoch": 444.0353982300885,
      "grad_norm": 8.5625,
      "learning_rate": 1.4875250195869653e-05,
      "loss": 0.9219,
      "num_input_tokens_seen": 14027328,
      "step": 25310
    },
    {
      "epoch": 444.12389380530976,
      "grad_norm": 13.0625,
      "learning_rate": 1.4866274650199862e-05,
      "loss": 1.1423,
      "num_input_tokens_seen": 14030368,
      "step": 25315
    },
    {
      "epoch": 444.21238938053096,
      "grad_norm": 7.53125,
      "learning_rate": 1.485730066727788e-05,
      "loss": 0.9437,
      "num_input_tokens_seen": 14032656,
      "step": 25320
    },
    {
      "epoch": 444.3008849557522,
      "grad_norm": 12.0,
      "learning_rate": 1.4848328248487586e-05,
      "loss": 1.18,
      "num_input_tokens_seen": 14035360,
      "step": 25325
    },
    {
      "epoch": 444.3893805309734,
      "grad_norm": 8.9375,
      "learning_rate": 1.4839357395212656e-05,
      "loss": 1.0246,
      "num_input_tokens_seen": 14038128,
      "step": 25330
    },
    {
      "epoch": 444.4778761061947,
      "grad_norm": 10.25,
      "learning_rate": 1.4830388108836502e-05,
      "loss": 1.0803,
      "num_input_tokens_seen": 14041088,
      "step": 25335
    },
    {
      "epoch": 444.56637168141594,
      "grad_norm": 7.71875,
      "learning_rate": 1.4821420390742299e-05,
      "loss": 1.08,
      "num_input_tokens_seen": 14044032,
      "step": 25340
    },
    {
      "epoch": 444.65486725663715,
      "grad_norm": 6.9375,
      "learning_rate": 1.4812454242312979e-05,
      "loss": 1.1141,
      "num_input_tokens_seen": 14046496,
      "step": 25345
    },
    {
      "epoch": 444.7433628318584,
      "grad_norm": 7.28125,
      "learning_rate": 1.4803489664931253e-05,
      "loss": 0.9806,
      "num_input_tokens_seen": 14049072,
      "step": 25350
    },
    {
      "epoch": 444.83185840707966,
      "grad_norm": 9.375,
      "learning_rate": 1.4794526659979544e-05,
      "loss": 0.7905,
      "num_input_tokens_seen": 14051808,
      "step": 25355
    },
    {
      "epoch": 444.92035398230087,
      "grad_norm": 9.875,
      "learning_rate": 1.4785565228840086e-05,
      "loss": 1.1543,
      "num_input_tokens_seen": 14054624,
      "step": 25360
    },
    {
      "epoch": 445.0,
      "grad_norm": 3.359375,
      "learning_rate": 1.4776605372894819e-05,
      "loss": 1.0279,
      "num_input_tokens_seen": 14056832,
      "step": 25365
    },
    {
      "epoch": 445.08849557522126,
      "grad_norm": 7.1875,
      "learning_rate": 1.4767647093525488e-05,
      "loss": 1.0025,
      "num_input_tokens_seen": 14059856,
      "step": 25370
    },
    {
      "epoch": 445.17699115044246,
      "grad_norm": 6.625,
      "learning_rate": 1.4758690392113566e-05,
      "loss": 1.1951,
      "num_input_tokens_seen": 14062352,
      "step": 25375
    },
    {
      "epoch": 445.2654867256637,
      "grad_norm": 7.34375,
      "learning_rate": 1.4749735270040276e-05,
      "loss": 0.9546,
      "num_input_tokens_seen": 14065312,
      "step": 25380
    },
    {
      "epoch": 445.353982300885,
      "grad_norm": 11.1875,
      "learning_rate": 1.4740781728686623e-05,
      "loss": 1.0527,
      "num_input_tokens_seen": 14068448,
      "step": 25385
    },
    {
      "epoch": 445.4424778761062,
      "grad_norm": 9.5,
      "learning_rate": 1.4731829769433358e-05,
      "loss": 1.0522,
      "num_input_tokens_seen": 14070944,
      "step": 25390
    },
    {
      "epoch": 445.53097345132744,
      "grad_norm": 8.125,
      "learning_rate": 1.4722879393660976e-05,
      "loss": 1.0919,
      "num_input_tokens_seen": 14074144,
      "step": 25395
    },
    {
      "epoch": 445.6194690265487,
      "grad_norm": 7.96875,
      "learning_rate": 1.4713930602749748e-05,
      "loss": 0.8437,
      "num_input_tokens_seen": 14076912,
      "step": 25400
    },
    {
      "epoch": 445.6194690265487,
      "eval_loss": 1.0310944318771362,
      "eval_runtime": 0.9354,
      "eval_samples_per_second": 26.726,
      "eval_steps_per_second": 13.898,
      "num_input_tokens_seen": 14076912,
      "step": 25400
    },
    {
      "epoch": 445.7079646017699,
      "grad_norm": 8.3125,
      "learning_rate": 1.470498339807968e-05,
      "loss": 0.9811,
      "num_input_tokens_seen": 14079296,
      "step": 25405
    },
    {
      "epoch": 445.79646017699116,
      "grad_norm": 9.3125,
      "learning_rate": 1.4696037781030542e-05,
      "loss": 1.2593,
      "num_input_tokens_seen": 14082224,
      "step": 25410
    },
    {
      "epoch": 445.88495575221236,
      "grad_norm": 8.375,
      "learning_rate": 1.4687093752981876e-05,
      "loss": 1.2647,
      "num_input_tokens_seen": 14084688,
      "step": 25415
    },
    {
      "epoch": 445.9734513274336,
      "grad_norm": 9.375,
      "learning_rate": 1.4678151315312943e-05,
      "loss": 0.903,
      "num_input_tokens_seen": 14087696,
      "step": 25420
    },
    {
      "epoch": 446.05309734513276,
      "grad_norm": 8.125,
      "learning_rate": 1.4669210469402789e-05,
      "loss": 1.0651,
      "num_input_tokens_seen": 14090136,
      "step": 25425
    },
    {
      "epoch": 446.14159292035396,
      "grad_norm": 10.8125,
      "learning_rate": 1.4660271216630218e-05,
      "loss": 0.9413,
      "num_input_tokens_seen": 14092936,
      "step": 25430
    },
    {
      "epoch": 446.2300884955752,
      "grad_norm": 7.78125,
      "learning_rate": 1.4651333558373748e-05,
      "loss": 1.2158,
      "num_input_tokens_seen": 14095512,
      "step": 25435
    },
    {
      "epoch": 446.3185840707965,
      "grad_norm": 8.0,
      "learning_rate": 1.4642397496011707e-05,
      "loss": 1.0314,
      "num_input_tokens_seen": 14098424,
      "step": 25440
    },
    {
      "epoch": 446.4070796460177,
      "grad_norm": 8.375,
      "learning_rate": 1.4633463030922129e-05,
      "loss": 0.9972,
      "num_input_tokens_seen": 14101352,
      "step": 25445
    },
    {
      "epoch": 446.49557522123894,
      "grad_norm": 7.03125,
      "learning_rate": 1.462453016448282e-05,
      "loss": 0.9461,
      "num_input_tokens_seen": 14104152,
      "step": 25450
    },
    {
      "epoch": 446.5840707964602,
      "grad_norm": 7.625,
      "learning_rate": 1.4615598898071354e-05,
      "loss": 1.0455,
      "num_input_tokens_seen": 14107288,
      "step": 25455
    },
    {
      "epoch": 446.6725663716814,
      "grad_norm": 8.4375,
      "learning_rate": 1.4606669233065026e-05,
      "loss": 1.0591,
      "num_input_tokens_seen": 14109800,
      "step": 25460
    },
    {
      "epoch": 446.76106194690266,
      "grad_norm": 8.25,
      "learning_rate": 1.4597741170840914e-05,
      "loss": 0.9251,
      "num_input_tokens_seen": 14112632,
      "step": 25465
    },
    {
      "epoch": 446.8495575221239,
      "grad_norm": 8.125,
      "learning_rate": 1.4588814712775853e-05,
      "loss": 1.3361,
      "num_input_tokens_seen": 14115448,
      "step": 25470
    },
    {
      "epoch": 446.9380530973451,
      "grad_norm": 8.0,
      "learning_rate": 1.4579889860246382e-05,
      "loss": 1.0219,
      "num_input_tokens_seen": 14118152,
      "step": 25475
    },
    {
      "epoch": 447.01769911504425,
      "grad_norm": 6.125,
      "learning_rate": 1.457096661462885e-05,
      "loss": 1.0056,
      "num_input_tokens_seen": 14120488,
      "step": 25480
    },
    {
      "epoch": 447.1061946902655,
      "grad_norm": 6.25,
      "learning_rate": 1.4562044977299322e-05,
      "loss": 1.0907,
      "num_input_tokens_seen": 14123144,
      "step": 25485
    },
    {
      "epoch": 447.1946902654867,
      "grad_norm": 7.375,
      "learning_rate": 1.4553124949633623e-05,
      "loss": 1.055,
      "num_input_tokens_seen": 14125672,
      "step": 25490
    },
    {
      "epoch": 447.283185840708,
      "grad_norm": 8.25,
      "learning_rate": 1.4544206533007354e-05,
      "loss": 0.9799,
      "num_input_tokens_seen": 14128712,
      "step": 25495
    },
    {
      "epoch": 447.37168141592923,
      "grad_norm": 7.75,
      "learning_rate": 1.4535289728795821e-05,
      "loss": 0.9496,
      "num_input_tokens_seen": 14131960,
      "step": 25500
    },
    {
      "epoch": 447.46017699115043,
      "grad_norm": 10.0,
      "learning_rate": 1.4526374538374132e-05,
      "loss": 1.3232,
      "num_input_tokens_seen": 14134552,
      "step": 25505
    },
    {
      "epoch": 447.5486725663717,
      "grad_norm": 7.40625,
      "learning_rate": 1.4517460963117097e-05,
      "loss": 0.9738,
      "num_input_tokens_seen": 14137064,
      "step": 25510
    },
    {
      "epoch": 447.6371681415929,
      "grad_norm": 10.9375,
      "learning_rate": 1.4508549004399314e-05,
      "loss": 1.1281,
      "num_input_tokens_seen": 14139976,
      "step": 25515
    },
    {
      "epoch": 447.72566371681415,
      "grad_norm": 7.125,
      "learning_rate": 1.449963866359513e-05,
      "loss": 1.0538,
      "num_input_tokens_seen": 14142776,
      "step": 25520
    },
    {
      "epoch": 447.8141592920354,
      "grad_norm": 8.75,
      "learning_rate": 1.4490729942078607e-05,
      "loss": 1.1335,
      "num_input_tokens_seen": 14145608,
      "step": 25525
    },
    {
      "epoch": 447.9026548672566,
      "grad_norm": 6.59375,
      "learning_rate": 1.4481822841223608e-05,
      "loss": 0.9111,
      "num_input_tokens_seen": 14148488,
      "step": 25530
    },
    {
      "epoch": 447.9911504424779,
      "grad_norm": 8.3125,
      "learning_rate": 1.4472917362403704e-05,
      "loss": 1.0609,
      "num_input_tokens_seen": 14151528,
      "step": 25535
    },
    {
      "epoch": 448.070796460177,
      "grad_norm": 8.0,
      "learning_rate": 1.4464013506992224e-05,
      "loss": 1.2588,
      "num_input_tokens_seen": 14153816,
      "step": 25540
    },
    {
      "epoch": 448.1592920353982,
      "grad_norm": 9.1875,
      "learning_rate": 1.4455111276362277e-05,
      "loss": 1.0357,
      "num_input_tokens_seen": 14157496,
      "step": 25545
    },
    {
      "epoch": 448.24778761061947,
      "grad_norm": 8.125,
      "learning_rate": 1.4446210671886676e-05,
      "loss": 1.0271,
      "num_input_tokens_seen": 14160264,
      "step": 25550
    },
    {
      "epoch": 448.3362831858407,
      "grad_norm": 6.625,
      "learning_rate": 1.4437311694938015e-05,
      "loss": 0.9186,
      "num_input_tokens_seen": 14163256,
      "step": 25555
    },
    {
      "epoch": 448.42477876106193,
      "grad_norm": 7.96875,
      "learning_rate": 1.442841434688864e-05,
      "loss": 0.9303,
      "num_input_tokens_seen": 14165848,
      "step": 25560
    },
    {
      "epoch": 448.5132743362832,
      "grad_norm": 10.6875,
      "learning_rate": 1.4419518629110615e-05,
      "loss": 1.0441,
      "num_input_tokens_seen": 14169096,
      "step": 25565
    },
    {
      "epoch": 448.60176991150445,
      "grad_norm": 8.875,
      "learning_rate": 1.4410624542975778e-05,
      "loss": 1.1502,
      "num_input_tokens_seen": 14171336,
      "step": 25570
    },
    {
      "epoch": 448.69026548672565,
      "grad_norm": 8.625,
      "learning_rate": 1.4401732089855724e-05,
      "loss": 1.0205,
      "num_input_tokens_seen": 14173928,
      "step": 25575
    },
    {
      "epoch": 448.7787610619469,
      "grad_norm": 8.8125,
      "learning_rate": 1.4392841271121754e-05,
      "loss": 1.2563,
      "num_input_tokens_seen": 14176936,
      "step": 25580
    },
    {
      "epoch": 448.86725663716817,
      "grad_norm": 6.8125,
      "learning_rate": 1.438395208814497e-05,
      "loss": 1.0925,
      "num_input_tokens_seen": 14179368,
      "step": 25585
    },
    {
      "epoch": 448.95575221238937,
      "grad_norm": 9.5625,
      "learning_rate": 1.4375064542296174e-05,
      "loss": 0.9302,
      "num_input_tokens_seen": 14182136,
      "step": 25590
    },
    {
      "epoch": 449.0353982300885,
      "grad_norm": 8.1875,
      "learning_rate": 1.4366178634945946e-05,
      "loss": 1.0395,
      "num_input_tokens_seen": 14184072,
      "step": 25595
    },
    {
      "epoch": 449.12389380530976,
      "grad_norm": 6.90625,
      "learning_rate": 1.4357294367464616e-05,
      "loss": 1.005,
      "num_input_tokens_seen": 14187144,
      "step": 25600
    },
    {
      "epoch": 449.12389380530976,
      "eval_loss": 1.0319925546646118,
      "eval_runtime": 0.932,
      "eval_samples_per_second": 26.824,
      "eval_steps_per_second": 13.948,
      "num_input_tokens_seen": 14187144,
      "step": 25600
    },
    {
      "epoch": 449.21238938053096,
      "grad_norm": 8.25,
      "learning_rate": 1.434841174122224e-05,
      "loss": 1.0061,
      "num_input_tokens_seen": 14189576,
      "step": 25605
    },
    {
      "epoch": 449.3008849557522,
      "grad_norm": 8.0,
      "learning_rate": 1.4339530757588615e-05,
      "loss": 1.0467,
      "num_input_tokens_seen": 14192024,
      "step": 25610
    },
    {
      "epoch": 449.3893805309734,
      "grad_norm": 8.5,
      "learning_rate": 1.433065141793333e-05,
      "loss": 1.1608,
      "num_input_tokens_seen": 14194584,
      "step": 25615
    },
    {
      "epoch": 449.4778761061947,
      "grad_norm": 6.09375,
      "learning_rate": 1.4321773723625665e-05,
      "loss": 0.8972,
      "num_input_tokens_seen": 14197432,
      "step": 25620
    },
    {
      "epoch": 449.56637168141594,
      "grad_norm": 7.65625,
      "learning_rate": 1.4312897676034693e-05,
      "loss": 1.1245,
      "num_input_tokens_seen": 14200184,
      "step": 25625
    },
    {
      "epoch": 449.65486725663715,
      "grad_norm": 11.1875,
      "learning_rate": 1.4304023276529188e-05,
      "loss": 1.1186,
      "num_input_tokens_seen": 14203592,
      "step": 25630
    },
    {
      "epoch": 449.7433628318584,
      "grad_norm": 8.5,
      "learning_rate": 1.4295150526477712e-05,
      "loss": 1.0559,
      "num_input_tokens_seen": 14206472,
      "step": 25635
    },
    {
      "epoch": 449.83185840707966,
      "grad_norm": 8.3125,
      "learning_rate": 1.4286279427248562e-05,
      "loss": 1.069,
      "num_input_tokens_seen": 14209640,
      "step": 25640
    },
    {
      "epoch": 449.92035398230087,
      "grad_norm": 8.625,
      "learning_rate": 1.4277409980209747e-05,
      "loss": 1.1907,
      "num_input_tokens_seen": 14212408,
      "step": 25645
    },
    {
      "epoch": 450.0,
      "grad_norm": 8.3125,
      "learning_rate": 1.4268542186729061e-05,
      "loss": 0.991,
      "num_input_tokens_seen": 14214880,
      "step": 25650
    },
    {
      "epoch": 450.08849557522126,
      "grad_norm": 8.625,
      "learning_rate": 1.4259676048174043e-05,
      "loss": 1.0775,
      "num_input_tokens_seen": 14217920,
      "step": 25655
    },
    {
      "epoch": 450.17699115044246,
      "grad_norm": 10.875,
      "learning_rate": 1.4250811565911937e-05,
      "loss": 0.9434,
      "num_input_tokens_seen": 14221040,
      "step": 25660
    },
    {
      "epoch": 450.2654867256637,
      "grad_norm": 16.0,
      "learning_rate": 1.4241948741309782e-05,
      "loss": 1.1248,
      "num_input_tokens_seen": 14223856,
      "step": 25665
    },
    {
      "epoch": 450.353982300885,
      "grad_norm": 14.125,
      "learning_rate": 1.4233087575734317e-05,
      "loss": 1.1138,
      "num_input_tokens_seen": 14226656,
      "step": 25670
    },
    {
      "epoch": 450.4424778761062,
      "grad_norm": 6.46875,
      "learning_rate": 1.422422807055206e-05,
      "loss": 0.8498,
      "num_input_tokens_seen": 14229408,
      "step": 25675
    },
    {
      "epoch": 450.53097345132744,
      "grad_norm": 8.375,
      "learning_rate": 1.4215370227129243e-05,
      "loss": 1.0923,
      "num_input_tokens_seen": 14232144,
      "step": 25680
    },
    {
      "epoch": 450.6194690265487,
      "grad_norm": 9.0,
      "learning_rate": 1.4206514046831876e-05,
      "loss": 0.9698,
      "num_input_tokens_seen": 14234960,
      "step": 25685
    },
    {
      "epoch": 450.7079646017699,
      "grad_norm": 7.4375,
      "learning_rate": 1.419765953102567e-05,
      "loss": 1.0756,
      "num_input_tokens_seen": 14237520,
      "step": 25690
    },
    {
      "epoch": 450.79646017699116,
      "grad_norm": 8.4375,
      "learning_rate": 1.4188806681076125e-05,
      "loss": 1.1264,
      "num_input_tokens_seen": 14240112,
      "step": 25695
    },
    {
      "epoch": 450.88495575221236,
      "grad_norm": 9.625,
      "learning_rate": 1.4179955498348443e-05,
      "loss": 1.167,
      "num_input_tokens_seen": 14243216,
      "step": 25700
    },
    {
      "epoch": 450.9734513274336,
      "grad_norm": 8.9375,
      "learning_rate": 1.4171105984207605e-05,
      "loss": 1.026,
      "num_input_tokens_seen": 14245808,
      "step": 25705
    },
    {
      "epoch": 451.05309734513276,
      "grad_norm": 8.0625,
      "learning_rate": 1.4162258140018304e-05,
      "loss": 1.4368,
      "num_input_tokens_seen": 14248024,
      "step": 25710
    },
    {
      "epoch": 451.14159292035396,
      "grad_norm": 9.5625,
      "learning_rate": 1.4153411967144986e-05,
      "loss": 1.2836,
      "num_input_tokens_seen": 14250552,
      "step": 25715
    },
    {
      "epoch": 451.2300884955752,
      "grad_norm": 8.625,
      "learning_rate": 1.4144567466951864e-05,
      "loss": 1.0894,
      "num_input_tokens_seen": 14252872,
      "step": 25720
    },
    {
      "epoch": 451.3185840707965,
      "grad_norm": 7.21875,
      "learning_rate": 1.4135724640802844e-05,
      "loss": 0.9909,
      "num_input_tokens_seen": 14256056,
      "step": 25725
    },
    {
      "epoch": 451.4070796460177,
      "grad_norm": 8.25,
      "learning_rate": 1.4126883490061615e-05,
      "loss": 1.1747,
      "num_input_tokens_seen": 14258744,
      "step": 25730
    },
    {
      "epoch": 451.49557522123894,
      "grad_norm": 11.5625,
      "learning_rate": 1.4118044016091603e-05,
      "loss": 1.0719,
      "num_input_tokens_seen": 14262152,
      "step": 25735
    },
    {
      "epoch": 451.5840707964602,
      "grad_norm": 7.5,
      "learning_rate": 1.410920622025594e-05,
      "loss": 0.8879,
      "num_input_tokens_seen": 14264824,
      "step": 25740
    },
    {
      "epoch": 451.6725663716814,
      "grad_norm": 7.6875,
      "learning_rate": 1.4100370103917554e-05,
      "loss": 1.0687,
      "num_input_tokens_seen": 14267832,
      "step": 25745
    },
    {
      "epoch": 451.76106194690266,
      "grad_norm": 6.53125,
      "learning_rate": 1.409153566843907e-05,
      "loss": 0.7312,
      "num_input_tokens_seen": 14270472,
      "step": 25750
    },
    {
      "epoch": 451.8495575221239,
      "grad_norm": 10.8125,
      "learning_rate": 1.408270291518286e-05,
      "loss": 1.0806,
      "num_input_tokens_seen": 14273448,
      "step": 25755
    },
    {
      "epoch": 451.9380530973451,
      "grad_norm": 9.5625,
      "learning_rate": 1.407387184551107e-05,
      "loss": 1.0617,
      "num_input_tokens_seen": 14276056,
      "step": 25760
    },
    {
      "epoch": 452.01769911504425,
      "grad_norm": 8.375,
      "learning_rate": 1.4065042460785532e-05,
      "loss": 1.0405,
      "num_input_tokens_seen": 14278736,
      "step": 25765
    },
    {
      "epoch": 452.1061946902655,
      "grad_norm": 6.1875,
      "learning_rate": 1.405621476236787e-05,
      "loss": 0.9277,
      "num_input_tokens_seen": 14281712,
      "step": 25770
    },
    {
      "epoch": 452.1946902654867,
      "grad_norm": 9.25,
      "learning_rate": 1.4047388751619423e-05,
      "loss": 1.1424,
      "num_input_tokens_seen": 14284320,
      "step": 25775
    },
    {
      "epoch": 452.283185840708,
      "grad_norm": 12.1875,
      "learning_rate": 1.4038564429901264e-05,
      "loss": 0.9526,
      "num_input_tokens_seen": 14287296,
      "step": 25780
    },
    {
      "epoch": 452.37168141592923,
      "grad_norm": 8.125,
      "learning_rate": 1.4029741798574227e-05,
      "loss": 1.0909,
      "num_input_tokens_seen": 14290112,
      "step": 25785
    },
    {
      "epoch": 452.46017699115043,
      "grad_norm": 7.125,
      "learning_rate": 1.402092085899886e-05,
      "loss": 0.8999,
      "num_input_tokens_seen": 14293168,
      "step": 25790
    },
    {
      "epoch": 452.5486725663717,
      "grad_norm": 8.8125,
      "learning_rate": 1.4012101612535464e-05,
      "loss": 1.1706,
      "num_input_tokens_seen": 14295680,
      "step": 25795
    },
    {
      "epoch": 452.6371681415929,
      "grad_norm": 7.3125,
      "learning_rate": 1.4003284060544092e-05,
      "loss": 0.9998,
      "num_input_tokens_seen": 14298896,
      "step": 25800
    },
    {
      "epoch": 452.6371681415929,
      "eval_loss": 1.0319925546646118,
      "eval_runtime": 0.9373,
      "eval_samples_per_second": 26.672,
      "eval_steps_per_second": 13.87,
      "num_input_tokens_seen": 14298896,
      "step": 25800
    },
    {
      "epoch": 452.72566371681415,
      "grad_norm": 6.96875,
      "learning_rate": 1.3994468204384504e-05,
      "loss": 0.8701,
      "num_input_tokens_seen": 14301664,
      "step": 25805
    },
    {
      "epoch": 452.8141592920354,
      "grad_norm": 6.9375,
      "learning_rate": 1.398565404541622e-05,
      "loss": 1.0801,
      "num_input_tokens_seen": 14304832,
      "step": 25810
    },
    {
      "epoch": 452.9026548672566,
      "grad_norm": 10.625,
      "learning_rate": 1.3976841584998513e-05,
      "loss": 1.4505,
      "num_input_tokens_seen": 14307280,
      "step": 25815
    },
    {
      "epoch": 452.9911504424779,
      "grad_norm": 8.0625,
      "learning_rate": 1.3968030824490352e-05,
      "loss": 1.0315,
      "num_input_tokens_seen": 14309696,
      "step": 25820
    },
    {
      "epoch": 453.070796460177,
      "grad_norm": 9.625,
      "learning_rate": 1.3959221765250469e-05,
      "loss": 1.0763,
      "num_input_tokens_seen": 14311912,
      "step": 25825
    },
    {
      "epoch": 453.1592920353982,
      "grad_norm": 9.75,
      "learning_rate": 1.3950414408637343e-05,
      "loss": 1.0979,
      "num_input_tokens_seen": 14314792,
      "step": 25830
    },
    {
      "epoch": 453.24778761061947,
      "grad_norm": 8.1875,
      "learning_rate": 1.3941608756009166e-05,
      "loss": 1.0793,
      "num_input_tokens_seen": 14317640,
      "step": 25835
    },
    {
      "epoch": 453.3362831858407,
      "grad_norm": 6.96875,
      "learning_rate": 1.3932804808723898e-05,
      "loss": 0.7989,
      "num_input_tokens_seen": 14320824,
      "step": 25840
    },
    {
      "epoch": 453.42477876106193,
      "grad_norm": 8.25,
      "learning_rate": 1.3924002568139194e-05,
      "loss": 0.9643,
      "num_input_tokens_seen": 14323272,
      "step": 25845
    },
    {
      "epoch": 453.5132743362832,
      "grad_norm": 8.3125,
      "learning_rate": 1.3915202035612485e-05,
      "loss": 0.9458,
      "num_input_tokens_seen": 14326024,
      "step": 25850
    },
    {
      "epoch": 453.60176991150445,
      "grad_norm": 7.5,
      "learning_rate": 1.3906403212500935e-05,
      "loss": 1.0506,
      "num_input_tokens_seen": 14329192,
      "step": 25855
    },
    {
      "epoch": 453.69026548672565,
      "grad_norm": 8.3125,
      "learning_rate": 1.3897606100161409e-05,
      "loss": 1.1081,
      "num_input_tokens_seen": 14331912,
      "step": 25860
    },
    {
      "epoch": 453.7787610619469,
      "grad_norm": 8.0,
      "learning_rate": 1.388881069995055e-05,
      "loss": 1.2107,
      "num_input_tokens_seen": 14334472,
      "step": 25865
    },
    {
      "epoch": 453.86725663716817,
      "grad_norm": 10.25,
      "learning_rate": 1.3880017013224708e-05,
      "loss": 1.1781,
      "num_input_tokens_seen": 14337176,
      "step": 25870
    },
    {
      "epoch": 453.95575221238937,
      "grad_norm": 11.0625,
      "learning_rate": 1.3871225041339984e-05,
      "loss": 1.1557,
      "num_input_tokens_seen": 14340040,
      "step": 25875
    },
    {
      "epoch": 454.0353982300885,
      "grad_norm": 10.625,
      "learning_rate": 1.386243478565222e-05,
      "loss": 1.0601,
      "num_input_tokens_seen": 14342640,
      "step": 25880
    },
    {
      "epoch": 454.12389380530976,
      "grad_norm": 7.59375,
      "learning_rate": 1.3853646247516966e-05,
      "loss": 0.9145,
      "num_input_tokens_seen": 14345056,
      "step": 25885
    },
    {
      "epoch": 454.21238938053096,
      "grad_norm": 7.9375,
      "learning_rate": 1.3844859428289545e-05,
      "loss": 1.0463,
      "num_input_tokens_seen": 14347616,
      "step": 25890
    },
    {
      "epoch": 454.3008849557522,
      "grad_norm": 7.125,
      "learning_rate": 1.3836074329324984e-05,
      "loss": 0.9008,
      "num_input_tokens_seen": 14350656,
      "step": 25895
    },
    {
      "epoch": 454.3893805309734,
      "grad_norm": 8.75,
      "learning_rate": 1.3827290951978044e-05,
      "loss": 0.9823,
      "num_input_tokens_seen": 14353232,
      "step": 25900
    },
    {
      "epoch": 454.4778761061947,
      "grad_norm": 6.84375,
      "learning_rate": 1.381850929760326e-05,
      "loss": 0.9865,
      "num_input_tokens_seen": 14355664,
      "step": 25905
    },
    {
      "epoch": 454.56637168141594,
      "grad_norm": 7.78125,
      "learning_rate": 1.3809729367554842e-05,
      "loss": 1.1941,
      "num_input_tokens_seen": 14358496,
      "step": 25910
    },
    {
      "epoch": 454.65486725663715,
      "grad_norm": 6.375,
      "learning_rate": 1.3800951163186784e-05,
      "loss": 1.1425,
      "num_input_tokens_seen": 14361424,
      "step": 25915
    },
    {
      "epoch": 454.7433628318584,
      "grad_norm": 6.90625,
      "learning_rate": 1.3792174685852801e-05,
      "loss": 1.0107,
      "num_input_tokens_seen": 14364000,
      "step": 25920
    },
    {
      "epoch": 454.83185840707966,
      "grad_norm": 9.625,
      "learning_rate": 1.378339993690632e-05,
      "loss": 1.038,
      "num_input_tokens_seen": 14367376,
      "step": 25925
    },
    {
      "epoch": 454.92035398230087,
      "grad_norm": 11.75,
      "learning_rate": 1.3774626917700523e-05,
      "loss": 1.109,
      "num_input_tokens_seen": 14370336,
      "step": 25930
    },
    {
      "epoch": 455.0,
      "grad_norm": 12.625,
      "learning_rate": 1.3765855629588334e-05,
      "loss": 1.122,
      "num_input_tokens_seen": 14372776,
      "step": 25935
    },
    {
      "epoch": 455.08849557522126,
      "grad_norm": 7.375,
      "learning_rate": 1.3757086073922374e-05,
      "loss": 1.1071,
      "num_input_tokens_seen": 14375192,
      "step": 25940
    },
    {
      "epoch": 455.17699115044246,
      "grad_norm": 8.9375,
      "learning_rate": 1.3748318252055038e-05,
      "loss": 1.0661,
      "num_input_tokens_seen": 14378376,
      "step": 25945
    },
    {
      "epoch": 455.2654867256637,
      "grad_norm": 8.625,
      "learning_rate": 1.3739552165338416e-05,
      "loss": 1.2582,
      "num_input_tokens_seen": 14381256,
      "step": 25950
    },
    {
      "epoch": 455.353982300885,
      "grad_norm": 8.1875,
      "learning_rate": 1.3730787815124354e-05,
      "loss": 1.0879,
      "num_input_tokens_seen": 14383928,
      "step": 25955
    },
    {
      "epoch": 455.4424778761062,
      "grad_norm": 6.34375,
      "learning_rate": 1.3722025202764443e-05,
      "loss": 0.8537,
      "num_input_tokens_seen": 14386744,
      "step": 25960
    },
    {
      "epoch": 455.53097345132744,
      "grad_norm": 10.25,
      "learning_rate": 1.371326432960997e-05,
      "loss": 0.8985,
      "num_input_tokens_seen": 14389432,
      "step": 25965
    },
    {
      "epoch": 455.6194690265487,
      "grad_norm": 8.25,
      "learning_rate": 1.3704505197011969e-05,
      "loss": 0.9514,
      "num_input_tokens_seen": 14391752,
      "step": 25970
    },
    {
      "epoch": 455.7079646017699,
      "grad_norm": 8.6875,
      "learning_rate": 1.3695747806321224e-05,
      "loss": 1.1902,
      "num_input_tokens_seen": 14394632,
      "step": 25975
    },
    {
      "epoch": 455.79646017699116,
      "grad_norm": 9.0,
      "learning_rate": 1.3686992158888212e-05,
      "loss": 1.0763,
      "num_input_tokens_seen": 14397592,
      "step": 25980
    },
    {
      "epoch": 455.88495575221236,
      "grad_norm": 7.9375,
      "learning_rate": 1.367823825606319e-05,
      "loss": 1.1183,
      "num_input_tokens_seen": 14400856,
      "step": 25985
    },
    {
      "epoch": 455.9734513274336,
      "grad_norm": 8.3125,
      "learning_rate": 1.36694860991961e-05,
      "loss": 0.9919,
      "num_input_tokens_seen": 14403656,
      "step": 25990
    },
    {
      "epoch": 456.05309734513276,
      "grad_norm": 7.53125,
      "learning_rate": 1.3660735689636636e-05,
      "loss": 1.0265,
      "num_input_tokens_seen": 14405984,
      "step": 25995
    },
    {
      "epoch": 456.14159292035396,
      "grad_norm": 7.96875,
      "learning_rate": 1.365198702873424e-05,
      "loss": 0.9302,
      "num_input_tokens_seen": 14408592,
      "step": 26000
    },
    {
      "epoch": 456.14159292035396,
      "eval_loss": 1.0287010669708252,
      "eval_runtime": 0.9403,
      "eval_samples_per_second": 26.588,
      "eval_steps_per_second": 13.826,
      "num_input_tokens_seen": 14408592,
      "step": 26000
    },
    {
      "epoch": 456.2300884955752,
      "grad_norm": 9.25,
      "learning_rate": 1.364324011783804e-05,
      "loss": 1.1252,
      "num_input_tokens_seen": 14411248,
      "step": 26005
    },
    {
      "epoch": 456.3185840707965,
      "grad_norm": 8.1875,
      "learning_rate": 1.3634494958296934e-05,
      "loss": 1.1646,
      "num_input_tokens_seen": 14414240,
      "step": 26010
    },
    {
      "epoch": 456.4070796460177,
      "grad_norm": 8.25,
      "learning_rate": 1.3625751551459542e-05,
      "loss": 1.1227,
      "num_input_tokens_seen": 14417344,
      "step": 26015
    },
    {
      "epoch": 456.49557522123894,
      "grad_norm": 8.375,
      "learning_rate": 1.3617009898674188e-05,
      "loss": 1.1376,
      "num_input_tokens_seen": 14420048,
      "step": 26020
    },
    {
      "epoch": 456.5840707964602,
      "grad_norm": 7.9375,
      "learning_rate": 1.3608270001288967e-05,
      "loss": 0.8511,
      "num_input_tokens_seen": 14423088,
      "step": 26025
    },
    {
      "epoch": 456.6725663716814,
      "grad_norm": 6.96875,
      "learning_rate": 1.359953186065166e-05,
      "loss": 0.9931,
      "num_input_tokens_seen": 14425968,
      "step": 26030
    },
    {
      "epoch": 456.76106194690266,
      "grad_norm": 8.5,
      "learning_rate": 1.3590795478109814e-05,
      "loss": 1.0463,
      "num_input_tokens_seen": 14428432,
      "step": 26035
    },
    {
      "epoch": 456.8495575221239,
      "grad_norm": 10.1875,
      "learning_rate": 1.3582060855010675e-05,
      "loss": 0.964,
      "num_input_tokens_seen": 14431200,
      "step": 26040
    },
    {
      "epoch": 456.9380530973451,
      "grad_norm": 10.125,
      "learning_rate": 1.3573327992701245e-05,
      "loss": 1.128,
      "num_input_tokens_seen": 14434384,
      "step": 26045
    },
    {
      "epoch": 457.01769911504425,
      "grad_norm": 9.75,
      "learning_rate": 1.356459689252823e-05,
      "loss": 1.1625,
      "num_input_tokens_seen": 14436616,
      "step": 26050
    },
    {
      "epoch": 457.1061946902655,
      "grad_norm": 10.5,
      "learning_rate": 1.3555867555838087e-05,
      "loss": 1.1892,
      "num_input_tokens_seen": 14439640,
      "step": 26055
    },
    {
      "epoch": 457.1946902654867,
      "grad_norm": 8.3125,
      "learning_rate": 1.3547139983976975e-05,
      "loss": 1.0321,
      "num_input_tokens_seen": 14442168,
      "step": 26060
    },
    {
      "epoch": 457.283185840708,
      "grad_norm": 9.0,
      "learning_rate": 1.3538414178290815e-05,
      "loss": 1.1451,
      "num_input_tokens_seen": 14444888,
      "step": 26065
    },
    {
      "epoch": 457.37168141592923,
      "grad_norm": 9.0,
      "learning_rate": 1.3529690140125209e-05,
      "loss": 0.9864,
      "num_input_tokens_seen": 14447800,
      "step": 26070
    },
    {
      "epoch": 457.46017699115043,
      "grad_norm": 9.25,
      "learning_rate": 1.352096787082553e-05,
      "loss": 1.0182,
      "num_input_tokens_seen": 14450344,
      "step": 26075
    },
    {
      "epoch": 457.5486725663717,
      "grad_norm": 8.8125,
      "learning_rate": 1.3512247371736871e-05,
      "loss": 1.0832,
      "num_input_tokens_seen": 14453064,
      "step": 26080
    },
    {
      "epoch": 457.6371681415929,
      "grad_norm": 7.40625,
      "learning_rate": 1.3503528644204022e-05,
      "loss": 1.0951,
      "num_input_tokens_seen": 14456312,
      "step": 26085
    },
    {
      "epoch": 457.72566371681415,
      "grad_norm": 8.375,
      "learning_rate": 1.349481168957153e-05,
      "loss": 1.0566,
      "num_input_tokens_seen": 14459064,
      "step": 26090
    },
    {
      "epoch": 457.8141592920354,
      "grad_norm": 9.0,
      "learning_rate": 1.3486096509183665e-05,
      "loss": 1.1051,
      "num_input_tokens_seen": 14462360,
      "step": 26095
    },
    {
      "epoch": 457.9026548672566,
      "grad_norm": 8.9375,
      "learning_rate": 1.3477383104384406e-05,
      "loss": 1.0704,
      "num_input_tokens_seen": 14464984,
      "step": 26100
    },
    {
      "epoch": 457.9911504424779,
      "grad_norm": 7.9375,
      "learning_rate": 1.3468671476517481e-05,
      "loss": 0.7978,
      "num_input_tokens_seen": 14467864,
      "step": 26105
    },
    {
      "epoch": 458.070796460177,
      "grad_norm": 8.375,
      "learning_rate": 1.3459961626926326e-05,
      "loss": 0.9517,
      "num_input_tokens_seen": 14470104,
      "step": 26110
    },
    {
      "epoch": 458.1592920353982,
      "grad_norm": 8.4375,
      "learning_rate": 1.3451253556954101e-05,
      "loss": 1.086,
      "num_input_tokens_seen": 14473208,
      "step": 26115
    },
    {
      "epoch": 458.24778761061947,
      "grad_norm": 7.3125,
      "learning_rate": 1.3442547267943717e-05,
      "loss": 1.054,
      "num_input_tokens_seen": 14476296,
      "step": 26120
    },
    {
      "epoch": 458.3362831858407,
      "grad_norm": 8.9375,
      "learning_rate": 1.3433842761237774e-05,
      "loss": 1.0828,
      "num_input_tokens_seen": 14478968,
      "step": 26125
    },
    {
      "epoch": 458.42477876106193,
      "grad_norm": 7.71875,
      "learning_rate": 1.3425140038178639e-05,
      "loss": 1.0243,
      "num_input_tokens_seen": 14481080,
      "step": 26130
    },
    {
      "epoch": 458.5132743362832,
      "grad_norm": 7.25,
      "learning_rate": 1.3416439100108358e-05,
      "loss": 1.0964,
      "num_input_tokens_seen": 14483880,
      "step": 26135
    },
    {
      "epoch": 458.60176991150445,
      "grad_norm": 7.21875,
      "learning_rate": 1.3407739948368734e-05,
      "loss": 0.8675,
      "num_input_tokens_seen": 14486504,
      "step": 26140
    },
    {
      "epoch": 458.69026548672565,
      "grad_norm": 9.3125,
      "learning_rate": 1.3399042584301298e-05,
      "loss": 1.1403,
      "num_input_tokens_seen": 14489208,
      "step": 26145
    },
    {
      "epoch": 458.7787610619469,
      "grad_norm": 8.125,
      "learning_rate": 1.3390347009247272e-05,
      "loss": 1.051,
      "num_input_tokens_seen": 14492424,
      "step": 26150
    },
    {
      "epoch": 458.86725663716817,
      "grad_norm": 9.5625,
      "learning_rate": 1.3381653224547635e-05,
      "loss": 0.9875,
      "num_input_tokens_seen": 14495896,
      "step": 26155
    },
    {
      "epoch": 458.95575221238937,
      "grad_norm": 10.875,
      "learning_rate": 1.3372961231543086e-05,
      "loss": 1.1017,
      "num_input_tokens_seen": 14498424,
      "step": 26160
    },
    {
      "epoch": 459.0353982300885,
      "grad_norm": 8.625,
      "learning_rate": 1.3364271031574016e-05,
      "loss": 1.0173,
      "num_input_tokens_seen": 14500584,
      "step": 26165
    },
    {
      "epoch": 459.12389380530976,
      "grad_norm": 8.3125,
      "learning_rate": 1.335558262598059e-05,
      "loss": 1.0591,
      "num_input_tokens_seen": 14503592,
      "step": 26170
    },
    {
      "epoch": 459.21238938053096,
      "grad_norm": 7.90625,
      "learning_rate": 1.3346896016102645e-05,
      "loss": 1.0481,
      "num_input_tokens_seen": 14506392,
      "step": 26175
    },
    {
      "epoch": 459.3008849557522,
      "grad_norm": 7.71875,
      "learning_rate": 1.3338211203279788e-05,
      "loss": 1.0099,
      "num_input_tokens_seen": 14509208,
      "step": 26180
    },
    {
      "epoch": 459.3893805309734,
      "grad_norm": 8.6875,
      "learning_rate": 1.3329528188851303e-05,
      "loss": 1.1287,
      "num_input_tokens_seen": 14511592,
      "step": 26185
    },
    {
      "epoch": 459.4778761061947,
      "grad_norm": 11.125,
      "learning_rate": 1.3320846974156242e-05,
      "loss": 1.1131,
      "num_input_tokens_seen": 14514344,
      "step": 26190
    },
    {
      "epoch": 459.56637168141594,
      "grad_norm": 7.5,
      "learning_rate": 1.3312167560533337e-05,
      "loss": 0.9253,
      "num_input_tokens_seen": 14516936,
      "step": 26195
    },
    {
      "epoch": 459.65486725663715,
      "grad_norm": 12.1875,
      "learning_rate": 1.3303489949321082e-05,
      "loss": 1.1934,
      "num_input_tokens_seen": 14519672,
      "step": 26200
    },
    {
      "epoch": 459.65486725663715,
      "eval_loss": 1.0287010669708252,
      "eval_runtime": 0.9381,
      "eval_samples_per_second": 26.65,
      "eval_steps_per_second": 13.858,
      "num_input_tokens_seen": 14519672,
      "step": 26200
    },
    {
      "epoch": 459.7433628318584,
      "grad_norm": 9.0,
      "learning_rate": 1.3294814141857653e-05,
      "loss": 1.1385,
      "num_input_tokens_seen": 14523224,
      "step": 26205
    },
    {
      "epoch": 459.83185840707966,
      "grad_norm": 7.78125,
      "learning_rate": 1.3286140139480992e-05,
      "loss": 0.9252,
      "num_input_tokens_seen": 14526344,
      "step": 26210
    },
    {
      "epoch": 459.92035398230087,
      "grad_norm": 9.0625,
      "learning_rate": 1.3277467943528719e-05,
      "loss": 1.0111,
      "num_input_tokens_seen": 14529192,
      "step": 26215
    },
    {
      "epoch": 460.0,
      "grad_norm": 15.1875,
      "learning_rate": 1.3268797555338203e-05,
      "loss": 0.9709,
      "num_input_tokens_seen": 14531304,
      "step": 26220
    },
    {
      "epoch": 460.08849557522126,
      "grad_norm": 9.9375,
      "learning_rate": 1.3260128976246533e-05,
      "loss": 1.0988,
      "num_input_tokens_seen": 14534216,
      "step": 26225
    },
    {
      "epoch": 460.17699115044246,
      "grad_norm": 8.6875,
      "learning_rate": 1.32514622075905e-05,
      "loss": 1.1156,
      "num_input_tokens_seen": 14536808,
      "step": 26230
    },
    {
      "epoch": 460.2654867256637,
      "grad_norm": 8.625,
      "learning_rate": 1.3242797250706638e-05,
      "loss": 1.012,
      "num_input_tokens_seen": 14539832,
      "step": 26235
    },
    {
      "epoch": 460.353982300885,
      "grad_norm": 8.6875,
      "learning_rate": 1.3234134106931195e-05,
      "loss": 1.0855,
      "num_input_tokens_seen": 14542392,
      "step": 26240
    },
    {
      "epoch": 460.4424778761062,
      "grad_norm": 11.3125,
      "learning_rate": 1.322547277760013e-05,
      "loss": 1.0793,
      "num_input_tokens_seen": 14545208,
      "step": 26245
    },
    {
      "epoch": 460.53097345132744,
      "grad_norm": 9.25,
      "learning_rate": 1.3216813264049132e-05,
      "loss": 1.0445,
      "num_input_tokens_seen": 14547768,
      "step": 26250
    },
    {
      "epoch": 460.6194690265487,
      "grad_norm": 13.9375,
      "learning_rate": 1.32081555676136e-05,
      "loss": 1.0061,
      "num_input_tokens_seen": 14550664,
      "step": 26255
    },
    {
      "epoch": 460.7079646017699,
      "grad_norm": 7.53125,
      "learning_rate": 1.3199499689628674e-05,
      "loss": 0.9742,
      "num_input_tokens_seen": 14553688,
      "step": 26260
    },
    {
      "epoch": 460.79646017699116,
      "grad_norm": 8.875,
      "learning_rate": 1.3190845631429192e-05,
      "loss": 1.1911,
      "num_input_tokens_seen": 14556440,
      "step": 26265
    },
    {
      "epoch": 460.88495575221236,
      "grad_norm": 6.59375,
      "learning_rate": 1.3182193394349704e-05,
      "loss": 0.8604,
      "num_input_tokens_seen": 14558616,
      "step": 26270
    },
    {
      "epoch": 460.9734513274336,
      "grad_norm": 9.3125,
      "learning_rate": 1.3173542979724507e-05,
      "loss": 1.0767,
      "num_input_tokens_seen": 14562072,
      "step": 26275
    },
    {
      "epoch": 461.05309734513276,
      "grad_norm": 6.96875,
      "learning_rate": 1.3164894388887617e-05,
      "loss": 0.7983,
      "num_input_tokens_seen": 14564288,
      "step": 26280
    },
    {
      "epoch": 461.14159292035396,
      "grad_norm": 8.8125,
      "learning_rate": 1.3156247623172727e-05,
      "loss": 1.3266,
      "num_input_tokens_seen": 14566992,
      "step": 26285
    },
    {
      "epoch": 461.2300884955752,
      "grad_norm": 7.84375,
      "learning_rate": 1.3147602683913302e-05,
      "loss": 0.9189,
      "num_input_tokens_seen": 14570112,
      "step": 26290
    },
    {
      "epoch": 461.3185840707965,
      "grad_norm": 14.3125,
      "learning_rate": 1.3138959572442481e-05,
      "loss": 1.1226,
      "num_input_tokens_seen": 14572464,
      "step": 26295
    },
    {
      "epoch": 461.4070796460177,
      "grad_norm": 8.5625,
      "learning_rate": 1.3130318290093146e-05,
      "loss": 0.9935,
      "num_input_tokens_seen": 14575472,
      "step": 26300
    },
    {
      "epoch": 461.49557522123894,
      "grad_norm": 7.28125,
      "learning_rate": 1.3121678838197909e-05,
      "loss": 0.9781,
      "num_input_tokens_seen": 14578448,
      "step": 26305
    },
    {
      "epoch": 461.5840707964602,
      "grad_norm": 9.4375,
      "learning_rate": 1.3113041218089056e-05,
      "loss": 0.9213,
      "num_input_tokens_seen": 14581248,
      "step": 26310
    },
    {
      "epoch": 461.6725663716814,
      "grad_norm": 8.875,
      "learning_rate": 1.3104405431098626e-05,
      "loss": 1.2105,
      "num_input_tokens_seen": 14584288,
      "step": 26315
    },
    {
      "epoch": 461.76106194690266,
      "grad_norm": 7.90625,
      "learning_rate": 1.3095771478558377e-05,
      "loss": 1.2998,
      "num_input_tokens_seen": 14587152,
      "step": 26320
    },
    {
      "epoch": 461.8495575221239,
      "grad_norm": 9.9375,
      "learning_rate": 1.3087139361799766e-05,
      "loss": 0.9618,
      "num_input_tokens_seen": 14590080,
      "step": 26325
    },
    {
      "epoch": 461.9380530973451,
      "grad_norm": 5.9375,
      "learning_rate": 1.3078509082153964e-05,
      "loss": 0.9717,
      "num_input_tokens_seen": 14592656,
      "step": 26330
    },
    {
      "epoch": 462.01769911504425,
      "grad_norm": 6.1875,
      "learning_rate": 1.3069880640951885e-05,
      "loss": 0.8678,
      "num_input_tokens_seen": 14594704,
      "step": 26335
    },
    {
      "epoch": 462.1061946902655,
      "grad_norm": 8.125,
      "learning_rate": 1.3061254039524123e-05,
      "loss": 0.9076,
      "num_input_tokens_seen": 14597104,
      "step": 26340
    },
    {
      "epoch": 462.1946902654867,
      "grad_norm": 6.875,
      "learning_rate": 1.3052629279201028e-05,
      "loss": 1.0334,
      "num_input_tokens_seen": 14599872,
      "step": 26345
    },
    {
      "epoch": 462.283185840708,
      "grad_norm": 7.90625,
      "learning_rate": 1.3044006361312633e-05,
      "loss": 1.2864,
      "num_input_tokens_seen": 14602240,
      "step": 26350
    },
    {
      "epoch": 462.37168141592923,
      "grad_norm": 12.125,
      "learning_rate": 1.30353852871887e-05,
      "loss": 1.0187,
      "num_input_tokens_seen": 14605696,
      "step": 26355
    },
    {
      "epoch": 462.46017699115043,
      "grad_norm": 8.125,
      "learning_rate": 1.302676605815873e-05,
      "loss": 1.0908,
      "num_input_tokens_seen": 14608432,
      "step": 26360
    },
    {
      "epoch": 462.5486725663717,
      "grad_norm": 8.375,
      "learning_rate": 1.3018148675551884e-05,
      "loss": 0.9177,
      "num_input_tokens_seen": 14611328,
      "step": 26365
    },
    {
      "epoch": 462.6371681415929,
      "grad_norm": 8.3125,
      "learning_rate": 1.3009533140697094e-05,
      "loss": 1.1497,
      "num_input_tokens_seen": 14613888,
      "step": 26370
    },
    {
      "epoch": 462.72566371681415,
      "grad_norm": 7.84375,
      "learning_rate": 1.3000919454922966e-05,
      "loss": 1.0147,
      "num_input_tokens_seen": 14616640,
      "step": 26375
    },
    {
      "epoch": 462.8141592920354,
      "grad_norm": 8.25,
      "learning_rate": 1.299230761955785e-05,
      "loss": 1.0887,
      "num_input_tokens_seen": 14619520,
      "step": 26380
    },
    {
      "epoch": 462.9026548672566,
      "grad_norm": 6.65625,
      "learning_rate": 1.2983697635929807e-05,
      "loss": 1.0225,
      "num_input_tokens_seen": 14622192,
      "step": 26385
    },
    {
      "epoch": 462.9911504424779,
      "grad_norm": 8.125,
      "learning_rate": 1.2975089505366584e-05,
      "loss": 1.0766,
      "num_input_tokens_seen": 14625392,
      "step": 26390
    },
    {
      "epoch": 463.070796460177,
      "grad_norm": 7.1875,
      "learning_rate": 1.2966483229195683e-05,
      "loss": 0.9714,
      "num_input_tokens_seen": 14627728,
      "step": 26395
    },
    {
      "epoch": 463.1592920353982,
      "grad_norm": 9.0,
      "learning_rate": 1.2957878808744283e-05,
      "loss": 1.0052,
      "num_input_tokens_seen": 14630736,
      "step": 26400
    },
    {
      "epoch": 463.1592920353982,
      "eval_loss": 1.0287010669708252,
      "eval_runtime": 0.9368,
      "eval_samples_per_second": 26.686,
      "eval_steps_per_second": 13.877,
      "num_input_tokens_seen": 14630736,
      "step": 26400
    },
    {
      "epoch": 463.24778761061947,
      "grad_norm": 7.9375,
      "learning_rate": 1.294927624533931e-05,
      "loss": 1.0022,
      "num_input_tokens_seen": 14633552,
      "step": 26405
    },
    {
      "epoch": 463.3362831858407,
      "grad_norm": 15.625,
      "learning_rate": 1.2940675540307378e-05,
      "loss": 1.3668,
      "num_input_tokens_seen": 14636368,
      "step": 26410
    },
    {
      "epoch": 463.42477876106193,
      "grad_norm": 10.125,
      "learning_rate": 1.2932076694974814e-05,
      "loss": 0.9473,
      "num_input_tokens_seen": 14639200,
      "step": 26415
    },
    {
      "epoch": 463.5132743362832,
      "grad_norm": 8.5,
      "learning_rate": 1.2923479710667682e-05,
      "loss": 1.1664,
      "num_input_tokens_seen": 14641808,
      "step": 26420
    },
    {
      "epoch": 463.60176991150445,
      "grad_norm": 9.125,
      "learning_rate": 1.2914884588711751e-05,
      "loss": 1.1464,
      "num_input_tokens_seen": 14644400,
      "step": 26425
    },
    {
      "epoch": 463.69026548672565,
      "grad_norm": 7.875,
      "learning_rate": 1.2906291330432475e-05,
      "loss": 1.0785,
      "num_input_tokens_seen": 14647104,
      "step": 26430
    },
    {
      "epoch": 463.7787610619469,
      "grad_norm": 7.40625,
      "learning_rate": 1.2897699937155055e-05,
      "loss": 1.0297,
      "num_input_tokens_seen": 14650208,
      "step": 26435
    },
    {
      "epoch": 463.86725663716817,
      "grad_norm": 8.5625,
      "learning_rate": 1.2889110410204403e-05,
      "loss": 0.863,
      "num_input_tokens_seen": 14653120,
      "step": 26440
    },
    {
      "epoch": 463.95575221238937,
      "grad_norm": 6.4375,
      "learning_rate": 1.2880522750905111e-05,
      "loss": 1.055,
      "num_input_tokens_seen": 14655712,
      "step": 26445
    },
    {
      "epoch": 464.0353982300885,
      "grad_norm": 7.15625,
      "learning_rate": 1.2871936960581523e-05,
      "loss": 0.9673,
      "num_input_tokens_seen": 14658288,
      "step": 26450
    },
    {
      "epoch": 464.12389380530976,
      "grad_norm": 9.0625,
      "learning_rate": 1.2863353040557658e-05,
      "loss": 0.9039,
      "num_input_tokens_seen": 14661360,
      "step": 26455
    },
    {
      "epoch": 464.21238938053096,
      "grad_norm": 7.8125,
      "learning_rate": 1.2854770992157273e-05,
      "loss": 0.7596,
      "num_input_tokens_seen": 14664400,
      "step": 26460
    },
    {
      "epoch": 464.3008849557522,
      "grad_norm": 8.0625,
      "learning_rate": 1.2846190816703835e-05,
      "loss": 1.0677,
      "num_input_tokens_seen": 14666720,
      "step": 26465
    },
    {
      "epoch": 464.3893805309734,
      "grad_norm": 7.71875,
      "learning_rate": 1.2837612515520498e-05,
      "loss": 0.9315,
      "num_input_tokens_seen": 14669840,
      "step": 26470
    },
    {
      "epoch": 464.4778761061947,
      "grad_norm": 7.875,
      "learning_rate": 1.2829036089930163e-05,
      "loss": 1.166,
      "num_input_tokens_seen": 14672736,
      "step": 26475
    },
    {
      "epoch": 464.56637168141594,
      "grad_norm": 7.4375,
      "learning_rate": 1.2820461541255412e-05,
      "loss": 1.149,
      "num_input_tokens_seen": 14675104,
      "step": 26480
    },
    {
      "epoch": 464.65486725663715,
      "grad_norm": 8.3125,
      "learning_rate": 1.2811888870818543e-05,
      "loss": 1.433,
      "num_input_tokens_seen": 14677584,
      "step": 26485
    },
    {
      "epoch": 464.7433628318584,
      "grad_norm": 9.1875,
      "learning_rate": 1.2803318079941581e-05,
      "loss": 1.1616,
      "num_input_tokens_seen": 14680064,
      "step": 26490
    },
    {
      "epoch": 464.83185840707966,
      "grad_norm": 8.125,
      "learning_rate": 1.2794749169946235e-05,
      "loss": 1.1255,
      "num_input_tokens_seen": 14683456,
      "step": 26495
    },
    {
      "epoch": 464.92035398230087,
      "grad_norm": 7.53125,
      "learning_rate": 1.2786182142153952e-05,
      "loss": 0.9116,
      "num_input_tokens_seen": 14686784,
      "step": 26500
    },
    {
      "epoch": 465.0,
      "grad_norm": 6.53125,
      "learning_rate": 1.2777616997885878e-05,
      "loss": 0.979,
      "num_input_tokens_seen": 14688896,
      "step": 26505
    },
    {
      "epoch": 465.08849557522126,
      "grad_norm": 8.5,
      "learning_rate": 1.2769053738462847e-05,
      "loss": 1.1001,
      "num_input_tokens_seen": 14691504,
      "step": 26510
    },
    {
      "epoch": 465.17699115044246,
      "grad_norm": 8.4375,
      "learning_rate": 1.2760492365205434e-05,
      "loss": 1.1948,
      "num_input_tokens_seen": 14694736,
      "step": 26515
    },
    {
      "epoch": 465.2654867256637,
      "grad_norm": 9.875,
      "learning_rate": 1.2751932879433919e-05,
      "loss": 1.0594,
      "num_input_tokens_seen": 14697648,
      "step": 26520
    },
    {
      "epoch": 465.353982300885,
      "grad_norm": 8.125,
      "learning_rate": 1.2743375282468267e-05,
      "loss": 1.189,
      "num_input_tokens_seen": 14700464,
      "step": 26525
    },
    {
      "epoch": 465.4424778761062,
      "grad_norm": 7.84375,
      "learning_rate": 1.2734819575628182e-05,
      "loss": 0.9493,
      "num_input_tokens_seen": 14703216,
      "step": 26530
    },
    {
      "epoch": 465.53097345132744,
      "grad_norm": 6.96875,
      "learning_rate": 1.2726265760233039e-05,
      "loss": 1.0408,
      "num_input_tokens_seen": 14705744,
      "step": 26535
    },
    {
      "epoch": 465.6194690265487,
      "grad_norm": 9.75,
      "learning_rate": 1.271771383760197e-05,
      "loss": 0.9349,
      "num_input_tokens_seen": 14708656,
      "step": 26540
    },
    {
      "epoch": 465.7079646017699,
      "grad_norm": 9.625,
      "learning_rate": 1.2709163809053764e-05,
      "loss": 0.9002,
      "num_input_tokens_seen": 14711008,
      "step": 26545
    },
    {
      "epoch": 465.79646017699116,
      "grad_norm": 9.875,
      "learning_rate": 1.2700615675906963e-05,
      "loss": 1.1876,
      "num_input_tokens_seen": 14713824,
      "step": 26550
    },
    {
      "epoch": 465.88495575221236,
      "grad_norm": 9.0625,
      "learning_rate": 1.269206943947978e-05,
      "loss": 1.0156,
      "num_input_tokens_seen": 14717264,
      "step": 26555
    },
    {
      "epoch": 465.9734513274336,
      "grad_norm": 8.6875,
      "learning_rate": 1.2683525101090177e-05,
      "loss": 0.9924,
      "num_input_tokens_seen": 14719760,
      "step": 26560
    },
    {
      "epoch": 466.05309734513276,
      "grad_norm": 8.8125,
      "learning_rate": 1.2674982662055765e-05,
      "loss": 0.9489,
      "num_input_tokens_seen": 14721824,
      "step": 26565
    },
    {
      "epoch": 466.14159292035396,
      "grad_norm": 6.625,
      "learning_rate": 1.2666442123693922e-05,
      "loss": 0.9144,
      "num_input_tokens_seen": 14724720,
      "step": 26570
    },
    {
      "epoch": 466.2300884955752,
      "grad_norm": 8.125,
      "learning_rate": 1.265790348732169e-05,
      "loss": 1.1058,
      "num_input_tokens_seen": 14727312,
      "step": 26575
    },
    {
      "epoch": 466.3185840707965,
      "grad_norm": 8.9375,
      "learning_rate": 1.264936675425584e-05,
      "loss": 1.0738,
      "num_input_tokens_seen": 14729936,
      "step": 26580
    },
    {
      "epoch": 466.4070796460177,
      "grad_norm": 9.5625,
      "learning_rate": 1.2640831925812852e-05,
      "loss": 0.9467,
      "num_input_tokens_seen": 14733104,
      "step": 26585
    },
    {
      "epoch": 466.49557522123894,
      "grad_norm": 9.25,
      "learning_rate": 1.263229900330889e-05,
      "loss": 1.2806,
      "num_input_tokens_seen": 14735616,
      "step": 26590
    },
    {
      "epoch": 466.5840707964602,
      "grad_norm": 10.875,
      "learning_rate": 1.2623767988059843e-05,
      "loss": 1.0533,
      "num_input_tokens_seen": 14738656,
      "step": 26595
    },
    {
      "epoch": 466.6725663716814,
      "grad_norm": 7.0625,
      "learning_rate": 1.2615238881381309e-05,
      "loss": 0.9753,
      "num_input_tokens_seen": 14741472,
      "step": 26600
    },
    {
      "epoch": 466.6725663716814,
      "eval_loss": 1.0287010669708252,
      "eval_runtime": 0.9369,
      "eval_samples_per_second": 26.685,
      "eval_steps_per_second": 13.876,
      "num_input_tokens_seen": 14741472,
      "step": 26600
    },
    {
      "epoch": 466.76106194690266,
      "grad_norm": 7.21875,
      "learning_rate": 1.2606711684588568e-05,
      "loss": 1.1242,
      "num_input_tokens_seen": 14744400,
      "step": 26605
    },
    {
      "epoch": 466.8495575221239,
      "grad_norm": 7.1875,
      "learning_rate": 1.2598186398996636e-05,
      "loss": 1.1574,
      "num_input_tokens_seen": 14747648,
      "step": 26610
    },
    {
      "epoch": 466.9380530973451,
      "grad_norm": 7.28125,
      "learning_rate": 1.2589663025920207e-05,
      "loss": 1.0696,
      "num_input_tokens_seen": 14750032,
      "step": 26615
    },
    {
      "epoch": 467.01769911504425,
      "grad_norm": 7.875,
      "learning_rate": 1.2581141566673705e-05,
      "loss": 1.0916,
      "num_input_tokens_seen": 14752344,
      "step": 26620
    },
    {
      "epoch": 467.1061946902655,
      "grad_norm": 10.1875,
      "learning_rate": 1.257262202257124e-05,
      "loss": 0.9139,
      "num_input_tokens_seen": 14755112,
      "step": 26625
    },
    {
      "epoch": 467.1946902654867,
      "grad_norm": 6.28125,
      "learning_rate": 1.2564104394926618e-05,
      "loss": 1.0898,
      "num_input_tokens_seen": 14757544,
      "step": 26630
    },
    {
      "epoch": 467.283185840708,
      "grad_norm": 10.0625,
      "learning_rate": 1.2555588685053383e-05,
      "loss": 1.3987,
      "num_input_tokens_seen": 14760104,
      "step": 26635
    },
    {
      "epoch": 467.37168141592923,
      "grad_norm": 6.9375,
      "learning_rate": 1.2547074894264762e-05,
      "loss": 1.0714,
      "num_input_tokens_seen": 14763064,
      "step": 26640
    },
    {
      "epoch": 467.46017699115043,
      "grad_norm": 6.0,
      "learning_rate": 1.2538563023873679e-05,
      "loss": 0.9518,
      "num_input_tokens_seen": 14765624,
      "step": 26645
    },
    {
      "epoch": 467.5486725663717,
      "grad_norm": 9.1875,
      "learning_rate": 1.2530053075192789e-05,
      "loss": 1.0771,
      "num_input_tokens_seen": 14768296,
      "step": 26650
    },
    {
      "epoch": 467.6371681415929,
      "grad_norm": 8.5625,
      "learning_rate": 1.252154504953441e-05,
      "loss": 0.9958,
      "num_input_tokens_seen": 14771608,
      "step": 26655
    },
    {
      "epoch": 467.72566371681415,
      "grad_norm": 10.6875,
      "learning_rate": 1.25130389482106e-05,
      "loss": 1.1126,
      "num_input_tokens_seen": 14774488,
      "step": 26660
    },
    {
      "epoch": 467.8141592920354,
      "grad_norm": 8.0625,
      "learning_rate": 1.2504534772533116e-05,
      "loss": 0.9747,
      "num_input_tokens_seen": 14777368,
      "step": 26665
    },
    {
      "epoch": 467.9026548672566,
      "grad_norm": 9.5625,
      "learning_rate": 1.2496032523813387e-05,
      "loss": 1.0764,
      "num_input_tokens_seen": 14780552,
      "step": 26670
    },
    {
      "epoch": 467.9911504424779,
      "grad_norm": 9.875,
      "learning_rate": 1.2487532203362576e-05,
      "loss": 1.0426,
      "num_input_tokens_seen": 14783224,
      "step": 26675
    },
    {
      "epoch": 468.070796460177,
      "grad_norm": 8.5625,
      "learning_rate": 1.247903381249155e-05,
      "loss": 0.7921,
      "num_input_tokens_seen": 14786008,
      "step": 26680
    },
    {
      "epoch": 468.1592920353982,
      "grad_norm": 8.3125,
      "learning_rate": 1.2470537352510853e-05,
      "loss": 1.0502,
      "num_input_tokens_seen": 14789032,
      "step": 26685
    },
    {
      "epoch": 468.24778761061947,
      "grad_norm": 7.40625,
      "learning_rate": 1.2462042824730758e-05,
      "loss": 1.0849,
      "num_input_tokens_seen": 14792472,
      "step": 26690
    },
    {
      "epoch": 468.3362831858407,
      "grad_norm": 8.625,
      "learning_rate": 1.245355023046122e-05,
      "loss": 1.113,
      "num_input_tokens_seen": 14794808,
      "step": 26695
    },
    {
      "epoch": 468.42477876106193,
      "grad_norm": 10.1875,
      "learning_rate": 1.2445059571011896e-05,
      "loss": 1.0368,
      "num_input_tokens_seen": 14797880,
      "step": 26700
    },
    {
      "epoch": 468.5132743362832,
      "grad_norm": 7.59375,
      "learning_rate": 1.2436570847692173e-05,
      "loss": 1.0527,
      "num_input_tokens_seen": 14800856,
      "step": 26705
    },
    {
      "epoch": 468.60176991150445,
      "grad_norm": 7.21875,
      "learning_rate": 1.2428084061811096e-05,
      "loss": 1.0253,
      "num_input_tokens_seen": 14804232,
      "step": 26710
    },
    {
      "epoch": 468.69026548672565,
      "grad_norm": 8.6875,
      "learning_rate": 1.2419599214677447e-05,
      "loss": 1.0326,
      "num_input_tokens_seen": 14807016,
      "step": 26715
    },
    {
      "epoch": 468.7787610619469,
      "grad_norm": 8.9375,
      "learning_rate": 1.2411116307599702e-05,
      "loss": 1.1341,
      "num_input_tokens_seen": 14809400,
      "step": 26720
    },
    {
      "epoch": 468.86725663716817,
      "grad_norm": 7.5,
      "learning_rate": 1.2402635341886016e-05,
      "loss": 1.0442,
      "num_input_tokens_seen": 14811912,
      "step": 26725
    },
    {
      "epoch": 468.95575221238937,
      "grad_norm": 7.25,
      "learning_rate": 1.2394156318844278e-05,
      "loss": 1.0495,
      "num_input_tokens_seen": 14814296,
      "step": 26730
    },
    {
      "epoch": 469.0353982300885,
      "grad_norm": 5.96875,
      "learning_rate": 1.2385679239782039e-05,
      "loss": 0.9792,
      "num_input_tokens_seen": 14816208,
      "step": 26735
    },
    {
      "epoch": 469.12389380530976,
      "grad_norm": 8.6875,
      "learning_rate": 1.2377204106006585e-05,
      "loss": 1.0173,
      "num_input_tokens_seen": 14818896,
      "step": 26740
    },
    {
      "epoch": 469.21238938053096,
      "grad_norm": 7.15625,
      "learning_rate": 1.2368730918824891e-05,
      "loss": 1.1639,
      "num_input_tokens_seen": 14821920,
      "step": 26745
    },
    {
      "epoch": 469.3008849557522,
      "grad_norm": 10.0625,
      "learning_rate": 1.236025967954362e-05,
      "loss": 1.1141,
      "num_input_tokens_seen": 14824560,
      "step": 26750
    },
    {
      "epoch": 469.3893805309734,
      "grad_norm": 6.96875,
      "learning_rate": 1.2351790389469153e-05,
      "loss": 0.9776,
      "num_input_tokens_seen": 14827584,
      "step": 26755
    },
    {
      "epoch": 469.4778761061947,
      "grad_norm": 9.625,
      "learning_rate": 1.234332304990755e-05,
      "loss": 1.1099,
      "num_input_tokens_seen": 14830400,
      "step": 26760
    },
    {
      "epoch": 469.56637168141594,
      "grad_norm": 8.6875,
      "learning_rate": 1.2334857662164593e-05,
      "loss": 0.9936,
      "num_input_tokens_seen": 14833024,
      "step": 26765
    },
    {
      "epoch": 469.65486725663715,
      "grad_norm": 7.90625,
      "learning_rate": 1.2326394227545743e-05,
      "loss": 1.0753,
      "num_input_tokens_seen": 14835696,
      "step": 26770
    },
    {
      "epoch": 469.7433628318584,
      "grad_norm": 7.84375,
      "learning_rate": 1.2317932747356162e-05,
      "loss": 1.0776,
      "num_input_tokens_seen": 14838576,
      "step": 26775
    },
    {
      "epoch": 469.83185840707966,
      "grad_norm": 6.40625,
      "learning_rate": 1.2309473222900726e-05,
      "loss": 1.0267,
      "num_input_tokens_seen": 14841584,
      "step": 26780
    },
    {
      "epoch": 469.92035398230087,
      "grad_norm": 7.90625,
      "learning_rate": 1.2301015655484006e-05,
      "loss": 0.9875,
      "num_input_tokens_seen": 14844624,
      "step": 26785
    },
    {
      "epoch": 470.0,
      "grad_norm": 9.1875,
      "learning_rate": 1.2292560046410245e-05,
      "loss": 1.1297,
      "num_input_tokens_seen": 14847040,
      "step": 26790
    },
    {
      "epoch": 470.08849557522126,
      "grad_norm": 7.75,
      "learning_rate": 1.228410639698343e-05,
      "loss": 0.9955,
      "num_input_tokens_seen": 14849872,
      "step": 26795
    },
    {
      "epoch": 470.17699115044246,
      "grad_norm": 10.375,
      "learning_rate": 1.2275654708507195e-05,
      "loss": 1.019,
      "num_input_tokens_seen": 14852816,
      "step": 26800
    },
    {
      "epoch": 470.17699115044246,
      "eval_loss": 1.0287010669708252,
      "eval_runtime": 0.937,
      "eval_samples_per_second": 26.679,
      "eval_steps_per_second": 13.873,
      "num_input_tokens_seen": 14852816,
      "step": 26800
    },
    {
      "epoch": 470.2654867256637,
      "grad_norm": 7.65625,
      "learning_rate": 1.2267204982284908e-05,
      "loss": 1.0126,
      "num_input_tokens_seen": 14856032,
      "step": 26805
    },
    {
      "epoch": 470.353982300885,
      "grad_norm": 7.0,
      "learning_rate": 1.2258757219619635e-05,
      "loss": 1.1255,
      "num_input_tokens_seen": 14858576,
      "step": 26810
    },
    {
      "epoch": 470.4424778761062,
      "grad_norm": 7.375,
      "learning_rate": 1.2250311421814104e-05,
      "loss": 0.9479,
      "num_input_tokens_seen": 14861168,
      "step": 26815
    },
    {
      "epoch": 470.53097345132744,
      "grad_norm": 7.75,
      "learning_rate": 1.2241867590170772e-05,
      "loss": 1.1316,
      "num_input_tokens_seen": 14864544,
      "step": 26820
    },
    {
      "epoch": 470.6194690265487,
      "grad_norm": 10.0625,
      "learning_rate": 1.2233425725991799e-05,
      "loss": 1.16,
      "num_input_tokens_seen": 14867072,
      "step": 26825
    },
    {
      "epoch": 470.7079646017699,
      "grad_norm": 10.4375,
      "learning_rate": 1.2224985830579003e-05,
      "loss": 1.1006,
      "num_input_tokens_seen": 14869952,
      "step": 26830
    },
    {
      "epoch": 470.79646017699116,
      "grad_norm": 7.375,
      "learning_rate": 1.2216547905233944e-05,
      "loss": 1.138,
      "num_input_tokens_seen": 14872448,
      "step": 26835
    },
    {
      "epoch": 470.88495575221236,
      "grad_norm": 8.6875,
      "learning_rate": 1.2208111951257842e-05,
      "loss": 0.9003,
      "num_input_tokens_seen": 14875072,
      "step": 26840
    },
    {
      "epoch": 470.9734513274336,
      "grad_norm": 8.6875,
      "learning_rate": 1.2199677969951622e-05,
      "loss": 1.0168,
      "num_input_tokens_seen": 14877904,
      "step": 26845
    },
    {
      "epoch": 471.05309734513276,
      "grad_norm": 10.9375,
      "learning_rate": 1.2191245962615927e-05,
      "loss": 0.9674,
      "num_input_tokens_seen": 14879904,
      "step": 26850
    },
    {
      "epoch": 471.14159292035396,
      "grad_norm": 10.875,
      "learning_rate": 1.218281593055106e-05,
      "loss": 1.0803,
      "num_input_tokens_seen": 14882704,
      "step": 26855
    },
    {
      "epoch": 471.2300884955752,
      "grad_norm": 9.75,
      "learning_rate": 1.217438787505705e-05,
      "loss": 0.8473,
      "num_input_tokens_seen": 14885792,
      "step": 26860
    },
    {
      "epoch": 471.3185840707965,
      "grad_norm": 7.90625,
      "learning_rate": 1.2165961797433615e-05,
      "loss": 1.1373,
      "num_input_tokens_seen": 14888768,
      "step": 26865
    },
    {
      "epoch": 471.4070796460177,
      "grad_norm": 7.28125,
      "learning_rate": 1.215753769898014e-05,
      "loss": 1.0222,
      "num_input_tokens_seen": 14891984,
      "step": 26870
    },
    {
      "epoch": 471.49557522123894,
      "grad_norm": 8.125,
      "learning_rate": 1.2149115580995755e-05,
      "loss": 1.2768,
      "num_input_tokens_seen": 14894448,
      "step": 26875
    },
    {
      "epoch": 471.5840707964602,
      "grad_norm": 6.53125,
      "learning_rate": 1.2140695444779227e-05,
      "loss": 1.0607,
      "num_input_tokens_seen": 14897424,
      "step": 26880
    },
    {
      "epoch": 471.6725663716814,
      "grad_norm": 7.59375,
      "learning_rate": 1.2132277291629066e-05,
      "loss": 0.8467,
      "num_input_tokens_seen": 14900080,
      "step": 26885
    },
    {
      "epoch": 471.76106194690266,
      "grad_norm": 8.5,
      "learning_rate": 1.2123861122843458e-05,
      "loss": 1.0791,
      "num_input_tokens_seen": 14902880,
      "step": 26890
    },
    {
      "epoch": 471.8495575221239,
      "grad_norm": 8.875,
      "learning_rate": 1.2115446939720271e-05,
      "loss": 1.1775,
      "num_input_tokens_seen": 14905824,
      "step": 26895
    },
    {
      "epoch": 471.9380530973451,
      "grad_norm": 7.90625,
      "learning_rate": 1.210703474355708e-05,
      "loss": 0.9989,
      "num_input_tokens_seen": 14908352,
      "step": 26900
    },
    {
      "epoch": 472.01769911504425,
      "grad_norm": 7.25,
      "learning_rate": 1.2098624535651164e-05,
      "loss": 1.0922,
      "num_input_tokens_seen": 14910672,
      "step": 26905
    },
    {
      "epoch": 472.1061946902655,
      "grad_norm": 10.125,
      "learning_rate": 1.2090216317299477e-05,
      "loss": 1.1452,
      "num_input_tokens_seen": 14914128,
      "step": 26910
    },
    {
      "epoch": 472.1946902654867,
      "grad_norm": 7.53125,
      "learning_rate": 1.2081810089798668e-05,
      "loss": 1.2008,
      "num_input_tokens_seen": 14916960,
      "step": 26915
    },
    {
      "epoch": 472.283185840708,
      "grad_norm": 9.25,
      "learning_rate": 1.2073405854445072e-05,
      "loss": 1.177,
      "num_input_tokens_seen": 14920016,
      "step": 26920
    },
    {
      "epoch": 472.37168141592923,
      "grad_norm": 7.0,
      "learning_rate": 1.206500361253474e-05,
      "loss": 0.8854,
      "num_input_tokens_seen": 14922448,
      "step": 26925
    },
    {
      "epoch": 472.46017699115043,
      "grad_norm": 10.0625,
      "learning_rate": 1.2056603365363409e-05,
      "loss": 0.9735,
      "num_input_tokens_seen": 14925312,
      "step": 26930
    },
    {
      "epoch": 472.5486725663717,
      "grad_norm": 8.4375,
      "learning_rate": 1.2048205114226487e-05,
      "loss": 1.0218,
      "num_input_tokens_seen": 14928112,
      "step": 26935
    },
    {
      "epoch": 472.6371681415929,
      "grad_norm": 9.25,
      "learning_rate": 1.2039808860419102e-05,
      "loss": 1.0347,
      "num_input_tokens_seen": 14930880,
      "step": 26940
    },
    {
      "epoch": 472.72566371681415,
      "grad_norm": 8.125,
      "learning_rate": 1.2031414605236066e-05,
      "loss": 0.9979,
      "num_input_tokens_seen": 14933616,
      "step": 26945
    },
    {
      "epoch": 472.8141592920354,
      "grad_norm": 7.34375,
      "learning_rate": 1.2023022349971862e-05,
      "loss": 0.9147,
      "num_input_tokens_seen": 14936736,
      "step": 26950
    },
    {
      "epoch": 472.9026548672566,
      "grad_norm": 8.375,
      "learning_rate": 1.20146320959207e-05,
      "loss": 1.1706,
      "num_input_tokens_seen": 14939552,
      "step": 26955
    },
    {
      "epoch": 472.9911504424779,
      "grad_norm": 11.5,
      "learning_rate": 1.2006243844376445e-05,
      "loss": 1.0858,
      "num_input_tokens_seen": 14941920,
      "step": 26960
    },
    {
      "epoch": 473.070796460177,
      "grad_norm": 8.6875,
      "learning_rate": 1.1997857596632678e-05,
      "loss": 0.9856,
      "num_input_tokens_seen": 14944728,
      "step": 26965
    },
    {
      "epoch": 473.1592920353982,
      "grad_norm": 9.5,
      "learning_rate": 1.1989473353982672e-05,
      "loss": 1.0822,
      "num_input_tokens_seen": 14947800,
      "step": 26970
    },
    {
      "epoch": 473.24778761061947,
      "grad_norm": 7.78125,
      "learning_rate": 1.198109111771937e-05,
      "loss": 0.9508,
      "num_input_tokens_seen": 14950776,
      "step": 26975
    },
    {
      "epoch": 473.3362831858407,
      "grad_norm": 7.09375,
      "learning_rate": 1.197271088913543e-05,
      "loss": 1.0362,
      "num_input_tokens_seen": 14953832,
      "step": 26980
    },
    {
      "epoch": 473.42477876106193,
      "grad_norm": 8.3125,
      "learning_rate": 1.1964332669523182e-05,
      "loss": 1.0256,
      "num_input_tokens_seen": 14956488,
      "step": 26985
    },
    {
      "epoch": 473.5132743362832,
      "grad_norm": 7.28125,
      "learning_rate": 1.1955956460174645e-05,
      "loss": 1.148,
      "num_input_tokens_seen": 14959368,
      "step": 26990
    },
    {
      "epoch": 473.60176991150445,
      "grad_norm": 6.375,
      "learning_rate": 1.1947582262381552e-05,
      "loss": 1.1415,
      "num_input_tokens_seen": 14961752,
      "step": 26995
    },
    {
      "epoch": 473.69026548672565,
      "grad_norm": 10.4375,
      "learning_rate": 1.1939210077435293e-05,
      "loss": 1.0584,
      "num_input_tokens_seen": 14964568,
      "step": 27000
    },
    {
      "epoch": 473.69026548672565,
      "eval_loss": 1.0287010669708252,
      "eval_runtime": 0.9422,
      "eval_samples_per_second": 26.534,
      "eval_steps_per_second": 13.798,
      "num_input_tokens_seen": 14964568,
      "step": 27000
    },
    {
      "epoch": 473.7787610619469,
      "grad_norm": 9.125,
      "learning_rate": 1.193083990662697e-05,
      "loss": 0.9194,
      "num_input_tokens_seen": 14967528,
      "step": 27005
    },
    {
      "epoch": 473.86725663716817,
      "grad_norm": 10.0,
      "learning_rate": 1.192247175124738e-05,
      "loss": 1.0917,
      "num_input_tokens_seen": 14969944,
      "step": 27010
    },
    {
      "epoch": 473.95575221238937,
      "grad_norm": 6.0625,
      "learning_rate": 1.191410561258698e-05,
      "loss": 1.2309,
      "num_input_tokens_seen": 14972536,
      "step": 27015
    },
    {
      "epoch": 474.0353982300885,
      "grad_norm": 7.65625,
      "learning_rate": 1.1905741491935944e-05,
      "loss": 1.0269,
      "num_input_tokens_seen": 14974680,
      "step": 27020
    },
    {
      "epoch": 474.12389380530976,
      "grad_norm": 8.6875,
      "learning_rate": 1.1897379390584129e-05,
      "loss": 1.0448,
      "num_input_tokens_seen": 14977896,
      "step": 27025
    },
    {
      "epoch": 474.21238938053096,
      "grad_norm": 10.9375,
      "learning_rate": 1.1889019309821062e-05,
      "loss": 1.0248,
      "num_input_tokens_seen": 14980872,
      "step": 27030
    },
    {
      "epoch": 474.3008849557522,
      "grad_norm": 8.0625,
      "learning_rate": 1.188066125093599e-05,
      "loss": 0.9389,
      "num_input_tokens_seen": 14983496,
      "step": 27035
    },
    {
      "epoch": 474.3893805309734,
      "grad_norm": 7.65625,
      "learning_rate": 1.1872305215217811e-05,
      "loss": 1.084,
      "num_input_tokens_seen": 14986584,
      "step": 27040
    },
    {
      "epoch": 474.4778761061947,
      "grad_norm": 8.0625,
      "learning_rate": 1.186395120395514e-05,
      "loss": 0.9329,
      "num_input_tokens_seen": 14989224,
      "step": 27045
    },
    {
      "epoch": 474.56637168141594,
      "grad_norm": 7.53125,
      "learning_rate": 1.1855599218436283e-05,
      "loss": 1.0723,
      "num_input_tokens_seen": 14991768,
      "step": 27050
    },
    {
      "epoch": 474.65486725663715,
      "grad_norm": 7.90625,
      "learning_rate": 1.1847249259949209e-05,
      "loss": 1.0007,
      "num_input_tokens_seen": 14994664,
      "step": 27055
    },
    {
      "epoch": 474.7433628318584,
      "grad_norm": 7.625,
      "learning_rate": 1.1838901329781574e-05,
      "loss": 1.0698,
      "num_input_tokens_seen": 14997352,
      "step": 27060
    },
    {
      "epoch": 474.83185840707966,
      "grad_norm": 9.9375,
      "learning_rate": 1.1830555429220758e-05,
      "loss": 1.0324,
      "num_input_tokens_seen": 15000552,
      "step": 27065
    },
    {
      "epoch": 474.92035398230087,
      "grad_norm": 8.4375,
      "learning_rate": 1.1822211559553784e-05,
      "loss": 1.211,
      "num_input_tokens_seen": 15003320,
      "step": 27070
    },
    {
      "epoch": 475.0,
      "grad_norm": 4.875,
      "learning_rate": 1.18138697220674e-05,
      "loss": 0.9746,
      "num_input_tokens_seen": 15005336,
      "step": 27075
    },
    {
      "epoch": 475.08849557522126,
      "grad_norm": 6.8125,
      "learning_rate": 1.1805529918048e-05,
      "loss": 1.1101,
      "num_input_tokens_seen": 15007976,
      "step": 27080
    },
    {
      "epoch": 475.17699115044246,
      "grad_norm": 9.0,
      "learning_rate": 1.1797192148781702e-05,
      "loss": 0.9711,
      "num_input_tokens_seen": 15010696,
      "step": 27085
    },
    {
      "epoch": 475.2654867256637,
      "grad_norm": 7.21875,
      "learning_rate": 1.1788856415554297e-05,
      "loss": 0.9247,
      "num_input_tokens_seen": 15013736,
      "step": 27090
    },
    {
      "epoch": 475.353982300885,
      "grad_norm": 11.4375,
      "learning_rate": 1.1780522719651249e-05,
      "loss": 1.1892,
      "num_input_tokens_seen": 15016744,
      "step": 27095
    },
    {
      "epoch": 475.4424778761062,
      "grad_norm": 11.25,
      "learning_rate": 1.1772191062357721e-05,
      "loss": 1.0099,
      "num_input_tokens_seen": 15019048,
      "step": 27100
    },
    {
      "epoch": 475.53097345132744,
      "grad_norm": 8.9375,
      "learning_rate": 1.1763861444958573e-05,
      "loss": 1.0772,
      "num_input_tokens_seen": 15021656,
      "step": 27105
    },
    {
      "epoch": 475.6194690265487,
      "grad_norm": 7.5,
      "learning_rate": 1.1755533868738317e-05,
      "loss": 1.0815,
      "num_input_tokens_seen": 15024456,
      "step": 27110
    },
    {
      "epoch": 475.7079646017699,
      "grad_norm": 7.875,
      "learning_rate": 1.1747208334981185e-05,
      "loss": 0.9927,
      "num_input_tokens_seen": 15027704,
      "step": 27115
    },
    {
      "epoch": 475.79646017699116,
      "grad_norm": 13.375,
      "learning_rate": 1.1738884844971067e-05,
      "loss": 1.1935,
      "num_input_tokens_seen": 15030584,
      "step": 27120
    },
    {
      "epoch": 475.88495575221236,
      "grad_norm": 7.53125,
      "learning_rate": 1.1730563399991563e-05,
      "loss": 0.9197,
      "num_input_tokens_seen": 15033560,
      "step": 27125
    },
    {
      "epoch": 475.9734513274336,
      "grad_norm": 9.3125,
      "learning_rate": 1.1722244001325938e-05,
      "loss": 1.0796,
      "num_input_tokens_seen": 15036344,
      "step": 27130
    },
    {
      "epoch": 476.05309734513276,
      "grad_norm": 8.875,
      "learning_rate": 1.1713926650257137e-05,
      "loss": 0.977,
      "num_input_tokens_seen": 15038712,
      "step": 27135
    },
    {
      "epoch": 476.14159292035396,
      "grad_norm": 9.375,
      "learning_rate": 1.170561134806781e-05,
      "loss": 1.0875,
      "num_input_tokens_seen": 15041640,
      "step": 27140
    },
    {
      "epoch": 476.2300884955752,
      "grad_norm": 8.375,
      "learning_rate": 1.1697298096040287e-05,
      "loss": 0.8156,
      "num_input_tokens_seen": 15044696,
      "step": 27145
    },
    {
      "epoch": 476.3185840707965,
      "grad_norm": 8.5625,
      "learning_rate": 1.1688986895456567e-05,
      "loss": 1.1064,
      "num_input_tokens_seen": 15048040,
      "step": 27150
    },
    {
      "epoch": 476.4070796460177,
      "grad_norm": 10.0625,
      "learning_rate": 1.1680677747598349e-05,
      "loss": 1.0161,
      "num_input_tokens_seen": 15051080,
      "step": 27155
    },
    {
      "epoch": 476.49557522123894,
      "grad_norm": 9.0625,
      "learning_rate": 1.1672370653746995e-05,
      "loss": 1.0476,
      "num_input_tokens_seen": 15053912,
      "step": 27160
    },
    {
      "epoch": 476.5840707964602,
      "grad_norm": 7.8125,
      "learning_rate": 1.166406561518357e-05,
      "loss": 0.9743,
      "num_input_tokens_seen": 15056760,
      "step": 27165
    },
    {
      "epoch": 476.6725663716814,
      "grad_norm": 8.375,
      "learning_rate": 1.1655762633188826e-05,
      "loss": 1.0296,
      "num_input_tokens_seen": 15059896,
      "step": 27170
    },
    {
      "epoch": 476.76106194690266,
      "grad_norm": 6.6875,
      "learning_rate": 1.1647461709043172e-05,
      "loss": 1.1346,
      "num_input_tokens_seen": 15062440,
      "step": 27175
    },
    {
      "epoch": 476.8495575221239,
      "grad_norm": 6.96875,
      "learning_rate": 1.1639162844026722e-05,
      "loss": 1.1036,
      "num_input_tokens_seen": 15064968,
      "step": 27180
    },
    {
      "epoch": 476.9380530973451,
      "grad_norm": 9.8125,
      "learning_rate": 1.163086603941927e-05,
      "loss": 1.3441,
      "num_input_tokens_seen": 15067432,
      "step": 27185
    },
    {
      "epoch": 477.01769911504425,
      "grad_norm": 7.875,
      "learning_rate": 1.1622571296500273e-05,
      "loss": 1.0214,
      "num_input_tokens_seen": 15069472,
      "step": 27190
    },
    {
      "epoch": 477.1061946902655,
      "grad_norm": 7.21875,
      "learning_rate": 1.1614278616548904e-05,
      "loss": 0.8355,
      "num_input_tokens_seen": 15072368,
      "step": 27195
    },
    {
      "epoch": 477.1946902654867,
      "grad_norm": 6.875,
      "learning_rate": 1.1605988000843986e-05,
      "loss": 0.9989,
      "num_input_tokens_seen": 15074912,
      "step": 27200
    },
    {
      "epoch": 477.1946902654867,
      "eval_loss": 1.0287010669708252,
      "eval_runtime": 0.9403,
      "eval_samples_per_second": 26.588,
      "eval_steps_per_second": 13.826,
      "num_input_tokens_seen": 15074912,
      "step": 27200
    },
    {
      "epoch": 477.283185840708,
      "grad_norm": 9.1875,
      "learning_rate": 1.1597699450664028e-05,
      "loss": 1.159,
      "num_input_tokens_seen": 15077968,
      "step": 27205
    },
    {
      "epoch": 477.37168141592923,
      "grad_norm": 6.59375,
      "learning_rate": 1.1589412967287252e-05,
      "loss": 1.2069,
      "num_input_tokens_seen": 15080496,
      "step": 27210
    },
    {
      "epoch": 477.46017699115043,
      "grad_norm": 8.0625,
      "learning_rate": 1.1581128551991514e-05,
      "loss": 1.0524,
      "num_input_tokens_seen": 15083408,
      "step": 27215
    },
    {
      "epoch": 477.5486725663717,
      "grad_norm": 8.25,
      "learning_rate": 1.1572846206054383e-05,
      "loss": 1.0674,
      "num_input_tokens_seen": 15086064,
      "step": 27220
    },
    {
      "epoch": 477.6371681415929,
      "grad_norm": 6.8125,
      "learning_rate": 1.1564565930753113e-05,
      "loss": 1.135,
      "num_input_tokens_seen": 15088512,
      "step": 27225
    },
    {
      "epoch": 477.72566371681415,
      "grad_norm": 7.34375,
      "learning_rate": 1.1556287727364606e-05,
      "loss": 1.1121,
      "num_input_tokens_seen": 15091600,
      "step": 27230
    },
    {
      "epoch": 477.8141592920354,
      "grad_norm": 8.1875,
      "learning_rate": 1.1548011597165489e-05,
      "loss": 1.088,
      "num_input_tokens_seen": 15094800,
      "step": 27235
    },
    {
      "epoch": 477.9026548672566,
      "grad_norm": 7.34375,
      "learning_rate": 1.1539737541432019e-05,
      "loss": 1.0129,
      "num_input_tokens_seen": 15097280,
      "step": 27240
    },
    {
      "epoch": 477.9911504424779,
      "grad_norm": 12.125,
      "learning_rate": 1.1531465561440174e-05,
      "loss": 1.0041,
      "num_input_tokens_seen": 15100176,
      "step": 27245
    },
    {
      "epoch": 478.070796460177,
      "grad_norm": 7.71875,
      "learning_rate": 1.1523195658465605e-05,
      "loss": 0.8273,
      "num_input_tokens_seen": 15102528,
      "step": 27250
    },
    {
      "epoch": 478.1592920353982,
      "grad_norm": 8.125,
      "learning_rate": 1.1514927833783618e-05,
      "loss": 1.085,
      "num_input_tokens_seen": 15105296,
      "step": 27255
    },
    {
      "epoch": 478.24778761061947,
      "grad_norm": 8.4375,
      "learning_rate": 1.150666208866922e-05,
      "loss": 1.2526,
      "num_input_tokens_seen": 15108064,
      "step": 27260
    },
    {
      "epoch": 478.3362831858407,
      "grad_norm": 12.0625,
      "learning_rate": 1.1498398424397106e-05,
      "loss": 1.2247,
      "num_input_tokens_seen": 15110496,
      "step": 27265
    },
    {
      "epoch": 478.42477876106193,
      "grad_norm": 8.5,
      "learning_rate": 1.1490136842241628e-05,
      "loss": 1.0644,
      "num_input_tokens_seen": 15113520,
      "step": 27270
    },
    {
      "epoch": 478.5132743362832,
      "grad_norm": 8.75,
      "learning_rate": 1.1481877343476813e-05,
      "loss": 1.2154,
      "num_input_tokens_seen": 15116112,
      "step": 27275
    },
    {
      "epoch": 478.60176991150445,
      "grad_norm": 6.90625,
      "learning_rate": 1.14736199293764e-05,
      "loss": 1.1284,
      "num_input_tokens_seen": 15119136,
      "step": 27280
    },
    {
      "epoch": 478.69026548672565,
      "grad_norm": 6.3125,
      "learning_rate": 1.1465364601213771e-05,
      "loss": 0.9007,
      "num_input_tokens_seen": 15122064,
      "step": 27285
    },
    {
      "epoch": 478.7787610619469,
      "grad_norm": 8.25,
      "learning_rate": 1.1457111360262012e-05,
      "loss": 0.9364,
      "num_input_tokens_seen": 15125024,
      "step": 27290
    },
    {
      "epoch": 478.86725663716817,
      "grad_norm": 10.5,
      "learning_rate": 1.1448860207793869e-05,
      "loss": 0.7226,
      "num_input_tokens_seen": 15127536,
      "step": 27295
    },
    {
      "epoch": 478.95575221238937,
      "grad_norm": 10.875,
      "learning_rate": 1.144061114508177e-05,
      "loss": 1.0219,
      "num_input_tokens_seen": 15130432,
      "step": 27300
    },
    {
      "epoch": 479.0353982300885,
      "grad_norm": 7.4375,
      "learning_rate": 1.1432364173397842e-05,
      "loss": 1.106,
      "num_input_tokens_seen": 15133056,
      "step": 27305
    },
    {
      "epoch": 479.12389380530976,
      "grad_norm": 11.875,
      "learning_rate": 1.1424119294013852e-05,
      "loss": 1.1794,
      "num_input_tokens_seen": 15135696,
      "step": 27310
    },
    {
      "epoch": 479.21238938053096,
      "grad_norm": 6.90625,
      "learning_rate": 1.1415876508201279e-05,
      "loss": 0.9034,
      "num_input_tokens_seen": 15138368,
      "step": 27315
    },
    {
      "epoch": 479.3008849557522,
      "grad_norm": 10.25,
      "learning_rate": 1.140763581723125e-05,
      "loss": 0.8298,
      "num_input_tokens_seen": 15141056,
      "step": 27320
    },
    {
      "epoch": 479.3893805309734,
      "grad_norm": 9.375,
      "learning_rate": 1.1399397222374588e-05,
      "loss": 1.076,
      "num_input_tokens_seen": 15144464,
      "step": 27325
    },
    {
      "epoch": 479.4778761061947,
      "grad_norm": 8.25,
      "learning_rate": 1.1391160724901804e-05,
      "loss": 1.2317,
      "num_input_tokens_seen": 15147408,
      "step": 27330
    },
    {
      "epoch": 479.56637168141594,
      "grad_norm": 7.375,
      "learning_rate": 1.138292632608304e-05,
      "loss": 0.9969,
      "num_input_tokens_seen": 15150336,
      "step": 27335
    },
    {
      "epoch": 479.65486725663715,
      "grad_norm": 6.90625,
      "learning_rate": 1.1374694027188174e-05,
      "loss": 0.9955,
      "num_input_tokens_seen": 15153056,
      "step": 27340
    },
    {
      "epoch": 479.7433628318584,
      "grad_norm": 8.125,
      "learning_rate": 1.1366463829486711e-05,
      "loss": 1.1143,
      "num_input_tokens_seen": 15155600,
      "step": 27345
    },
    {
      "epoch": 479.83185840707966,
      "grad_norm": 5.40625,
      "learning_rate": 1.1358235734247849e-05,
      "loss": 1.0287,
      "num_input_tokens_seen": 15158096,
      "step": 27350
    },
    {
      "epoch": 479.92035398230087,
      "grad_norm": 8.375,
      "learning_rate": 1.1350009742740478e-05,
      "loss": 1.1094,
      "num_input_tokens_seen": 15161344,
      "step": 27355
    },
    {
      "epoch": 480.0,
      "grad_norm": 16.5,
      "learning_rate": 1.134178585623313e-05,
      "loss": 1.1887,
      "num_input_tokens_seen": 15163576,
      "step": 27360
    },
    {
      "epoch": 480.08849557522126,
      "grad_norm": 9.125,
      "learning_rate": 1.1333564075994047e-05,
      "loss": 1.1109,
      "num_input_tokens_seen": 15166248,
      "step": 27365
    },
    {
      "epoch": 480.17699115044246,
      "grad_norm": 6.75,
      "learning_rate": 1.1325344403291133e-05,
      "loss": 0.9401,
      "num_input_tokens_seen": 15169784,
      "step": 27370
    },
    {
      "epoch": 480.2654867256637,
      "grad_norm": 8.8125,
      "learning_rate": 1.1317126839391951e-05,
      "loss": 0.97,
      "num_input_tokens_seen": 15172712,
      "step": 27375
    },
    {
      "epoch": 480.353982300885,
      "grad_norm": 7.5,
      "learning_rate": 1.1308911385563766e-05,
      "loss": 0.8889,
      "num_input_tokens_seen": 15176008,
      "step": 27380
    },
    {
      "epoch": 480.4424778761062,
      "grad_norm": 15.1875,
      "learning_rate": 1.1300698043073494e-05,
      "loss": 1.3117,
      "num_input_tokens_seen": 15178568,
      "step": 27385
    },
    {
      "epoch": 480.53097345132744,
      "grad_norm": 9.25,
      "learning_rate": 1.1292486813187736e-05,
      "loss": 1.0044,
      "num_input_tokens_seen": 15181224,
      "step": 27390
    },
    {
      "epoch": 480.6194690265487,
      "grad_norm": 6.59375,
      "learning_rate": 1.1284277697172782e-05,
      "loss": 1.184,
      "num_input_tokens_seen": 15183736,
      "step": 27395
    },
    {
      "epoch": 480.7079646017699,
      "grad_norm": 10.625,
      "learning_rate": 1.127607069629456e-05,
      "loss": 1.1818,
      "num_input_tokens_seen": 15186488,
      "step": 27400
    },
    {
      "epoch": 480.7079646017699,
      "eval_loss": 1.0287010669708252,
      "eval_runtime": 0.9362,
      "eval_samples_per_second": 26.704,
      "eval_steps_per_second": 13.886,
      "num_input_tokens_seen": 15186488,
      "step": 27400
    },
    {
      "epoch": 480.79646017699116,
      "grad_norm": 10.0625,
      "learning_rate": 1.1267865811818701e-05,
      "loss": 1.0344,
      "num_input_tokens_seen": 15188968,
      "step": 27405
    },
    {
      "epoch": 480.88495575221236,
      "grad_norm": 9.0,
      "learning_rate": 1.1259663045010513e-05,
      "loss": 0.9932,
      "num_input_tokens_seen": 15191960,
      "step": 27410
    },
    {
      "epoch": 480.9734513274336,
      "grad_norm": 7.53125,
      "learning_rate": 1.1251462397134957e-05,
      "loss": 1.0923,
      "num_input_tokens_seen": 15194552,
      "step": 27415
    },
    {
      "epoch": 481.05309734513276,
      "grad_norm": 7.1875,
      "learning_rate": 1.1243263869456664e-05,
      "loss": 0.8634,
      "num_input_tokens_seen": 15197416,
      "step": 27420
    },
    {
      "epoch": 481.14159292035396,
      "grad_norm": 9.625,
      "learning_rate": 1.1235067463239967e-05,
      "loss": 1.2898,
      "num_input_tokens_seen": 15199864,
      "step": 27425
    },
    {
      "epoch": 481.2300884955752,
      "grad_norm": 8.8125,
      "learning_rate": 1.122687317974884e-05,
      "loss": 1.0533,
      "num_input_tokens_seen": 15202632,
      "step": 27430
    },
    {
      "epoch": 481.3185840707965,
      "grad_norm": 8.25,
      "learning_rate": 1.1218681020246963e-05,
      "loss": 0.958,
      "num_input_tokens_seen": 15205560,
      "step": 27435
    },
    {
      "epoch": 481.4070796460177,
      "grad_norm": 9.125,
      "learning_rate": 1.1210490985997652e-05,
      "loss": 1.201,
      "num_input_tokens_seen": 15208264,
      "step": 27440
    },
    {
      "epoch": 481.49557522123894,
      "grad_norm": 8.4375,
      "learning_rate": 1.1202303078263917e-05,
      "loss": 1.0032,
      "num_input_tokens_seen": 15210984,
      "step": 27445
    },
    {
      "epoch": 481.5840707964602,
      "grad_norm": 8.4375,
      "learning_rate": 1.1194117298308451e-05,
      "loss": 1.1526,
      "num_input_tokens_seen": 15213512,
      "step": 27450
    },
    {
      "epoch": 481.6725663716814,
      "grad_norm": 8.5,
      "learning_rate": 1.1185933647393585e-05,
      "loss": 0.9114,
      "num_input_tokens_seen": 15216520,
      "step": 27455
    },
    {
      "epoch": 481.76106194690266,
      "grad_norm": 8.9375,
      "learning_rate": 1.1177752126781354e-05,
      "loss": 1.1412,
      "num_input_tokens_seen": 15219512,
      "step": 27460
    },
    {
      "epoch": 481.8495575221239,
      "grad_norm": 6.90625,
      "learning_rate": 1.1169572737733441e-05,
      "loss": 0.9096,
      "num_input_tokens_seen": 15222232,
      "step": 27465
    },
    {
      "epoch": 481.9380530973451,
      "grad_norm": 7.21875,
      "learning_rate": 1.1161395481511216e-05,
      "loss": 0.9679,
      "num_input_tokens_seen": 15224984,
      "step": 27470
    },
    {
      "epoch": 482.01769911504425,
      "grad_norm": 7.3125,
      "learning_rate": 1.1153220359375722e-05,
      "loss": 1.0823,
      "num_input_tokens_seen": 15227456,
      "step": 27475
    },
    {
      "epoch": 482.1061946902655,
      "grad_norm": 9.5,
      "learning_rate": 1.114504737258765e-05,
      "loss": 1.161,
      "num_input_tokens_seen": 15230176,
      "step": 27480
    },
    {
      "epoch": 482.1946902654867,
      "grad_norm": 9.4375,
      "learning_rate": 1.1136876522407393e-05,
      "loss": 1.1083,
      "num_input_tokens_seen": 15233024,
      "step": 27485
    },
    {
      "epoch": 482.283185840708,
      "grad_norm": 8.1875,
      "learning_rate": 1.1128707810094985e-05,
      "loss": 0.8887,
      "num_input_tokens_seen": 15235808,
      "step": 27490
    },
    {
      "epoch": 482.37168141592923,
      "grad_norm": 7.03125,
      "learning_rate": 1.1120541236910157e-05,
      "loss": 0.9942,
      "num_input_tokens_seen": 15238400,
      "step": 27495
    },
    {
      "epoch": 482.46017699115043,
      "grad_norm": 8.0625,
      "learning_rate": 1.111237680411229e-05,
      "loss": 1.1655,
      "num_input_tokens_seen": 15240928,
      "step": 27500
    },
    {
      "epoch": 482.5486725663717,
      "grad_norm": 7.90625,
      "learning_rate": 1.1104214512960433e-05,
      "loss": 0.9784,
      "num_input_tokens_seen": 15243936,
      "step": 27505
    },
    {
      "epoch": 482.6371681415929,
      "grad_norm": 4.40625,
      "learning_rate": 1.1096054364713327e-05,
      "loss": 0.9373,
      "num_input_tokens_seen": 15246832,
      "step": 27510
    },
    {
      "epoch": 482.72566371681415,
      "grad_norm": 12.1875,
      "learning_rate": 1.1087896360629371e-05,
      "loss": 1.1475,
      "num_input_tokens_seen": 15250208,
      "step": 27515
    },
    {
      "epoch": 482.8141592920354,
      "grad_norm": 12.6875,
      "learning_rate": 1.107974050196662e-05,
      "loss": 1.1069,
      "num_input_tokens_seen": 15252864,
      "step": 27520
    },
    {
      "epoch": 482.9026548672566,
      "grad_norm": 8.1875,
      "learning_rate": 1.1071586789982816e-05,
      "loss": 0.9708,
      "num_input_tokens_seen": 15255744,
      "step": 27525
    },
    {
      "epoch": 482.9911504424779,
      "grad_norm": 9.0625,
      "learning_rate": 1.1063435225935373e-05,
      "loss": 1.1773,
      "num_input_tokens_seen": 15258400,
      "step": 27530
    },
    {
      "epoch": 483.070796460177,
      "grad_norm": 8.4375,
      "learning_rate": 1.1055285811081348e-05,
      "loss": 1.0005,
      "num_input_tokens_seen": 15260808,
      "step": 27535
    },
    {
      "epoch": 483.1592920353982,
      "grad_norm": 6.1875,
      "learning_rate": 1.1047138546677499e-05,
      "loss": 0.9322,
      "num_input_tokens_seen": 15263784,
      "step": 27540
    },
    {
      "epoch": 483.24778761061947,
      "grad_norm": 10.375,
      "learning_rate": 1.1038993433980219e-05,
      "loss": 1.0733,
      "num_input_tokens_seen": 15266632,
      "step": 27545
    },
    {
      "epoch": 483.3362831858407,
      "grad_norm": 5.8125,
      "learning_rate": 1.1030850474245597e-05,
      "loss": 0.9046,
      "num_input_tokens_seen": 15269064,
      "step": 27550
    },
    {
      "epoch": 483.42477876106193,
      "grad_norm": 11.0,
      "learning_rate": 1.102270966872939e-05,
      "loss": 1.224,
      "num_input_tokens_seen": 15272296,
      "step": 27555
    },
    {
      "epoch": 483.5132743362832,
      "grad_norm": 8.3125,
      "learning_rate": 1.1014571018687e-05,
      "loss": 1.1571,
      "num_input_tokens_seen": 15274888,
      "step": 27560
    },
    {
      "epoch": 483.60176991150445,
      "grad_norm": 10.1875,
      "learning_rate": 1.1006434525373502e-05,
      "loss": 1.165,
      "num_input_tokens_seen": 15277416,
      "step": 27565
    },
    {
      "epoch": 483.69026548672565,
      "grad_norm": 6.9375,
      "learning_rate": 1.0998300190043664e-05,
      "loss": 0.9781,
      "num_input_tokens_seen": 15280312,
      "step": 27570
    },
    {
      "epoch": 483.7787610619469,
      "grad_norm": 9.3125,
      "learning_rate": 1.0990168013951882e-05,
      "loss": 1.0613,
      "num_input_tokens_seen": 15282776,
      "step": 27575
    },
    {
      "epoch": 483.86725663716817,
      "grad_norm": 6.8125,
      "learning_rate": 1.0982037998352263e-05,
      "loss": 0.9124,
      "num_input_tokens_seen": 15285384,
      "step": 27580
    },
    {
      "epoch": 483.95575221238937,
      "grad_norm": 9.1875,
      "learning_rate": 1.0973910144498534e-05,
      "loss": 1.209,
      "num_input_tokens_seen": 15288168,
      "step": 27585
    },
    {
      "epoch": 484.0353982300885,
      "grad_norm": 9.5,
      "learning_rate": 1.0965784453644123e-05,
      "loss": 1.2062,
      "num_input_tokens_seen": 15290832,
      "step": 27590
    },
    {
      "epoch": 484.12389380530976,
      "grad_norm": 6.78125,
      "learning_rate": 1.0957660927042127e-05,
      "loss": 0.9258,
      "num_input_tokens_seen": 15294256,
      "step": 27595
    },
    {
      "epoch": 484.21238938053096,
      "grad_norm": 7.90625,
      "learning_rate": 1.094953956594527e-05,
      "loss": 1.2094,
      "num_input_tokens_seen": 15297600,
      "step": 27600
    },
    {
      "epoch": 484.21238938053096,
      "eval_loss": 1.0287010669708252,
      "eval_runtime": 0.9345,
      "eval_samples_per_second": 26.752,
      "eval_steps_per_second": 13.911,
      "num_input_tokens_seen": 15297600,
      "step": 27600
    },
    {
      "epoch": 484.3008849557522,
      "grad_norm": 8.25,
      "learning_rate": 1.0941420371605981e-05,
      "loss": 0.9731,
      "num_input_tokens_seen": 15300480,
      "step": 27605
    },
    {
      "epoch": 484.3893805309734,
      "grad_norm": 8.375,
      "learning_rate": 1.0933303345276354e-05,
      "loss": 1.1356,
      "num_input_tokens_seen": 15303040,
      "step": 27610
    },
    {
      "epoch": 484.4778761061947,
      "grad_norm": 7.21875,
      "learning_rate": 1.0925188488208112e-05,
      "loss": 0.9195,
      "num_input_tokens_seen": 15305552,
      "step": 27615
    },
    {
      "epoch": 484.56637168141594,
      "grad_norm": 9.375,
      "learning_rate": 1.0917075801652694e-05,
      "loss": 1.1624,
      "num_input_tokens_seen": 15308272,
      "step": 27620
    },
    {
      "epoch": 484.65486725663715,
      "grad_norm": 6.90625,
      "learning_rate": 1.0908965286861151e-05,
      "loss": 0.9332,
      "num_input_tokens_seen": 15311168,
      "step": 27625
    },
    {
      "epoch": 484.7433628318584,
      "grad_norm": 9.8125,
      "learning_rate": 1.090085694508425e-05,
      "loss": 1.173,
      "num_input_tokens_seen": 15313872,
      "step": 27630
    },
    {
      "epoch": 484.83185840707966,
      "grad_norm": 7.96875,
      "learning_rate": 1.089275077757238e-05,
      "loss": 1.0743,
      "num_input_tokens_seen": 15316704,
      "step": 27635
    },
    {
      "epoch": 484.92035398230087,
      "grad_norm": 12.25,
      "learning_rate": 1.0884646785575633e-05,
      "loss": 1.1568,
      "num_input_tokens_seen": 15319216,
      "step": 27640
    },
    {
      "epoch": 485.0,
      "grad_norm": 13.9375,
      "learning_rate": 1.0876544970343728e-05,
      "loss": 0.969,
      "num_input_tokens_seen": 15321608,
      "step": 27645
    },
    {
      "epoch": 485.08849557522126,
      "grad_norm": 8.75,
      "learning_rate": 1.0868445333126082e-05,
      "loss": 0.8807,
      "num_input_tokens_seen": 15324568,
      "step": 27650
    },
    {
      "epoch": 485.17699115044246,
      "grad_norm": 8.8125,
      "learning_rate": 1.0860347875171745e-05,
      "loss": 1.1732,
      "num_input_tokens_seen": 15327112,
      "step": 27655
    },
    {
      "epoch": 485.2654867256637,
      "grad_norm": 7.28125,
      "learning_rate": 1.0852252597729465e-05,
      "loss": 1.1243,
      "num_input_tokens_seen": 15329576,
      "step": 27660
    },
    {
      "epoch": 485.353982300885,
      "grad_norm": 8.0625,
      "learning_rate": 1.0844159502047615e-05,
      "loss": 0.9051,
      "num_input_tokens_seen": 15332168,
      "step": 27665
    },
    {
      "epoch": 485.4424778761062,
      "grad_norm": 6.34375,
      "learning_rate": 1.0836068589374265e-05,
      "loss": 1.0602,
      "num_input_tokens_seen": 15334728,
      "step": 27670
    },
    {
      "epoch": 485.53097345132744,
      "grad_norm": 7.0625,
      "learning_rate": 1.0827979860957144e-05,
      "loss": 1.2038,
      "num_input_tokens_seen": 15337336,
      "step": 27675
    },
    {
      "epoch": 485.6194690265487,
      "grad_norm": 8.3125,
      "learning_rate": 1.0819893318043615e-05,
      "loss": 1.0153,
      "num_input_tokens_seen": 15340232,
      "step": 27680
    },
    {
      "epoch": 485.7079646017699,
      "grad_norm": 8.625,
      "learning_rate": 1.0811808961880734e-05,
      "loss": 1.0162,
      "num_input_tokens_seen": 15343368,
      "step": 27685
    },
    {
      "epoch": 485.79646017699116,
      "grad_norm": 9.3125,
      "learning_rate": 1.080372679371522e-05,
      "loss": 1.0529,
      "num_input_tokens_seen": 15346488,
      "step": 27690
    },
    {
      "epoch": 485.88495575221236,
      "grad_norm": 7.28125,
      "learning_rate": 1.0795646814793428e-05,
      "loss": 0.9867,
      "num_input_tokens_seen": 15349240,
      "step": 27695
    },
    {
      "epoch": 485.9734513274336,
      "grad_norm": 9.6875,
      "learning_rate": 1.078756902636141e-05,
      "loss": 1.0565,
      "num_input_tokens_seen": 15352072,
      "step": 27700
    },
    {
      "epoch": 486.05309734513276,
      "grad_norm": 8.875,
      "learning_rate": 1.077949342966485e-05,
      "loss": 1.5002,
      "num_input_tokens_seen": 15354512,
      "step": 27705
    },
    {
      "epoch": 486.14159292035396,
      "grad_norm": 7.375,
      "learning_rate": 1.0771420025949103e-05,
      "loss": 1.0416,
      "num_input_tokens_seen": 15357232,
      "step": 27710
    },
    {
      "epoch": 486.2300884955752,
      "grad_norm": 8.8125,
      "learning_rate": 1.0763348816459204e-05,
      "loss": 1.0543,
      "num_input_tokens_seen": 15360352,
      "step": 27715
    },
    {
      "epoch": 486.3185840707965,
      "grad_norm": 7.21875,
      "learning_rate": 1.0755279802439816e-05,
      "loss": 1.0401,
      "num_input_tokens_seen": 15363120,
      "step": 27720
    },
    {
      "epoch": 486.4070796460177,
      "grad_norm": 8.8125,
      "learning_rate": 1.0747212985135293e-05,
      "loss": 1.0812,
      "num_input_tokens_seen": 15365760,
      "step": 27725
    },
    {
      "epoch": 486.49557522123894,
      "grad_norm": 8.4375,
      "learning_rate": 1.073914836578965e-05,
      "loss": 0.9524,
      "num_input_tokens_seen": 15367968,
      "step": 27730
    },
    {
      "epoch": 486.5840707964602,
      "grad_norm": 6.84375,
      "learning_rate": 1.0731085945646529e-05,
      "loss": 1.0846,
      "num_input_tokens_seen": 15371088,
      "step": 27735
    },
    {
      "epoch": 486.6725663716814,
      "grad_norm": 8.125,
      "learning_rate": 1.0723025725949285e-05,
      "loss": 0.9848,
      "num_input_tokens_seen": 15374064,
      "step": 27740
    },
    {
      "epoch": 486.76106194690266,
      "grad_norm": 8.3125,
      "learning_rate": 1.0714967707940875e-05,
      "loss": 1.1319,
      "num_input_tokens_seen": 15376656,
      "step": 27745
    },
    {
      "epoch": 486.8495575221239,
      "grad_norm": 7.6875,
      "learning_rate": 1.0706911892863963e-05,
      "loss": 1.0142,
      "num_input_tokens_seen": 15379408,
      "step": 27750
    },
    {
      "epoch": 486.9380530973451,
      "grad_norm": 9.1875,
      "learning_rate": 1.0698858281960866e-05,
      "loss": 1.046,
      "num_input_tokens_seen": 15382432,
      "step": 27755
    },
    {
      "epoch": 487.01769911504425,
      "grad_norm": 6.90625,
      "learning_rate": 1.069080687647353e-05,
      "loss": 0.8901,
      "num_input_tokens_seen": 15385064,
      "step": 27760
    },
    {
      "epoch": 487.1061946902655,
      "grad_norm": 6.875,
      "learning_rate": 1.0682757677643596e-05,
      "loss": 1.1117,
      "num_input_tokens_seen": 15387640,
      "step": 27765
    },
    {
      "epoch": 487.1946902654867,
      "grad_norm": 9.0625,
      "learning_rate": 1.0674710686712359e-05,
      "loss": 1.0627,
      "num_input_tokens_seen": 15390472,
      "step": 27770
    },
    {
      "epoch": 487.283185840708,
      "grad_norm": 8.6875,
      "learning_rate": 1.0666665904920756e-05,
      "loss": 1.0857,
      "num_input_tokens_seen": 15393176,
      "step": 27775
    },
    {
      "epoch": 487.37168141592923,
      "grad_norm": 8.25,
      "learning_rate": 1.0658623333509385e-05,
      "loss": 1.0199,
      "num_input_tokens_seen": 15396280,
      "step": 27780
    },
    {
      "epoch": 487.46017699115043,
      "grad_norm": 9.625,
      "learning_rate": 1.0650582973718532e-05,
      "loss": 1.0517,
      "num_input_tokens_seen": 15399320,
      "step": 27785
    },
    {
      "epoch": 487.5486725663717,
      "grad_norm": 6.96875,
      "learning_rate": 1.0642544826788098e-05,
      "loss": 1.1865,
      "num_input_tokens_seen": 15401976,
      "step": 27790
    },
    {
      "epoch": 487.6371681415929,
      "grad_norm": 9.75,
      "learning_rate": 1.063450889395769e-05,
      "loss": 1.1655,
      "num_input_tokens_seen": 15404472,
      "step": 27795
    },
    {
      "epoch": 487.72566371681415,
      "grad_norm": 7.8125,
      "learning_rate": 1.062647517646653e-05,
      "loss": 0.9714,
      "num_input_tokens_seen": 15407784,
      "step": 27800
    },
    {
      "epoch": 487.72566371681415,
      "eval_loss": 1.0287010669708252,
      "eval_runtime": 0.9382,
      "eval_samples_per_second": 26.646,
      "eval_steps_per_second": 13.856,
      "num_input_tokens_seen": 15407784,
      "step": 27800
    },
    {
      "epoch": 487.8141592920354,
      "grad_norm": 8.3125,
      "learning_rate": 1.0618443675553527e-05,
      "loss": 1.1562,
      "num_input_tokens_seen": 15410920,
      "step": 27805
    },
    {
      "epoch": 487.9026548672566,
      "grad_norm": 8.75,
      "learning_rate": 1.0610414392457247e-05,
      "loss": 0.9831,
      "num_input_tokens_seen": 15413384,
      "step": 27810
    },
    {
      "epoch": 487.9911504424779,
      "grad_norm": 7.65625,
      "learning_rate": 1.0602387328415888e-05,
      "loss": 0.8359,
      "num_input_tokens_seen": 15416520,
      "step": 27815
    },
    {
      "epoch": 488.070796460177,
      "grad_norm": 7.15625,
      "learning_rate": 1.0594362484667347e-05,
      "loss": 1.1505,
      "num_input_tokens_seen": 15418544,
      "step": 27820
    },
    {
      "epoch": 488.1592920353982,
      "grad_norm": 7.375,
      "learning_rate": 1.0586339862449132e-05,
      "loss": 1.112,
      "num_input_tokens_seen": 15421152,
      "step": 27825
    },
    {
      "epoch": 488.24778761061947,
      "grad_norm": 9.5,
      "learning_rate": 1.0578319462998445e-05,
      "loss": 0.877,
      "num_input_tokens_seen": 15424080,
      "step": 27830
    },
    {
      "epoch": 488.3362831858407,
      "grad_norm": 10.4375,
      "learning_rate": 1.057030128755214e-05,
      "loss": 1.003,
      "num_input_tokens_seen": 15426976,
      "step": 27835
    },
    {
      "epoch": 488.42477876106193,
      "grad_norm": 11.25,
      "learning_rate": 1.0562285337346703e-05,
      "loss": 1.0414,
      "num_input_tokens_seen": 15430368,
      "step": 27840
    },
    {
      "epoch": 488.5132743362832,
      "grad_norm": 8.9375,
      "learning_rate": 1.0554271613618308e-05,
      "loss": 1.0253,
      "num_input_tokens_seen": 15433728,
      "step": 27845
    },
    {
      "epoch": 488.60176991150445,
      "grad_norm": 7.59375,
      "learning_rate": 1.054626011760276e-05,
      "loss": 1.0823,
      "num_input_tokens_seen": 15436608,
      "step": 27850
    },
    {
      "epoch": 488.69026548672565,
      "grad_norm": 10.125,
      "learning_rate": 1.0538250850535549e-05,
      "loss": 0.9551,
      "num_input_tokens_seen": 15439568,
      "step": 27855
    },
    {
      "epoch": 488.7787610619469,
      "grad_norm": 5.6875,
      "learning_rate": 1.0530243813651794e-05,
      "loss": 1.2004,
      "num_input_tokens_seen": 15441856,
      "step": 27860
    },
    {
      "epoch": 488.86725663716817,
      "grad_norm": 8.9375,
      "learning_rate": 1.0522239008186271e-05,
      "loss": 0.948,
      "num_input_tokens_seen": 15444240,
      "step": 27865
    },
    {
      "epoch": 488.95575221238937,
      "grad_norm": 8.75,
      "learning_rate": 1.0514236435373434e-05,
      "loss": 1.1568,
      "num_input_tokens_seen": 15446992,
      "step": 27870
    },
    {
      "epoch": 489.0353982300885,
      "grad_norm": 8.5625,
      "learning_rate": 1.0506236096447386e-05,
      "loss": 1.1291,
      "num_input_tokens_seen": 15449392,
      "step": 27875
    },
    {
      "epoch": 489.12389380530976,
      "grad_norm": 9.875,
      "learning_rate": 1.049823799264186e-05,
      "loss": 1.1758,
      "num_input_tokens_seen": 15452048,
      "step": 27880
    },
    {
      "epoch": 489.21238938053096,
      "grad_norm": 8.1875,
      "learning_rate": 1.049024212519028e-05,
      "loss": 0.96,
      "num_input_tokens_seen": 15455072,
      "step": 27885
    },
    {
      "epoch": 489.3008849557522,
      "grad_norm": 9.6875,
      "learning_rate": 1.0482248495325713e-05,
      "loss": 1.1413,
      "num_input_tokens_seen": 15457584,
      "step": 27890
    },
    {
      "epoch": 489.3893805309734,
      "grad_norm": 7.75,
      "learning_rate": 1.047425710428086e-05,
      "loss": 1.1667,
      "num_input_tokens_seen": 15460304,
      "step": 27895
    },
    {
      "epoch": 489.4778761061947,
      "grad_norm": 6.59375,
      "learning_rate": 1.0466267953288114e-05,
      "loss": 1.0101,
      "num_input_tokens_seen": 15462848,
      "step": 27900
    },
    {
      "epoch": 489.56637168141594,
      "grad_norm": 6.0,
      "learning_rate": 1.0458281043579482e-05,
      "loss": 0.9385,
      "num_input_tokens_seen": 15465920,
      "step": 27905
    },
    {
      "epoch": 489.65486725663715,
      "grad_norm": 8.625,
      "learning_rate": 1.0450296376386657e-05,
      "loss": 1.0202,
      "num_input_tokens_seen": 15469024,
      "step": 27910
    },
    {
      "epoch": 489.7433628318584,
      "grad_norm": 8.8125,
      "learning_rate": 1.044231395294098e-05,
      "loss": 0.8762,
      "num_input_tokens_seen": 15471312,
      "step": 27915
    },
    {
      "epoch": 489.83185840707966,
      "grad_norm": 9.1875,
      "learning_rate": 1.0434333774473435e-05,
      "loss": 1.0393,
      "num_input_tokens_seen": 15474720,
      "step": 27920
    },
    {
      "epoch": 489.92035398230087,
      "grad_norm": 8.1875,
      "learning_rate": 1.0426355842214657e-05,
      "loss": 1.1874,
      "num_input_tokens_seen": 15477168,
      "step": 27925
    },
    {
      "epoch": 490.0,
      "grad_norm": 7.84375,
      "learning_rate": 1.0418380157394963e-05,
      "loss": 1.0873,
      "num_input_tokens_seen": 15479480,
      "step": 27930
    },
    {
      "epoch": 490.08849557522126,
      "grad_norm": 7.0625,
      "learning_rate": 1.0410406721244281e-05,
      "loss": 1.0816,
      "num_input_tokens_seen": 15481720,
      "step": 27935
    },
    {
      "epoch": 490.17699115044246,
      "grad_norm": 10.1875,
      "learning_rate": 1.0402435534992238e-05,
      "loss": 1.0472,
      "num_input_tokens_seen": 15484136,
      "step": 27940
    },
    {
      "epoch": 490.2654867256637,
      "grad_norm": 6.09375,
      "learning_rate": 1.0394466599868071e-05,
      "loss": 0.9833,
      "num_input_tokens_seen": 15487096,
      "step": 27945
    },
    {
      "epoch": 490.353982300885,
      "grad_norm": 8.375,
      "learning_rate": 1.0386499917100697e-05,
      "loss": 1.1649,
      "num_input_tokens_seen": 15490040,
      "step": 27950
    },
    {
      "epoch": 490.4424778761062,
      "grad_norm": 10.0,
      "learning_rate": 1.0378535487918692e-05,
      "loss": 1.1471,
      "num_input_tokens_seen": 15492712,
      "step": 27955
    },
    {
      "epoch": 490.53097345132744,
      "grad_norm": 8.0,
      "learning_rate": 1.037057331355025e-05,
      "loss": 1.1345,
      "num_input_tokens_seen": 15495480,
      "step": 27960
    },
    {
      "epoch": 490.6194690265487,
      "grad_norm": 6.9375,
      "learning_rate": 1.0362613395223247e-05,
      "loss": 1.0766,
      "num_input_tokens_seen": 15498584,
      "step": 27965
    },
    {
      "epoch": 490.7079646017699,
      "grad_norm": 9.875,
      "learning_rate": 1.0354655734165212e-05,
      "loss": 1.094,
      "num_input_tokens_seen": 15501768,
      "step": 27970
    },
    {
      "epoch": 490.79646017699116,
      "grad_norm": 7.28125,
      "learning_rate": 1.03467003316033e-05,
      "loss": 0.8924,
      "num_input_tokens_seen": 15504392,
      "step": 27975
    },
    {
      "epoch": 490.88495575221236,
      "grad_norm": 13.875,
      "learning_rate": 1.033874718876435e-05,
      "loss": 0.9378,
      "num_input_tokens_seen": 15507688,
      "step": 27980
    },
    {
      "epoch": 490.9734513274336,
      "grad_norm": 7.75,
      "learning_rate": 1.0330796306874818e-05,
      "loss": 1.0276,
      "num_input_tokens_seen": 15510280,
      "step": 27985
    },
    {
      "epoch": 491.05309734513276,
      "grad_norm": 12.125,
      "learning_rate": 1.032284768716085e-05,
      "loss": 1.2198,
      "num_input_tokens_seen": 15512592,
      "step": 27990
    },
    {
      "epoch": 491.14159292035396,
      "grad_norm": 9.5,
      "learning_rate": 1.0314901330848206e-05,
      "loss": 1.1064,
      "num_input_tokens_seen": 15515904,
      "step": 27995
    },
    {
      "epoch": 491.2300884955752,
      "grad_norm": 5.34375,
      "learning_rate": 1.030695723916233e-05,
      "loss": 0.9373,
      "num_input_tokens_seen": 15518800,
      "step": 28000
    },
    {
      "epoch": 491.2300884955752,
      "eval_loss": 1.0287010669708252,
      "eval_runtime": 0.9359,
      "eval_samples_per_second": 26.713,
      "eval_steps_per_second": 13.891,
      "num_input_tokens_seen": 15518800,
      "step": 28000
    },
    {
      "epoch": 491.3185840707965,
      "grad_norm": 9.1875,
      "learning_rate": 1.0299015413328289e-05,
      "loss": 1.0162,
      "num_input_tokens_seen": 15522032,
      "step": 28005
    },
    {
      "epoch": 491.4070796460177,
      "grad_norm": 8.4375,
      "learning_rate": 1.0291075854570809e-05,
      "loss": 0.9933,
      "num_input_tokens_seen": 15524832,
      "step": 28010
    },
    {
      "epoch": 491.49557522123894,
      "grad_norm": 9.5,
      "learning_rate": 1.0283138564114275e-05,
      "loss": 1.0431,
      "num_input_tokens_seen": 15527552,
      "step": 28015
    },
    {
      "epoch": 491.5840707964602,
      "grad_norm": 8.0,
      "learning_rate": 1.027520354318273e-05,
      "loss": 1.0297,
      "num_input_tokens_seen": 15530160,
      "step": 28020
    },
    {
      "epoch": 491.6725663716814,
      "grad_norm": 7.21875,
      "learning_rate": 1.0267270792999828e-05,
      "loss": 0.979,
      "num_input_tokens_seen": 15532800,
      "step": 28025
    },
    {
      "epoch": 491.76106194690266,
      "grad_norm": 8.4375,
      "learning_rate": 1.0259340314788919e-05,
      "loss": 0.9972,
      "num_input_tokens_seen": 15535776,
      "step": 28030
    },
    {
      "epoch": 491.8495575221239,
      "grad_norm": 8.25,
      "learning_rate": 1.0251412109772979e-05,
      "loss": 0.9543,
      "num_input_tokens_seen": 15538576,
      "step": 28035
    },
    {
      "epoch": 491.9380530973451,
      "grad_norm": 12.875,
      "learning_rate": 1.0243486179174627e-05,
      "loss": 1.2005,
      "num_input_tokens_seen": 15540768,
      "step": 28040
    },
    {
      "epoch": 492.01769911504425,
      "grad_norm": 8.25,
      "learning_rate": 1.0235562524216158e-05,
      "loss": 1.1423,
      "num_input_tokens_seen": 15543088,
      "step": 28045
    },
    {
      "epoch": 492.1061946902655,
      "grad_norm": 6.375,
      "learning_rate": 1.022764114611948e-05,
      "loss": 1.0761,
      "num_input_tokens_seen": 15545664,
      "step": 28050
    },
    {
      "epoch": 492.1946902654867,
      "grad_norm": 7.71875,
      "learning_rate": 1.0219722046106178e-05,
      "loss": 1.0665,
      "num_input_tokens_seen": 15548512,
      "step": 28055
    },
    {
      "epoch": 492.283185840708,
      "grad_norm": 7.96875,
      "learning_rate": 1.0211805225397486e-05,
      "loss": 1.0197,
      "num_input_tokens_seen": 15551072,
      "step": 28060
    },
    {
      "epoch": 492.37168141592923,
      "grad_norm": 8.875,
      "learning_rate": 1.020389068521426e-05,
      "loss": 1.0633,
      "num_input_tokens_seen": 15554480,
      "step": 28065
    },
    {
      "epoch": 492.46017699115043,
      "grad_norm": 7.375,
      "learning_rate": 1.0195978426777039e-05,
      "loss": 0.9311,
      "num_input_tokens_seen": 15557056,
      "step": 28070
    },
    {
      "epoch": 492.5486725663717,
      "grad_norm": 7.59375,
      "learning_rate": 1.0188068451305982e-05,
      "loss": 1.0174,
      "num_input_tokens_seen": 15560208,
      "step": 28075
    },
    {
      "epoch": 492.6371681415929,
      "grad_norm": 10.4375,
      "learning_rate": 1.0180160760020902e-05,
      "loss": 1.1443,
      "num_input_tokens_seen": 15563056,
      "step": 28080
    },
    {
      "epoch": 492.72566371681415,
      "grad_norm": 9.0625,
      "learning_rate": 1.0172255354141278e-05,
      "loss": 0.9102,
      "num_input_tokens_seen": 15565904,
      "step": 28085
    },
    {
      "epoch": 492.8141592920354,
      "grad_norm": 9.375,
      "learning_rate": 1.0164352234886205e-05,
      "loss": 1.0687,
      "num_input_tokens_seen": 15568496,
      "step": 28090
    },
    {
      "epoch": 492.9026548672566,
      "grad_norm": 8.875,
      "learning_rate": 1.0156451403474454e-05,
      "loss": 1.0552,
      "num_input_tokens_seen": 15571216,
      "step": 28095
    },
    {
      "epoch": 492.9911504424779,
      "grad_norm": 9.0,
      "learning_rate": 1.0148552861124443e-05,
      "loss": 1.273,
      "num_input_tokens_seen": 15574160,
      "step": 28100
    },
    {
      "epoch": 493.070796460177,
      "grad_norm": 8.0625,
      "learning_rate": 1.0140656609054205e-05,
      "loss": 1.1139,
      "num_input_tokens_seen": 15576784,
      "step": 28105
    },
    {
      "epoch": 493.1592920353982,
      "grad_norm": 6.90625,
      "learning_rate": 1.0132762648481455e-05,
      "loss": 1.1125,
      "num_input_tokens_seen": 15579456,
      "step": 28110
    },
    {
      "epoch": 493.24778761061947,
      "grad_norm": 12.1875,
      "learning_rate": 1.0124870980623543e-05,
      "loss": 1.0391,
      "num_input_tokens_seen": 15582704,
      "step": 28115
    },
    {
      "epoch": 493.3362831858407,
      "grad_norm": 10.0625,
      "learning_rate": 1.0116981606697453e-05,
      "loss": 0.9574,
      "num_input_tokens_seen": 15585424,
      "step": 28120
    },
    {
      "epoch": 493.42477876106193,
      "grad_norm": 10.9375,
      "learning_rate": 1.0109094527919838e-05,
      "loss": 0.904,
      "num_input_tokens_seen": 15588400,
      "step": 28125
    },
    {
      "epoch": 493.5132743362832,
      "grad_norm": 8.4375,
      "learning_rate": 1.010120974550697e-05,
      "loss": 1.0859,
      "num_input_tokens_seen": 15591744,
      "step": 28130
    },
    {
      "epoch": 493.60176991150445,
      "grad_norm": 9.1875,
      "learning_rate": 1.0093327260674795e-05,
      "loss": 0.927,
      "num_input_tokens_seen": 15594224,
      "step": 28135
    },
    {
      "epoch": 493.69026548672565,
      "grad_norm": 8.6875,
      "learning_rate": 1.0085447074638878e-05,
      "loss": 1.0672,
      "num_input_tokens_seen": 15597280,
      "step": 28140
    },
    {
      "epoch": 493.7787610619469,
      "grad_norm": 7.90625,
      "learning_rate": 1.0077569188614461e-05,
      "loss": 0.9676,
      "num_input_tokens_seen": 15599920,
      "step": 28145
    },
    {
      "epoch": 493.86725663716817,
      "grad_norm": 8.375,
      "learning_rate": 1.0069693603816393e-05,
      "loss": 1.133,
      "num_input_tokens_seen": 15602752,
      "step": 28150
    },
    {
      "epoch": 493.95575221238937,
      "grad_norm": 8.5625,
      "learning_rate": 1.0061820321459204e-05,
      "loss": 1.1759,
      "num_input_tokens_seen": 15605232,
      "step": 28155
    },
    {
      "epoch": 494.0353982300885,
      "grad_norm": 8.75,
      "learning_rate": 1.0053949342757038e-05,
      "loss": 1.0741,
      "num_input_tokens_seen": 15607552,
      "step": 28160
    },
    {
      "epoch": 494.12389380530976,
      "grad_norm": 7.625,
      "learning_rate": 1.0046080668923717e-05,
      "loss": 1.1605,
      "num_input_tokens_seen": 15610240,
      "step": 28165
    },
    {
      "epoch": 494.21238938053096,
      "grad_norm": 10.0625,
      "learning_rate": 1.003821430117267e-05,
      "loss": 1.1756,
      "num_input_tokens_seen": 15612976,
      "step": 28170
    },
    {
      "epoch": 494.3008849557522,
      "grad_norm": 10.8125,
      "learning_rate": 1.0030350240716999e-05,
      "loss": 1.0103,
      "num_input_tokens_seen": 15615472,
      "step": 28175
    },
    {
      "epoch": 494.3893805309734,
      "grad_norm": 11.25,
      "learning_rate": 1.0022488488769449e-05,
      "loss": 0.9672,
      "num_input_tokens_seen": 15617872,
      "step": 28180
    },
    {
      "epoch": 494.4778761061947,
      "grad_norm": 7.84375,
      "learning_rate": 1.0014629046542387e-05,
      "loss": 0.8302,
      "num_input_tokens_seen": 15620592,
      "step": 28185
    },
    {
      "epoch": 494.56637168141594,
      "grad_norm": 8.375,
      "learning_rate": 1.0006771915247842e-05,
      "loss": 1.2165,
      "num_input_tokens_seen": 15623200,
      "step": 28190
    },
    {
      "epoch": 494.65486725663715,
      "grad_norm": 6.25,
      "learning_rate": 9.998917096097495e-06,
      "loss": 1.0551,
      "num_input_tokens_seen": 15626464,
      "step": 28195
    },
    {
      "epoch": 494.7433628318584,
      "grad_norm": 8.125,
      "learning_rate": 9.991064590302638e-06,
      "loss": 1.0237,
      "num_input_tokens_seen": 15629392,
      "step": 28200
    },
    {
      "epoch": 494.7433628318584,
      "eval_loss": 1.0310944318771362,
      "eval_runtime": 0.9432,
      "eval_samples_per_second": 26.506,
      "eval_steps_per_second": 13.783,
      "num_input_tokens_seen": 15629392,
      "step": 28200
    },
    {
      "epoch": 494.83185840707966,
      "grad_norm": 7.6875,
      "learning_rate": 9.983214399074241e-06,
      "loss": 1.0817,
      "num_input_tokens_seen": 15632272,
      "step": 28205
    },
    {
      "epoch": 494.92035398230087,
      "grad_norm": 8.0,
      "learning_rate": 9.975366523622893e-06,
      "loss": 1.0557,
      "num_input_tokens_seen": 15634880,
      "step": 28210
    },
    {
      "epoch": 495.0,
      "grad_norm": 15.1875,
      "learning_rate": 9.967520965158841e-06,
      "loss": 0.9927,
      "num_input_tokens_seen": 15637680,
      "step": 28215
    },
    {
      "epoch": 495.08849557522126,
      "grad_norm": 8.75,
      "learning_rate": 9.95967772489197e-06,
      "loss": 1.1127,
      "num_input_tokens_seen": 15640672,
      "step": 28220
    },
    {
      "epoch": 495.17699115044246,
      "grad_norm": 6.21875,
      "learning_rate": 9.951836804031794e-06,
      "loss": 0.8906,
      "num_input_tokens_seen": 15643360,
      "step": 28225
    },
    {
      "epoch": 495.2654867256637,
      "grad_norm": 9.125,
      "learning_rate": 9.943998203787489e-06,
      "loss": 1.076,
      "num_input_tokens_seen": 15646080,
      "step": 28230
    },
    {
      "epoch": 495.353982300885,
      "grad_norm": 10.6875,
      "learning_rate": 9.936161925367874e-06,
      "loss": 1.0811,
      "num_input_tokens_seen": 15648656,
      "step": 28235
    },
    {
      "epoch": 495.4424778761062,
      "grad_norm": 8.125,
      "learning_rate": 9.928327969981386e-06,
      "loss": 0.9808,
      "num_input_tokens_seen": 15651696,
      "step": 28240
    },
    {
      "epoch": 495.53097345132744,
      "grad_norm": 7.28125,
      "learning_rate": 9.920496338836135e-06,
      "loss": 1.0024,
      "num_input_tokens_seen": 15654352,
      "step": 28245
    },
    {
      "epoch": 495.6194690265487,
      "grad_norm": 10.25,
      "learning_rate": 9.912667033139844e-06,
      "loss": 1.1364,
      "num_input_tokens_seen": 15657296,
      "step": 28250
    },
    {
      "epoch": 495.7079646017699,
      "grad_norm": 7.3125,
      "learning_rate": 9.904840054099893e-06,
      "loss": 1.0182,
      "num_input_tokens_seen": 15660096,
      "step": 28255
    },
    {
      "epoch": 495.79646017699116,
      "grad_norm": 8.375,
      "learning_rate": 9.897015402923312e-06,
      "loss": 1.1291,
      "num_input_tokens_seen": 15662912,
      "step": 28260
    },
    {
      "epoch": 495.88495575221236,
      "grad_norm": 12.25,
      "learning_rate": 9.889193080816744e-06,
      "loss": 1.0495,
      "num_input_tokens_seen": 15666128,
      "step": 28265
    },
    {
      "epoch": 495.9734513274336,
      "grad_norm": 8.3125,
      "learning_rate": 9.881373088986498e-06,
      "loss": 0.9924,
      "num_input_tokens_seen": 15668624,
      "step": 28270
    },
    {
      "epoch": 496.05309734513276,
      "grad_norm": 7.53125,
      "learning_rate": 9.873555428638523e-06,
      "loss": 1.2444,
      "num_input_tokens_seen": 15671136,
      "step": 28275
    },
    {
      "epoch": 496.14159292035396,
      "grad_norm": 10.0625,
      "learning_rate": 9.865740100978383e-06,
      "loss": 1.0011,
      "num_input_tokens_seen": 15673664,
      "step": 28280
    },
    {
      "epoch": 496.2300884955752,
      "grad_norm": 6.5,
      "learning_rate": 9.857927107211315e-06,
      "loss": 0.9987,
      "num_input_tokens_seen": 15676688,
      "step": 28285
    },
    {
      "epoch": 496.3185840707965,
      "grad_norm": 8.1875,
      "learning_rate": 9.850116448542177e-06,
      "loss": 1.0423,
      "num_input_tokens_seen": 15679280,
      "step": 28290
    },
    {
      "epoch": 496.4070796460177,
      "grad_norm": 9.3125,
      "learning_rate": 9.842308126175457e-06,
      "loss": 1.1414,
      "num_input_tokens_seen": 15681760,
      "step": 28295
    },
    {
      "epoch": 496.49557522123894,
      "grad_norm": 6.90625,
      "learning_rate": 9.834502141315315e-06,
      "loss": 1.1541,
      "num_input_tokens_seen": 15684128,
      "step": 28300
    },
    {
      "epoch": 496.5840707964602,
      "grad_norm": 9.0625,
      "learning_rate": 9.82669849516552e-06,
      "loss": 1.205,
      "num_input_tokens_seen": 15687152,
      "step": 28305
    },
    {
      "epoch": 496.6725663716814,
      "grad_norm": 10.125,
      "learning_rate": 9.818897188929493e-06,
      "loss": 1.1755,
      "num_input_tokens_seen": 15689840,
      "step": 28310
    },
    {
      "epoch": 496.76106194690266,
      "grad_norm": 10.125,
      "learning_rate": 9.811098223810309e-06,
      "loss": 0.983,
      "num_input_tokens_seen": 15692928,
      "step": 28315
    },
    {
      "epoch": 496.8495575221239,
      "grad_norm": 8.875,
      "learning_rate": 9.803301601010641e-06,
      "loss": 0.9407,
      "num_input_tokens_seen": 15696112,
      "step": 28320
    },
    {
      "epoch": 496.9380530973451,
      "grad_norm": 5.71875,
      "learning_rate": 9.795507321732853e-06,
      "loss": 0.9827,
      "num_input_tokens_seen": 15699120,
      "step": 28325
    },
    {
      "epoch": 497.01769911504425,
      "grad_norm": 8.375,
      "learning_rate": 9.787715387178898e-06,
      "loss": 0.9274,
      "num_input_tokens_seen": 15701720,
      "step": 28330
    },
    {
      "epoch": 497.1061946902655,
      "grad_norm": 7.0,
      "learning_rate": 9.779925798550399e-06,
      "loss": 0.9838,
      "num_input_tokens_seen": 15704616,
      "step": 28335
    },
    {
      "epoch": 497.1946902654867,
      "grad_norm": 8.25,
      "learning_rate": 9.772138557048619e-06,
      "loss": 1.0156,
      "num_input_tokens_seen": 15707000,
      "step": 28340
    },
    {
      "epoch": 497.283185840708,
      "grad_norm": 10.0,
      "learning_rate": 9.764353663874426e-06,
      "loss": 1.1363,
      "num_input_tokens_seen": 15709992,
      "step": 28345
    },
    {
      "epoch": 497.37168141592923,
      "grad_norm": 10.3125,
      "learning_rate": 9.756571120228375e-06,
      "loss": 1.4163,
      "num_input_tokens_seen": 15712664,
      "step": 28350
    },
    {
      "epoch": 497.46017699115043,
      "grad_norm": 7.71875,
      "learning_rate": 9.748790927310605e-06,
      "loss": 0.9584,
      "num_input_tokens_seen": 15715640,
      "step": 28355
    },
    {
      "epoch": 497.5486725663717,
      "grad_norm": 8.8125,
      "learning_rate": 9.741013086320946e-06,
      "loss": 0.8334,
      "num_input_tokens_seen": 15719176,
      "step": 28360
    },
    {
      "epoch": 497.6371681415929,
      "grad_norm": 7.71875,
      "learning_rate": 9.733237598458821e-06,
      "loss": 0.9214,
      "num_input_tokens_seen": 15721704,
      "step": 28365
    },
    {
      "epoch": 497.72566371681415,
      "grad_norm": 12.25,
      "learning_rate": 9.725464464923308e-06,
      "loss": 1.0949,
      "num_input_tokens_seen": 15724424,
      "step": 28370
    },
    {
      "epoch": 497.8141592920354,
      "grad_norm": 6.59375,
      "learning_rate": 9.717693686913123e-06,
      "loss": 1.0448,
      "num_input_tokens_seen": 15727144,
      "step": 28375
    },
    {
      "epoch": 497.9026548672566,
      "grad_norm": 8.25,
      "learning_rate": 9.709925265626632e-06,
      "loss": 1.1052,
      "num_input_tokens_seen": 15729832,
      "step": 28380
    },
    {
      "epoch": 497.9911504424779,
      "grad_norm": 8.9375,
      "learning_rate": 9.702159202261801e-06,
      "loss": 1.0931,
      "num_input_tokens_seen": 15732520,
      "step": 28385
    },
    {
      "epoch": 498.070796460177,
      "grad_norm": 6.84375,
      "learning_rate": 9.694395498016268e-06,
      "loss": 1.0369,
      "num_input_tokens_seen": 15734760,
      "step": 28390
    },
    {
      "epoch": 498.1592920353982,
      "grad_norm": 8.6875,
      "learning_rate": 9.686634154087298e-06,
      "loss": 1.1073,
      "num_input_tokens_seen": 15737368,
      "step": 28395
    },
    {
      "epoch": 498.24778761061947,
      "grad_norm": 8.5625,
      "learning_rate": 9.678875171671776e-06,
      "loss": 0.997,
      "num_input_tokens_seen": 15740552,
      "step": 28400
    },
    {
      "epoch": 498.24778761061947,
      "eval_loss": 1.0310944318771362,
      "eval_runtime": 0.9429,
      "eval_samples_per_second": 26.514,
      "eval_steps_per_second": 13.787,
      "num_input_tokens_seen": 15740552,
      "step": 28400
    },
    {
      "epoch": 498.3362831858407,
      "grad_norm": 7.84375,
      "learning_rate": 9.671118551966246e-06,
      "loss": 0.8468,
      "num_input_tokens_seen": 15743704,
      "step": 28405
    },
    {
      "epoch": 498.42477876106193,
      "grad_norm": 8.875,
      "learning_rate": 9.66336429616686e-06,
      "loss": 1.3423,
      "num_input_tokens_seen": 15746584,
      "step": 28410
    },
    {
      "epoch": 498.5132743362832,
      "grad_norm": 7.90625,
      "learning_rate": 9.655612405469436e-06,
      "loss": 0.9505,
      "num_input_tokens_seen": 15749384,
      "step": 28415
    },
    {
      "epoch": 498.60176991150445,
      "grad_norm": 7.625,
      "learning_rate": 9.647862881069413e-06,
      "loss": 1.101,
      "num_input_tokens_seen": 15752328,
      "step": 28420
    },
    {
      "epoch": 498.69026548672565,
      "grad_norm": 8.0,
      "learning_rate": 9.640115724161855e-06,
      "loss": 1.0805,
      "num_input_tokens_seen": 15754920,
      "step": 28425
    },
    {
      "epoch": 498.7787610619469,
      "grad_norm": 11.375,
      "learning_rate": 9.632370935941483e-06,
      "loss": 1.0324,
      "num_input_tokens_seen": 15757752,
      "step": 28430
    },
    {
      "epoch": 498.86725663716817,
      "grad_norm": 7.9375,
      "learning_rate": 9.624628517602634e-06,
      "loss": 1.0391,
      "num_input_tokens_seen": 15760600,
      "step": 28435
    },
    {
      "epoch": 498.95575221238937,
      "grad_norm": 12.1875,
      "learning_rate": 9.61688847033928e-06,
      "loss": 1.037,
      "num_input_tokens_seen": 15763480,
      "step": 28440
    },
    {
      "epoch": 499.0353982300885,
      "grad_norm": 7.03125,
      "learning_rate": 9.609150795345051e-06,
      "loss": 1.0751,
      "num_input_tokens_seen": 15765408,
      "step": 28445
    },
    {
      "epoch": 499.12389380530976,
      "grad_norm": 8.3125,
      "learning_rate": 9.601415493813171e-06,
      "loss": 1.2127,
      "num_input_tokens_seen": 15767840,
      "step": 28450
    },
    {
      "epoch": 499.21238938053096,
      "grad_norm": 8.1875,
      "learning_rate": 9.593682566936533e-06,
      "loss": 0.895,
      "num_input_tokens_seen": 15770704,
      "step": 28455
    },
    {
      "epoch": 499.3008849557522,
      "grad_norm": 8.1875,
      "learning_rate": 9.58595201590766e-06,
      "loss": 0.9974,
      "num_input_tokens_seen": 15773424,
      "step": 28460
    },
    {
      "epoch": 499.3893805309734,
      "grad_norm": 10.1875,
      "learning_rate": 9.578223841918681e-06,
      "loss": 0.991,
      "num_input_tokens_seen": 15775936,
      "step": 28465
    },
    {
      "epoch": 499.4778761061947,
      "grad_norm": 7.90625,
      "learning_rate": 9.570498046161389e-06,
      "loss": 1.1705,
      "num_input_tokens_seen": 15778736,
      "step": 28470
    },
    {
      "epoch": 499.56637168141594,
      "grad_norm": 9.5,
      "learning_rate": 9.562774629827206e-06,
      "loss": 1.1453,
      "num_input_tokens_seen": 15782160,
      "step": 28475
    },
    {
      "epoch": 499.65486725663715,
      "grad_norm": 7.65625,
      "learning_rate": 9.555053594107163e-06,
      "loss": 0.9709,
      "num_input_tokens_seen": 15784992,
      "step": 28480
    },
    {
      "epoch": 499.7433628318584,
      "grad_norm": 10.375,
      "learning_rate": 9.547334940191957e-06,
      "loss": 1.0496,
      "num_input_tokens_seen": 15788096,
      "step": 28485
    },
    {
      "epoch": 499.83185840707966,
      "grad_norm": 7.65625,
      "learning_rate": 9.539618669271886e-06,
      "loss": 0.9269,
      "num_input_tokens_seen": 15791056,
      "step": 28490
    },
    {
      "epoch": 499.92035398230087,
      "grad_norm": 8.3125,
      "learning_rate": 9.531904782536904e-06,
      "loss": 0.9901,
      "num_input_tokens_seen": 15794208,
      "step": 28495
    },
    {
      "epoch": 500.0,
      "grad_norm": 7.34375,
      "learning_rate": 9.524193281176597e-06,
      "loss": 1.1447,
      "num_input_tokens_seen": 15796304,
      "step": 28500
    },
    {
      "epoch": 500.08849557522126,
      "grad_norm": 8.0,
      "learning_rate": 9.516484166380165e-06,
      "loss": 0.8731,
      "num_input_tokens_seen": 15799488,
      "step": 28505
    },
    {
      "epoch": 500.17699115044246,
      "grad_norm": 7.5,
      "learning_rate": 9.508777439336447e-06,
      "loss": 0.8759,
      "num_input_tokens_seen": 15802080,
      "step": 28510
    },
    {
      "epoch": 500.2654867256637,
      "grad_norm": 7.21875,
      "learning_rate": 9.50107310123393e-06,
      "loss": 0.9572,
      "num_input_tokens_seen": 15805088,
      "step": 28515
    },
    {
      "epoch": 500.353982300885,
      "grad_norm": 9.8125,
      "learning_rate": 9.493371153260702e-06,
      "loss": 1.1948,
      "num_input_tokens_seen": 15807696,
      "step": 28520
    },
    {
      "epoch": 500.4424778761062,
      "grad_norm": 8.125,
      "learning_rate": 9.485671596604523e-06,
      "loss": 1.1036,
      "num_input_tokens_seen": 15810576,
      "step": 28525
    },
    {
      "epoch": 500.53097345132744,
      "grad_norm": 8.6875,
      "learning_rate": 9.477974432452738e-06,
      "loss": 1.2366,
      "num_input_tokens_seen": 15813504,
      "step": 28530
    },
    {
      "epoch": 500.6194690265487,
      "grad_norm": 8.125,
      "learning_rate": 9.470279661992356e-06,
      "loss": 1.0016,
      "num_input_tokens_seen": 15816000,
      "step": 28535
    },
    {
      "epoch": 500.7079646017699,
      "grad_norm": 13.625,
      "learning_rate": 9.462587286410021e-06,
      "loss": 1.0732,
      "num_input_tokens_seen": 15818992,
      "step": 28540
    },
    {
      "epoch": 500.79646017699116,
      "grad_norm": 7.75,
      "learning_rate": 9.454897306891972e-06,
      "loss": 1.0609,
      "num_input_tokens_seen": 15821744,
      "step": 28545
    },
    {
      "epoch": 500.88495575221236,
      "grad_norm": 8.4375,
      "learning_rate": 9.44720972462411e-06,
      "loss": 1.0882,
      "num_input_tokens_seen": 15824752,
      "step": 28550
    },
    {
      "epoch": 500.9734513274336,
      "grad_norm": 8.1875,
      "learning_rate": 9.439524540791964e-06,
      "loss": 1.0847,
      "num_input_tokens_seen": 15827232,
      "step": 28555
    },
    {
      "epoch": 501.05309734513276,
      "grad_norm": 7.75,
      "learning_rate": 9.431841756580673e-06,
      "loss": 0.8331,
      "num_input_tokens_seen": 15829664,
      "step": 28560
    },
    {
      "epoch": 501.14159292035396,
      "grad_norm": 8.6875,
      "learning_rate": 9.42416137317503e-06,
      "loss": 1.0372,
      "num_input_tokens_seen": 15832656,
      "step": 28565
    },
    {
      "epoch": 501.2300884955752,
      "grad_norm": 7.8125,
      "learning_rate": 9.416483391759437e-06,
      "loss": 1.0373,
      "num_input_tokens_seen": 15835440,
      "step": 28570
    },
    {
      "epoch": 501.3185840707965,
      "grad_norm": 10.25,
      "learning_rate": 9.408807813517945e-06,
      "loss": 1.1745,
      "num_input_tokens_seen": 15838144,
      "step": 28575
    },
    {
      "epoch": 501.4070796460177,
      "grad_norm": 8.875,
      "learning_rate": 9.401134639634221e-06,
      "loss": 0.7657,
      "num_input_tokens_seen": 15840848,
      "step": 28580
    },
    {
      "epoch": 501.49557522123894,
      "grad_norm": 8.3125,
      "learning_rate": 9.393463871291555e-06,
      "loss": 1.0947,
      "num_input_tokens_seen": 15843664,
      "step": 28585
    },
    {
      "epoch": 501.5840707964602,
      "grad_norm": 8.875,
      "learning_rate": 9.385795509672881e-06,
      "loss": 1.2797,
      "num_input_tokens_seen": 15846480,
      "step": 28590
    },
    {
      "epoch": 501.6725663716814,
      "grad_norm": 9.1875,
      "learning_rate": 9.378129555960771e-06,
      "loss": 1.1084,
      "num_input_tokens_seen": 15849344,
      "step": 28595
    },
    {
      "epoch": 501.76106194690266,
      "grad_norm": 8.5,
      "learning_rate": 9.370466011337392e-06,
      "loss": 1.062,
      "num_input_tokens_seen": 15852112,
      "step": 28600
    },
    {
      "epoch": 501.76106194690266,
      "eval_loss": 1.0310944318771362,
      "eval_runtime": 0.9352,
      "eval_samples_per_second": 26.731,
      "eval_steps_per_second": 13.9,
      "num_input_tokens_seen": 15852112,
      "step": 28600
    },
    {
      "epoch": 501.8495575221239,
      "grad_norm": 8.8125,
      "learning_rate": 9.362804876984573e-06,
      "loss": 1.0001,
      "num_input_tokens_seen": 15855248,
      "step": 28605
    },
    {
      "epoch": 501.9380530973451,
      "grad_norm": 9.75,
      "learning_rate": 9.355146154083747e-06,
      "loss": 0.9865,
      "num_input_tokens_seen": 15857776,
      "step": 28610
    },
    {
      "epoch": 502.01769911504425,
      "grad_norm": 8.375,
      "learning_rate": 9.347489843815987e-06,
      "loss": 1.0935,
      "num_input_tokens_seen": 15860096,
      "step": 28615
    },
    {
      "epoch": 502.1061946902655,
      "grad_norm": 8.125,
      "learning_rate": 9.339835947362002e-06,
      "loss": 1.2378,
      "num_input_tokens_seen": 15862624,
      "step": 28620
    },
    {
      "epoch": 502.1946902654867,
      "grad_norm": 6.8125,
      "learning_rate": 9.332184465902105e-06,
      "loss": 1.1655,
      "num_input_tokens_seen": 15865376,
      "step": 28625
    },
    {
      "epoch": 502.283185840708,
      "grad_norm": 6.09375,
      "learning_rate": 9.324535400616266e-06,
      "loss": 0.9208,
      "num_input_tokens_seen": 15868304,
      "step": 28630
    },
    {
      "epoch": 502.37168141592923,
      "grad_norm": 10.5,
      "learning_rate": 9.31688875268405e-06,
      "loss": 1.0376,
      "num_input_tokens_seen": 15871168,
      "step": 28635
    },
    {
      "epoch": 502.46017699115043,
      "grad_norm": 8.3125,
      "learning_rate": 9.309244523284674e-06,
      "loss": 1.0416,
      "num_input_tokens_seen": 15873824,
      "step": 28640
    },
    {
      "epoch": 502.5486725663717,
      "grad_norm": 8.6875,
      "learning_rate": 9.301602713596982e-06,
      "loss": 0.9269,
      "num_input_tokens_seen": 15876512,
      "step": 28645
    },
    {
      "epoch": 502.6371681415929,
      "grad_norm": 6.9375,
      "learning_rate": 9.293963324799432e-06,
      "loss": 1.0807,
      "num_input_tokens_seen": 15879232,
      "step": 28650
    },
    {
      "epoch": 502.72566371681415,
      "grad_norm": 6.8125,
      "learning_rate": 9.286326358070104e-06,
      "loss": 1.0556,
      "num_input_tokens_seen": 15882096,
      "step": 28655
    },
    {
      "epoch": 502.8141592920354,
      "grad_norm": 13.625,
      "learning_rate": 9.278691814586729e-06,
      "loss": 1.037,
      "num_input_tokens_seen": 15885312,
      "step": 28660
    },
    {
      "epoch": 502.9026548672566,
      "grad_norm": 12.5,
      "learning_rate": 9.271059695526635e-06,
      "loss": 1.2185,
      "num_input_tokens_seen": 15888032,
      "step": 28665
    },
    {
      "epoch": 502.9911504424779,
      "grad_norm": 7.84375,
      "learning_rate": 9.263430002066805e-06,
      "loss": 0.9624,
      "num_input_tokens_seen": 15890784,
      "step": 28670
    },
    {
      "epoch": 503.070796460177,
      "grad_norm": 7.96875,
      "learning_rate": 9.25580273538382e-06,
      "loss": 0.9227,
      "num_input_tokens_seen": 15893200,
      "step": 28675
    },
    {
      "epoch": 503.1592920353982,
      "grad_norm": 7.84375,
      "learning_rate": 9.248177896653907e-06,
      "loss": 1.0628,
      "num_input_tokens_seen": 15895680,
      "step": 28680
    },
    {
      "epoch": 503.24778761061947,
      "grad_norm": 12.625,
      "learning_rate": 9.240555487052918e-06,
      "loss": 0.9888,
      "num_input_tokens_seen": 15898896,
      "step": 28685
    },
    {
      "epoch": 503.3362831858407,
      "grad_norm": 10.25,
      "learning_rate": 9.232935507756313e-06,
      "loss": 1.0881,
      "num_input_tokens_seen": 15901168,
      "step": 28690
    },
    {
      "epoch": 503.42477876106193,
      "grad_norm": 7.28125,
      "learning_rate": 9.225317959939193e-06,
      "loss": 0.9237,
      "num_input_tokens_seen": 15903968,
      "step": 28695
    },
    {
      "epoch": 503.5132743362832,
      "grad_norm": 9.9375,
      "learning_rate": 9.217702844776287e-06,
      "loss": 1.0608,
      "num_input_tokens_seen": 15906768,
      "step": 28700
    },
    {
      "epoch": 503.60176991150445,
      "grad_norm": 7.0625,
      "learning_rate": 9.210090163441929e-06,
      "loss": 1.0233,
      "num_input_tokens_seen": 15909392,
      "step": 28705
    },
    {
      "epoch": 503.69026548672565,
      "grad_norm": 8.25,
      "learning_rate": 9.202479917110105e-06,
      "loss": 1.063,
      "num_input_tokens_seen": 15912480,
      "step": 28710
    },
    {
      "epoch": 503.7787610619469,
      "grad_norm": 5.46875,
      "learning_rate": 9.194872106954392e-06,
      "loss": 1.1646,
      "num_input_tokens_seen": 15915104,
      "step": 28715
    },
    {
      "epoch": 503.86725663716817,
      "grad_norm": 14.1875,
      "learning_rate": 9.187266734148029e-06,
      "loss": 0.9238,
      "num_input_tokens_seen": 15918192,
      "step": 28720
    },
    {
      "epoch": 503.95575221238937,
      "grad_norm": 7.59375,
      "learning_rate": 9.179663799863849e-06,
      "loss": 1.118,
      "num_input_tokens_seen": 15921312,
      "step": 28725
    },
    {
      "epoch": 504.0353982300885,
      "grad_norm": 8.5625,
      "learning_rate": 9.172063305274317e-06,
      "loss": 1.0802,
      "num_input_tokens_seen": 15924000,
      "step": 28730
    },
    {
      "epoch": 504.12389380530976,
      "grad_norm": 7.59375,
      "learning_rate": 9.164465251551527e-06,
      "loss": 0.9974,
      "num_input_tokens_seen": 15926464,
      "step": 28735
    },
    {
      "epoch": 504.21238938053096,
      "grad_norm": 8.625,
      "learning_rate": 9.156869639867205e-06,
      "loss": 0.7931,
      "num_input_tokens_seen": 15929040,
      "step": 28740
    },
    {
      "epoch": 504.3008849557522,
      "grad_norm": 7.125,
      "learning_rate": 9.149276471392677e-06,
      "loss": 1.0347,
      "num_input_tokens_seen": 15931712,
      "step": 28745
    },
    {
      "epoch": 504.3893805309734,
      "grad_norm": 6.9375,
      "learning_rate": 9.141685747298914e-06,
      "loss": 1.0756,
      "num_input_tokens_seen": 15934784,
      "step": 28750
    },
    {
      "epoch": 504.4778761061947,
      "grad_norm": 10.125,
      "learning_rate": 9.13409746875649e-06,
      "loss": 1.1567,
      "num_input_tokens_seen": 15937232,
      "step": 28755
    },
    {
      "epoch": 504.56637168141594,
      "grad_norm": 6.65625,
      "learning_rate": 9.12651163693562e-06,
      "loss": 0.9069,
      "num_input_tokens_seen": 15940192,
      "step": 28760
    },
    {
      "epoch": 504.65486725663715,
      "grad_norm": 9.5625,
      "learning_rate": 9.11892825300614e-06,
      "loss": 1.025,
      "num_input_tokens_seen": 15943216,
      "step": 28765
    },
    {
      "epoch": 504.7433628318584,
      "grad_norm": 11.75,
      "learning_rate": 9.111347318137491e-06,
      "loss": 1.1962,
      "num_input_tokens_seen": 15946320,
      "step": 28770
    },
    {
      "epoch": 504.83185840707966,
      "grad_norm": 7.40625,
      "learning_rate": 9.103768833498755e-06,
      "loss": 1.0712,
      "num_input_tokens_seen": 15949280,
      "step": 28775
    },
    {
      "epoch": 504.92035398230087,
      "grad_norm": 6.84375,
      "learning_rate": 9.096192800258639e-06,
      "loss": 1.219,
      "num_input_tokens_seen": 15952128,
      "step": 28780
    },
    {
      "epoch": 505.0,
      "grad_norm": 17.375,
      "learning_rate": 9.088619219585443e-06,
      "loss": 1.126,
      "num_input_tokens_seen": 15954680,
      "step": 28785
    },
    {
      "epoch": 505.08849557522126,
      "grad_norm": 8.1875,
      "learning_rate": 9.081048092647127e-06,
      "loss": 0.8565,
      "num_input_tokens_seen": 15957208,
      "step": 28790
    },
    {
      "epoch": 505.17699115044246,
      "grad_norm": 7.59375,
      "learning_rate": 9.073479420611245e-06,
      "loss": 0.8975,
      "num_input_tokens_seen": 15960136,
      "step": 28795
    },
    {
      "epoch": 505.2654867256637,
      "grad_norm": 11.375,
      "learning_rate": 9.065913204644974e-06,
      "loss": 1.2497,
      "num_input_tokens_seen": 15962600,
      "step": 28800
    },
    {
      "epoch": 505.2654867256637,
      "eval_loss": 1.0310944318771362,
      "eval_runtime": 0.9359,
      "eval_samples_per_second": 26.711,
      "eval_steps_per_second": 13.89,
      "num_input_tokens_seen": 15962600,
      "step": 28800
    },
    {
      "epoch": 505.353982300885,
      "grad_norm": 8.4375,
      "learning_rate": 9.058349445915135e-06,
      "loss": 1.0974,
      "num_input_tokens_seen": 15965176,
      "step": 28805
    },
    {
      "epoch": 505.4424778761062,
      "grad_norm": 7.25,
      "learning_rate": 9.050788145588138e-06,
      "loss": 0.8949,
      "num_input_tokens_seen": 15968072,
      "step": 28810
    },
    {
      "epoch": 505.53097345132744,
      "grad_norm": 9.0,
      "learning_rate": 9.043229304830039e-06,
      "loss": 1.1658,
      "num_input_tokens_seen": 15971048,
      "step": 28815
    },
    {
      "epoch": 505.6194690265487,
      "grad_norm": 4.5,
      "learning_rate": 9.035672924806515e-06,
      "loss": 0.951,
      "num_input_tokens_seen": 15973896,
      "step": 28820
    },
    {
      "epoch": 505.7079646017699,
      "grad_norm": 6.8125,
      "learning_rate": 9.028119006682839e-06,
      "loss": 1.062,
      "num_input_tokens_seen": 15976616,
      "step": 28825
    },
    {
      "epoch": 505.79646017699116,
      "grad_norm": 8.875,
      "learning_rate": 9.020567551623935e-06,
      "loss": 1.068,
      "num_input_tokens_seen": 15979576,
      "step": 28830
    },
    {
      "epoch": 505.88495575221236,
      "grad_norm": 9.5,
      "learning_rate": 9.013018560794318e-06,
      "loss": 1.0872,
      "num_input_tokens_seen": 15982200,
      "step": 28835
    },
    {
      "epoch": 505.9734513274336,
      "grad_norm": 12.1875,
      "learning_rate": 9.005472035358139e-06,
      "loss": 1.1534,
      "num_input_tokens_seen": 15985160,
      "step": 28840
    },
    {
      "epoch": 506.05309734513276,
      "grad_norm": 8.4375,
      "learning_rate": 8.997927976479185e-06,
      "loss": 1.1603,
      "num_input_tokens_seen": 15987960,
      "step": 28845
    },
    {
      "epoch": 506.14159292035396,
      "grad_norm": 9.5625,
      "learning_rate": 8.99038638532082e-06,
      "loss": 1.0019,
      "num_input_tokens_seen": 15991000,
      "step": 28850
    },
    {
      "epoch": 506.2300884955752,
      "grad_norm": 7.25,
      "learning_rate": 8.982847263046065e-06,
      "loss": 1.259,
      "num_input_tokens_seen": 15993464,
      "step": 28855
    },
    {
      "epoch": 506.3185840707965,
      "grad_norm": 8.125,
      "learning_rate": 8.975310610817555e-06,
      "loss": 1.0605,
      "num_input_tokens_seen": 15996584,
      "step": 28860
    },
    {
      "epoch": 506.4070796460177,
      "grad_norm": 7.375,
      "learning_rate": 8.967776429797528e-06,
      "loss": 0.987,
      "num_input_tokens_seen": 15999592,
      "step": 28865
    },
    {
      "epoch": 506.49557522123894,
      "grad_norm": 6.5,
      "learning_rate": 8.960244721147842e-06,
      "loss": 0.9616,
      "num_input_tokens_seen": 16002168,
      "step": 28870
    },
    {
      "epoch": 506.5840707964602,
      "grad_norm": 8.75,
      "learning_rate": 8.952715486029995e-06,
      "loss": 1.0126,
      "num_input_tokens_seen": 16005240,
      "step": 28875
    },
    {
      "epoch": 506.6725663716814,
      "grad_norm": 9.5,
      "learning_rate": 8.945188725605075e-06,
      "loss": 1.2822,
      "num_input_tokens_seen": 16008232,
      "step": 28880
    },
    {
      "epoch": 506.76106194690266,
      "grad_norm": 8.9375,
      "learning_rate": 8.937664441033817e-06,
      "loss": 1.0512,
      "num_input_tokens_seen": 16010520,
      "step": 28885
    },
    {
      "epoch": 506.8495575221239,
      "grad_norm": 7.1875,
      "learning_rate": 8.930142633476549e-06,
      "loss": 1.0407,
      "num_input_tokens_seen": 16013384,
      "step": 28890
    },
    {
      "epoch": 506.9380530973451,
      "grad_norm": 7.875,
      "learning_rate": 8.92262330409323e-06,
      "loss": 0.882,
      "num_input_tokens_seen": 16016264,
      "step": 28895
    },
    {
      "epoch": 507.01769911504425,
      "grad_norm": 7.71875,
      "learning_rate": 8.915106454043448e-06,
      "loss": 0.9363,
      "num_input_tokens_seen": 16018688,
      "step": 28900
    },
    {
      "epoch": 507.1061946902655,
      "grad_norm": 7.5,
      "learning_rate": 8.90759208448638e-06,
      "loss": 0.9614,
      "num_input_tokens_seen": 16021712,
      "step": 28905
    },
    {
      "epoch": 507.1946902654867,
      "grad_norm": 9.1875,
      "learning_rate": 8.900080196580848e-06,
      "loss": 1.1298,
      "num_input_tokens_seen": 16024032,
      "step": 28910
    },
    {
      "epoch": 507.283185840708,
      "grad_norm": 9.5625,
      "learning_rate": 8.892570791485267e-06,
      "loss": 1.0753,
      "num_input_tokens_seen": 16026960,
      "step": 28915
    },
    {
      "epoch": 507.37168141592923,
      "grad_norm": 7.3125,
      "learning_rate": 8.885063870357688e-06,
      "loss": 0.9458,
      "num_input_tokens_seen": 16029248,
      "step": 28920
    },
    {
      "epoch": 507.46017699115043,
      "grad_norm": 8.5,
      "learning_rate": 8.87755943435578e-06,
      "loss": 1.0551,
      "num_input_tokens_seen": 16032288,
      "step": 28925
    },
    {
      "epoch": 507.5486725663717,
      "grad_norm": 7.4375,
      "learning_rate": 8.87005748463681e-06,
      "loss": 1.0257,
      "num_input_tokens_seen": 16034832,
      "step": 28930
    },
    {
      "epoch": 507.6371681415929,
      "grad_norm": 7.625,
      "learning_rate": 8.862558022357681e-06,
      "loss": 1.0131,
      "num_input_tokens_seen": 16037696,
      "step": 28935
    },
    {
      "epoch": 507.72566371681415,
      "grad_norm": 8.875,
      "learning_rate": 8.855061048674903e-06,
      "loss": 1.0779,
      "num_input_tokens_seen": 16040256,
      "step": 28940
    },
    {
      "epoch": 507.8141592920354,
      "grad_norm": 8.0625,
      "learning_rate": 8.847566564744595e-06,
      "loss": 1.141,
      "num_input_tokens_seen": 16042800,
      "step": 28945
    },
    {
      "epoch": 507.9026548672566,
      "grad_norm": 10.9375,
      "learning_rate": 8.840074571722512e-06,
      "loss": 1.0185,
      "num_input_tokens_seen": 16045488,
      "step": 28950
    },
    {
      "epoch": 507.9911504424779,
      "grad_norm": 10.875,
      "learning_rate": 8.832585070764002e-06,
      "loss": 1.1208,
      "num_input_tokens_seen": 16048912,
      "step": 28955
    },
    {
      "epoch": 508.070796460177,
      "grad_norm": 8.125,
      "learning_rate": 8.825098063024045e-06,
      "loss": 1.2289,
      "num_input_tokens_seen": 16051288,
      "step": 28960
    },
    {
      "epoch": 508.1592920353982,
      "grad_norm": 6.96875,
      "learning_rate": 8.817613549657244e-06,
      "loss": 1.066,
      "num_input_tokens_seen": 16053816,
      "step": 28965
    },
    {
      "epoch": 508.24778761061947,
      "grad_norm": 8.4375,
      "learning_rate": 8.810131531817783e-06,
      "loss": 1.1112,
      "num_input_tokens_seen": 16056808,
      "step": 28970
    },
    {
      "epoch": 508.3362831858407,
      "grad_norm": 9.875,
      "learning_rate": 8.802652010659496e-06,
      "loss": 1.0533,
      "num_input_tokens_seen": 16059672,
      "step": 28975
    },
    {
      "epoch": 508.42477876106193,
      "grad_norm": 9.125,
      "learning_rate": 8.795174987335827e-06,
      "loss": 0.9846,
      "num_input_tokens_seen": 16062824,
      "step": 28980
    },
    {
      "epoch": 508.5132743362832,
      "grad_norm": 7.65625,
      "learning_rate": 8.787700462999807e-06,
      "loss": 0.9594,
      "num_input_tokens_seen": 16065864,
      "step": 28985
    },
    {
      "epoch": 508.60176991150445,
      "grad_norm": 9.625,
      "learning_rate": 8.780228438804122e-06,
      "loss": 1.1715,
      "num_input_tokens_seen": 16068696,
      "step": 28990
    },
    {
      "epoch": 508.69026548672565,
      "grad_norm": 7.375,
      "learning_rate": 8.772758915901032e-06,
      "loss": 1.1859,
      "num_input_tokens_seen": 16071112,
      "step": 28995
    },
    {
      "epoch": 508.7787610619469,
      "grad_norm": 7.09375,
      "learning_rate": 8.765291895442443e-06,
      "loss": 0.85,
      "num_input_tokens_seen": 16073896,
      "step": 29000
    },
    {
      "epoch": 508.7787610619469,
      "eval_loss": 1.0310944318771362,
      "eval_runtime": 0.9434,
      "eval_samples_per_second": 26.501,
      "eval_steps_per_second": 13.781,
      "num_input_tokens_seen": 16073896,
      "step": 29000
    },
    {
      "epoch": 508.86725663716817,
      "grad_norm": 8.6875,
      "learning_rate": 8.75782737857987e-06,
      "loss": 0.9936,
      "num_input_tokens_seen": 16076936,
      "step": 29005
    },
    {
      "epoch": 508.95575221238937,
      "grad_norm": 7.625,
      "learning_rate": 8.750365366464425e-06,
      "loss": 0.976,
      "num_input_tokens_seen": 16079560,
      "step": 29010
    },
    {
      "epoch": 509.0353982300885,
      "grad_norm": 8.4375,
      "learning_rate": 8.742905860246838e-06,
      "loss": 1.057,
      "num_input_tokens_seen": 16081696,
      "step": 29015
    },
    {
      "epoch": 509.12389380530976,
      "grad_norm": 9.375,
      "learning_rate": 8.735448861077478e-06,
      "loss": 1.1376,
      "num_input_tokens_seen": 16084272,
      "step": 29020
    },
    {
      "epoch": 509.21238938053096,
      "grad_norm": 6.03125,
      "learning_rate": 8.727994370106288e-06,
      "loss": 1.0009,
      "num_input_tokens_seen": 16086880,
      "step": 29025
    },
    {
      "epoch": 509.3008849557522,
      "grad_norm": 8.3125,
      "learning_rate": 8.720542388482861e-06,
      "loss": 1.0832,
      "num_input_tokens_seen": 16089616,
      "step": 29030
    },
    {
      "epoch": 509.3893805309734,
      "grad_norm": 9.75,
      "learning_rate": 8.71309291735637e-06,
      "loss": 1.0909,
      "num_input_tokens_seen": 16092784,
      "step": 29035
    },
    {
      "epoch": 509.4778761061947,
      "grad_norm": 10.875,
      "learning_rate": 8.705645957875621e-06,
      "loss": 1.0966,
      "num_input_tokens_seen": 16095616,
      "step": 29040
    },
    {
      "epoch": 509.56637168141594,
      "grad_norm": 10.125,
      "learning_rate": 8.698201511189048e-06,
      "loss": 1.0409,
      "num_input_tokens_seen": 16098688,
      "step": 29045
    },
    {
      "epoch": 509.65486725663715,
      "grad_norm": 7.375,
      "learning_rate": 8.690759578444649e-06,
      "loss": 0.97,
      "num_input_tokens_seen": 16101472,
      "step": 29050
    },
    {
      "epoch": 509.7433628318584,
      "grad_norm": 8.8125,
      "learning_rate": 8.68332016079008e-06,
      "loss": 0.9343,
      "num_input_tokens_seen": 16104624,
      "step": 29055
    },
    {
      "epoch": 509.83185840707966,
      "grad_norm": 8.5625,
      "learning_rate": 8.6758832593726e-06,
      "loss": 1.0575,
      "num_input_tokens_seen": 16107232,
      "step": 29060
    },
    {
      "epoch": 509.92035398230087,
      "grad_norm": 10.5625,
      "learning_rate": 8.668448875339053e-06,
      "loss": 1.1512,
      "num_input_tokens_seen": 16109680,
      "step": 29065
    },
    {
      "epoch": 510.0,
      "grad_norm": 15.8125,
      "learning_rate": 8.661017009835933e-06,
      "loss": 0.9796,
      "num_input_tokens_seen": 16112056,
      "step": 29070
    },
    {
      "epoch": 510.08849557522126,
      "grad_norm": 8.25,
      "learning_rate": 8.653587664009311e-06,
      "loss": 1.0874,
      "num_input_tokens_seen": 16115016,
      "step": 29075
    },
    {
      "epoch": 510.17699115044246,
      "grad_norm": 13.3125,
      "learning_rate": 8.646160839004902e-06,
      "loss": 1.0176,
      "num_input_tokens_seen": 16117880,
      "step": 29080
    },
    {
      "epoch": 510.2654867256637,
      "grad_norm": 11.6875,
      "learning_rate": 8.638736535967998e-06,
      "loss": 1.0296,
      "num_input_tokens_seen": 16120632,
      "step": 29085
    },
    {
      "epoch": 510.353982300885,
      "grad_norm": 7.90625,
      "learning_rate": 8.631314756043535e-06,
      "loss": 1.1027,
      "num_input_tokens_seen": 16123592,
      "step": 29090
    },
    {
      "epoch": 510.4424778761062,
      "grad_norm": 9.8125,
      "learning_rate": 8.62389550037603e-06,
      "loss": 1.1651,
      "num_input_tokens_seen": 16126424,
      "step": 29095
    },
    {
      "epoch": 510.53097345132744,
      "grad_norm": 8.25,
      "learning_rate": 8.616478770109646e-06,
      "loss": 1.108,
      "num_input_tokens_seen": 16129496,
      "step": 29100
    },
    {
      "epoch": 510.6194690265487,
      "grad_norm": 10.375,
      "learning_rate": 8.609064566388111e-06,
      "loss": 1.125,
      "num_input_tokens_seen": 16132056,
      "step": 29105
    },
    {
      "epoch": 510.7079646017699,
      "grad_norm": 7.59375,
      "learning_rate": 8.601652890354815e-06,
      "loss": 1.0126,
      "num_input_tokens_seen": 16135272,
      "step": 29110
    },
    {
      "epoch": 510.79646017699116,
      "grad_norm": 6.0625,
      "learning_rate": 8.594243743152705e-06,
      "loss": 0.7946,
      "num_input_tokens_seen": 16137960,
      "step": 29115
    },
    {
      "epoch": 510.88495575221236,
      "grad_norm": 5.8125,
      "learning_rate": 8.58683712592438e-06,
      "loss": 0.9542,
      "num_input_tokens_seen": 16140488,
      "step": 29120
    },
    {
      "epoch": 510.9734513274336,
      "grad_norm": 13.5,
      "learning_rate": 8.579433039812037e-06,
      "loss": 1.1418,
      "num_input_tokens_seen": 16143272,
      "step": 29125
    },
    {
      "epoch": 511.05309734513276,
      "grad_norm": 10.4375,
      "learning_rate": 8.572031485957466e-06,
      "loss": 1.1517,
      "num_input_tokens_seen": 16145488,
      "step": 29130
    },
    {
      "epoch": 511.14159292035396,
      "grad_norm": 8.125,
      "learning_rate": 8.564632465502084e-06,
      "loss": 1.0269,
      "num_input_tokens_seen": 16148048,
      "step": 29135
    },
    {
      "epoch": 511.2300884955752,
      "grad_norm": 10.4375,
      "learning_rate": 8.557235979586928e-06,
      "loss": 0.9555,
      "num_input_tokens_seen": 16151088,
      "step": 29140
    },
    {
      "epoch": 511.3185840707965,
      "grad_norm": 8.375,
      "learning_rate": 8.549842029352606e-06,
      "loss": 1.0091,
      "num_input_tokens_seen": 16153536,
      "step": 29145
    },
    {
      "epoch": 511.4070796460177,
      "grad_norm": 9.625,
      "learning_rate": 8.542450615939376e-06,
      "loss": 1.2648,
      "num_input_tokens_seen": 16156304,
      "step": 29150
    },
    {
      "epoch": 511.49557522123894,
      "grad_norm": 7.625,
      "learning_rate": 8.535061740487082e-06,
      "loss": 1.2026,
      "num_input_tokens_seen": 16159168,
      "step": 29155
    },
    {
      "epoch": 511.5840707964602,
      "grad_norm": 8.25,
      "learning_rate": 8.527675404135168e-06,
      "loss": 0.9152,
      "num_input_tokens_seen": 16162064,
      "step": 29160
    },
    {
      "epoch": 511.6725663716814,
      "grad_norm": 8.6875,
      "learning_rate": 8.520291608022724e-06,
      "loss": 0.929,
      "num_input_tokens_seen": 16164608,
      "step": 29165
    },
    {
      "epoch": 511.76106194690266,
      "grad_norm": 7.03125,
      "learning_rate": 8.512910353288398e-06,
      "loss": 1.025,
      "num_input_tokens_seen": 16167616,
      "step": 29170
    },
    {
      "epoch": 511.8495575221239,
      "grad_norm": 8.875,
      "learning_rate": 8.505531641070486e-06,
      "loss": 1.0043,
      "num_input_tokens_seen": 16171072,
      "step": 29175
    },
    {
      "epoch": 511.9380530973451,
      "grad_norm": 7.65625,
      "learning_rate": 8.498155472506885e-06,
      "loss": 0.9939,
      "num_input_tokens_seen": 16173776,
      "step": 29180
    },
    {
      "epoch": 512.0176991150443,
      "grad_norm": 9.6875,
      "learning_rate": 8.49078184873508e-06,
      "loss": 1.2275,
      "num_input_tokens_seen": 16176216,
      "step": 29185
    },
    {
      "epoch": 512.1061946902655,
      "grad_norm": 6.25,
      "learning_rate": 8.483410770892188e-06,
      "loss": 0.8625,
      "num_input_tokens_seen": 16178936,
      "step": 29190
    },
    {
      "epoch": 512.1946902654868,
      "grad_norm": 8.8125,
      "learning_rate": 8.476042240114909e-06,
      "loss": 0.9964,
      "num_input_tokens_seen": 16181992,
      "step": 29195
    },
    {
      "epoch": 512.2831858407079,
      "grad_norm": 12.625,
      "learning_rate": 8.468676257539568e-06,
      "loss": 1.0928,
      "num_input_tokens_seen": 16184680,
      "step": 29200
    },
    {
      "epoch": 512.2831858407079,
      "eval_loss": 1.0310944318771362,
      "eval_runtime": 0.9383,
      "eval_samples_per_second": 26.643,
      "eval_steps_per_second": 13.854,
      "num_input_tokens_seen": 16184680,
      "step": 29200
    },
    {
      "epoch": 512.3716814159292,
      "grad_norm": 8.875,
      "learning_rate": 8.4613128243021e-06,
      "loss": 1.1334,
      "num_input_tokens_seen": 16187544,
      "step": 29205
    },
    {
      "epoch": 512.4601769911504,
      "grad_norm": 9.0625,
      "learning_rate": 8.453951941538028e-06,
      "loss": 1.0862,
      "num_input_tokens_seen": 16190952,
      "step": 29210
    },
    {
      "epoch": 512.5486725663717,
      "grad_norm": 10.5625,
      "learning_rate": 8.446593610382495e-06,
      "loss": 1.2188,
      "num_input_tokens_seen": 16193144,
      "step": 29215
    },
    {
      "epoch": 512.637168141593,
      "grad_norm": 7.84375,
      "learning_rate": 8.439237831970259e-06,
      "loss": 0.9369,
      "num_input_tokens_seen": 16195960,
      "step": 29220
    },
    {
      "epoch": 512.7256637168142,
      "grad_norm": 7.59375,
      "learning_rate": 8.431884607435667e-06,
      "loss": 0.9152,
      "num_input_tokens_seen": 16198920,
      "step": 29225
    },
    {
      "epoch": 512.8141592920354,
      "grad_norm": 7.59375,
      "learning_rate": 8.424533937912665e-06,
      "loss": 1.0759,
      "num_input_tokens_seen": 16201704,
      "step": 29230
    },
    {
      "epoch": 512.9026548672566,
      "grad_norm": 7.5,
      "learning_rate": 8.41718582453484e-06,
      "loss": 1.1016,
      "num_input_tokens_seen": 16204520,
      "step": 29235
    },
    {
      "epoch": 512.9911504424779,
      "grad_norm": 7.34375,
      "learning_rate": 8.409840268435346e-06,
      "loss": 1.0437,
      "num_input_tokens_seen": 16207368,
      "step": 29240
    },
    {
      "epoch": 513.070796460177,
      "grad_norm": 8.9375,
      "learning_rate": 8.402497270746976e-06,
      "loss": 1.1386,
      "num_input_tokens_seen": 16210104,
      "step": 29245
    },
    {
      "epoch": 513.1592920353983,
      "grad_norm": 7.5,
      "learning_rate": 8.395156832602095e-06,
      "loss": 1.1805,
      "num_input_tokens_seen": 16212744,
      "step": 29250
    },
    {
      "epoch": 513.2477876106195,
      "grad_norm": 9.0625,
      "learning_rate": 8.387818955132707e-06,
      "loss": 1.0404,
      "num_input_tokens_seen": 16215352,
      "step": 29255
    },
    {
      "epoch": 513.3362831858407,
      "grad_norm": 8.0,
      "learning_rate": 8.38048363947039e-06,
      "loss": 1.0195,
      "num_input_tokens_seen": 16218376,
      "step": 29260
    },
    {
      "epoch": 513.4247787610619,
      "grad_norm": 9.1875,
      "learning_rate": 8.373150886746351e-06,
      "loss": 1.0875,
      "num_input_tokens_seen": 16221064,
      "step": 29265
    },
    {
      "epoch": 513.5132743362832,
      "grad_norm": 8.3125,
      "learning_rate": 8.365820698091397e-06,
      "loss": 0.9433,
      "num_input_tokens_seen": 16223976,
      "step": 29270
    },
    {
      "epoch": 513.6017699115044,
      "grad_norm": 8.0625,
      "learning_rate": 8.358493074635922e-06,
      "loss": 1.2215,
      "num_input_tokens_seen": 16226600,
      "step": 29275
    },
    {
      "epoch": 513.6902654867257,
      "grad_norm": 7.84375,
      "learning_rate": 8.351168017509948e-06,
      "loss": 1.0826,
      "num_input_tokens_seen": 16229416,
      "step": 29280
    },
    {
      "epoch": 513.7787610619469,
      "grad_norm": 9.5625,
      "learning_rate": 8.343845527843094e-06,
      "loss": 0.8223,
      "num_input_tokens_seen": 16232392,
      "step": 29285
    },
    {
      "epoch": 513.8672566371681,
      "grad_norm": 7.15625,
      "learning_rate": 8.336525606764566e-06,
      "loss": 0.9661,
      "num_input_tokens_seen": 16234744,
      "step": 29290
    },
    {
      "epoch": 513.9557522123894,
      "grad_norm": 10.4375,
      "learning_rate": 8.329208255403204e-06,
      "loss": 1.1216,
      "num_input_tokens_seen": 16237896,
      "step": 29295
    },
    {
      "epoch": 514.0353982300885,
      "grad_norm": 7.4375,
      "learning_rate": 8.321893474887426e-06,
      "loss": 0.8224,
      "num_input_tokens_seen": 16240040,
      "step": 29300
    },
    {
      "epoch": 514.1238938053098,
      "grad_norm": 10.8125,
      "learning_rate": 8.31458126634526e-06,
      "loss": 1.2821,
      "num_input_tokens_seen": 16242520,
      "step": 29305
    },
    {
      "epoch": 514.212389380531,
      "grad_norm": 8.25,
      "learning_rate": 8.30727163090435e-06,
      "loss": 0.9602,
      "num_input_tokens_seen": 16244824,
      "step": 29310
    },
    {
      "epoch": 514.3008849557522,
      "grad_norm": 6.75,
      "learning_rate": 8.29996456969192e-06,
      "loss": 0.9108,
      "num_input_tokens_seen": 16247560,
      "step": 29315
    },
    {
      "epoch": 514.3893805309734,
      "grad_norm": 7.0625,
      "learning_rate": 8.292660083834818e-06,
      "loss": 1.0824,
      "num_input_tokens_seen": 16250200,
      "step": 29320
    },
    {
      "epoch": 514.4778761061947,
      "grad_norm": 7.6875,
      "learning_rate": 8.2853581744595e-06,
      "loss": 0.8831,
      "num_input_tokens_seen": 16253256,
      "step": 29325
    },
    {
      "epoch": 514.566371681416,
      "grad_norm": 9.6875,
      "learning_rate": 8.278058842691991e-06,
      "loss": 1.0907,
      "num_input_tokens_seen": 16256328,
      "step": 29330
    },
    {
      "epoch": 514.6548672566372,
      "grad_norm": 6.15625,
      "learning_rate": 8.27076208965796e-06,
      "loss": 0.8669,
      "num_input_tokens_seen": 16259400,
      "step": 29335
    },
    {
      "epoch": 514.7433628318585,
      "grad_norm": 9.9375,
      "learning_rate": 8.263467916482637e-06,
      "loss": 1.0798,
      "num_input_tokens_seen": 16262360,
      "step": 29340
    },
    {
      "epoch": 514.8318584070796,
      "grad_norm": 12.1875,
      "learning_rate": 8.256176324290885e-06,
      "loss": 1.2419,
      "num_input_tokens_seen": 16265224,
      "step": 29345
    },
    {
      "epoch": 514.9203539823009,
      "grad_norm": 8.0,
      "learning_rate": 8.248887314207168e-06,
      "loss": 1.0855,
      "num_input_tokens_seen": 16268312,
      "step": 29350
    },
    {
      "epoch": 515.0,
      "grad_norm": 15.8125,
      "learning_rate": 8.24160088735553e-06,
      "loss": 1.1832,
      "num_input_tokens_seen": 16270496,
      "step": 29355
    },
    {
      "epoch": 515.0884955752213,
      "grad_norm": 8.5625,
      "learning_rate": 8.234317044859629e-06,
      "loss": 0.9865,
      "num_input_tokens_seen": 16272912,
      "step": 29360
    },
    {
      "epoch": 515.1769911504425,
      "grad_norm": 8.4375,
      "learning_rate": 8.227035787842744e-06,
      "loss": 0.9053,
      "num_input_tokens_seen": 16275504,
      "step": 29365
    },
    {
      "epoch": 515.2654867256637,
      "grad_norm": 6.875,
      "learning_rate": 8.219757117427721e-06,
      "loss": 0.8472,
      "num_input_tokens_seen": 16278176,
      "step": 29370
    },
    {
      "epoch": 515.3539823008849,
      "grad_norm": 5.53125,
      "learning_rate": 8.212481034737014e-06,
      "loss": 1.0436,
      "num_input_tokens_seen": 16281488,
      "step": 29375
    },
    {
      "epoch": 515.4424778761062,
      "grad_norm": 12.8125,
      "learning_rate": 8.205207540892707e-06,
      "loss": 1.0978,
      "num_input_tokens_seen": 16284576,
      "step": 29380
    },
    {
      "epoch": 515.5309734513274,
      "grad_norm": 12.3125,
      "learning_rate": 8.197936637016442e-06,
      "loss": 1.2479,
      "num_input_tokens_seen": 16287104,
      "step": 29385
    },
    {
      "epoch": 515.6194690265487,
      "grad_norm": 7.84375,
      "learning_rate": 8.190668324229508e-06,
      "loss": 1.0563,
      "num_input_tokens_seen": 16289936,
      "step": 29390
    },
    {
      "epoch": 515.70796460177,
      "grad_norm": 8.0625,
      "learning_rate": 8.183402603652749e-06,
      "loss": 1.0893,
      "num_input_tokens_seen": 16293120,
      "step": 29395
    },
    {
      "epoch": 515.7964601769911,
      "grad_norm": 9.375,
      "learning_rate": 8.176139476406635e-06,
      "loss": 1.1411,
      "num_input_tokens_seen": 16295584,
      "step": 29400
    },
    {
      "epoch": 515.7964601769911,
      "eval_loss": 1.0310944318771362,
      "eval_runtime": 0.9376,
      "eval_samples_per_second": 26.664,
      "eval_steps_per_second": 13.865,
      "num_input_tokens_seen": 16295584,
      "step": 29400
    },
    {
      "epoch": 515.8849557522124,
      "grad_norm": 8.75,
      "learning_rate": 8.16887894361125e-06,
      "loss": 0.9218,
      "num_input_tokens_seen": 16298672,
      "step": 29405
    },
    {
      "epoch": 515.9734513274336,
      "grad_norm": 8.625,
      "learning_rate": 8.161621006386233e-06,
      "loss": 1.1576,
      "num_input_tokens_seen": 16301408,
      "step": 29410
    },
    {
      "epoch": 516.0530973451328,
      "grad_norm": 10.4375,
      "learning_rate": 8.154365665850869e-06,
      "loss": 1.3915,
      "num_input_tokens_seen": 16303416,
      "step": 29415
    },
    {
      "epoch": 516.141592920354,
      "grad_norm": 11.125,
      "learning_rate": 8.147112923124005e-06,
      "loss": 1.074,
      "num_input_tokens_seen": 16305976,
      "step": 29420
    },
    {
      "epoch": 516.2300884955753,
      "grad_norm": 7.625,
      "learning_rate": 8.13986277932412e-06,
      "loss": 0.9668,
      "num_input_tokens_seen": 16309320,
      "step": 29425
    },
    {
      "epoch": 516.3185840707964,
      "grad_norm": 9.25,
      "learning_rate": 8.132615235569277e-06,
      "loss": 1.1172,
      "num_input_tokens_seen": 16312232,
      "step": 29430
    },
    {
      "epoch": 516.4070796460177,
      "grad_norm": 6.28125,
      "learning_rate": 8.125370292977124e-06,
      "loss": 1.0165,
      "num_input_tokens_seen": 16314728,
      "step": 29435
    },
    {
      "epoch": 516.4955752212389,
      "grad_norm": 14.5625,
      "learning_rate": 8.118127952664944e-06,
      "loss": 0.9518,
      "num_input_tokens_seen": 16317576,
      "step": 29440
    },
    {
      "epoch": 516.5840707964602,
      "grad_norm": 8.375,
      "learning_rate": 8.110888215749574e-06,
      "loss": 1.0774,
      "num_input_tokens_seen": 16320184,
      "step": 29445
    },
    {
      "epoch": 516.6725663716815,
      "grad_norm": 9.375,
      "learning_rate": 8.10365108334749e-06,
      "loss": 1.0472,
      "num_input_tokens_seen": 16323048,
      "step": 29450
    },
    {
      "epoch": 516.7610619469026,
      "grad_norm": 12.1875,
      "learning_rate": 8.096416556574743e-06,
      "loss": 0.983,
      "num_input_tokens_seen": 16325624,
      "step": 29455
    },
    {
      "epoch": 516.8495575221239,
      "grad_norm": 8.125,
      "learning_rate": 8.08918463654698e-06,
      "loss": 0.9735,
      "num_input_tokens_seen": 16328760,
      "step": 29460
    },
    {
      "epoch": 516.9380530973451,
      "grad_norm": 7.21875,
      "learning_rate": 8.081955324379458e-06,
      "loss": 1.1063,
      "num_input_tokens_seen": 16331640,
      "step": 29465
    },
    {
      "epoch": 517.0176991150443,
      "grad_norm": 7.65625,
      "learning_rate": 8.074728621187039e-06,
      "loss": 1.1232,
      "num_input_tokens_seen": 16333728,
      "step": 29470
    },
    {
      "epoch": 517.1061946902655,
      "grad_norm": 5.9375,
      "learning_rate": 8.067504528084158e-06,
      "loss": 1.0793,
      "num_input_tokens_seen": 16336864,
      "step": 29475
    },
    {
      "epoch": 517.1946902654868,
      "grad_norm": 10.5,
      "learning_rate": 8.060283046184861e-06,
      "loss": 1.1882,
      "num_input_tokens_seen": 16339584,
      "step": 29480
    },
    {
      "epoch": 517.2831858407079,
      "grad_norm": 8.6875,
      "learning_rate": 8.053064176602806e-06,
      "loss": 1.0112,
      "num_input_tokens_seen": 16342528,
      "step": 29485
    },
    {
      "epoch": 517.3716814159292,
      "grad_norm": 11.4375,
      "learning_rate": 8.045847920451216e-06,
      "loss": 1.1965,
      "num_input_tokens_seen": 16345392,
      "step": 29490
    },
    {
      "epoch": 517.4601769911504,
      "grad_norm": 11.0625,
      "learning_rate": 8.038634278842944e-06,
      "loss": 1.0426,
      "num_input_tokens_seen": 16347968,
      "step": 29495
    },
    {
      "epoch": 517.5486725663717,
      "grad_norm": 7.03125,
      "learning_rate": 8.031423252890408e-06,
      "loss": 1.0471,
      "num_input_tokens_seen": 16350304,
      "step": 29500
    },
    {
      "epoch": 517.637168141593,
      "grad_norm": 8.3125,
      "learning_rate": 8.024214843705646e-06,
      "loss": 1.0718,
      "num_input_tokens_seen": 16353184,
      "step": 29505
    },
    {
      "epoch": 517.7256637168142,
      "grad_norm": 8.375,
      "learning_rate": 8.017009052400295e-06,
      "loss": 0.9311,
      "num_input_tokens_seen": 16355664,
      "step": 29510
    },
    {
      "epoch": 517.8141592920354,
      "grad_norm": 6.84375,
      "learning_rate": 8.00980588008557e-06,
      "loss": 1.0508,
      "num_input_tokens_seen": 16358496,
      "step": 29515
    },
    {
      "epoch": 517.9026548672566,
      "grad_norm": 7.875,
      "learning_rate": 8.002605327872282e-06,
      "loss": 0.863,
      "num_input_tokens_seen": 16361456,
      "step": 29520
    },
    {
      "epoch": 517.9911504424779,
      "grad_norm": 8.1875,
      "learning_rate": 7.995407396870862e-06,
      "loss": 1.1108,
      "num_input_tokens_seen": 16364432,
      "step": 29525
    },
    {
      "epoch": 518.070796460177,
      "grad_norm": 7.90625,
      "learning_rate": 7.988212088191307e-06,
      "loss": 1.2756,
      "num_input_tokens_seen": 16366984,
      "step": 29530
    },
    {
      "epoch": 518.1592920353983,
      "grad_norm": 9.0625,
      "learning_rate": 7.98101940294324e-06,
      "loss": 1.0093,
      "num_input_tokens_seen": 16369816,
      "step": 29535
    },
    {
      "epoch": 518.2477876106195,
      "grad_norm": 9.4375,
      "learning_rate": 7.973829342235847e-06,
      "loss": 1.0061,
      "num_input_tokens_seen": 16372760,
      "step": 29540
    },
    {
      "epoch": 518.3362831858407,
      "grad_norm": 9.125,
      "learning_rate": 7.966641907177936e-06,
      "loss": 0.8965,
      "num_input_tokens_seen": 16376264,
      "step": 29545
    },
    {
      "epoch": 518.4247787610619,
      "grad_norm": 6.28125,
      "learning_rate": 7.959457098877901e-06,
      "loss": 1.0627,
      "num_input_tokens_seen": 16379080,
      "step": 29550
    },
    {
      "epoch": 518.5132743362832,
      "grad_norm": 11.375,
      "learning_rate": 7.952274918443719e-06,
      "loss": 1.209,
      "num_input_tokens_seen": 16381304,
      "step": 29555
    },
    {
      "epoch": 518.6017699115044,
      "grad_norm": 8.6875,
      "learning_rate": 7.945095366982983e-06,
      "loss": 0.7289,
      "num_input_tokens_seen": 16383832,
      "step": 29560
    },
    {
      "epoch": 518.6902654867257,
      "grad_norm": 11.3125,
      "learning_rate": 7.937918445602871e-06,
      "loss": 1.1641,
      "num_input_tokens_seen": 16386728,
      "step": 29565
    },
    {
      "epoch": 518.7787610619469,
      "grad_norm": 8.0625,
      "learning_rate": 7.930744155410145e-06,
      "loss": 1.0842,
      "num_input_tokens_seen": 16389688,
      "step": 29570
    },
    {
      "epoch": 518.8672566371681,
      "grad_norm": 8.125,
      "learning_rate": 7.923572497511181e-06,
      "loss": 1.1792,
      "num_input_tokens_seen": 16392728,
      "step": 29575
    },
    {
      "epoch": 518.9557522123894,
      "grad_norm": 7.65625,
      "learning_rate": 7.916403473011927e-06,
      "loss": 1.053,
      "num_input_tokens_seen": 16395256,
      "step": 29580
    },
    {
      "epoch": 519.0353982300885,
      "grad_norm": 8.375,
      "learning_rate": 7.909237083017953e-06,
      "loss": 1.1632,
      "num_input_tokens_seen": 16397512,
      "step": 29585
    },
    {
      "epoch": 519.1238938053098,
      "grad_norm": 7.3125,
      "learning_rate": 7.902073328634389e-06,
      "loss": 0.8887,
      "num_input_tokens_seen": 16400840,
      "step": 29590
    },
    {
      "epoch": 519.212389380531,
      "grad_norm": 8.5625,
      "learning_rate": 7.894912210965987e-06,
      "loss": 0.9984,
      "num_input_tokens_seen": 16403816,
      "step": 29595
    },
    {
      "epoch": 519.3008849557522,
      "grad_norm": 8.1875,
      "learning_rate": 7.887753731117075e-06,
      "loss": 1.1284,
      "num_input_tokens_seen": 16406536,
      "step": 29600
    },
    {
      "epoch": 519.3008849557522,
      "eval_loss": 1.0310944318771362,
      "eval_runtime": 0.9344,
      "eval_samples_per_second": 26.755,
      "eval_steps_per_second": 13.913,
      "num_input_tokens_seen": 16406536,
      "step": 29600
    },
    {
      "epoch": 519.3893805309734,
      "grad_norm": 8.0,
      "learning_rate": 7.880597890191587e-06,
      "loss": 1.048,
      "num_input_tokens_seen": 16409096,
      "step": 29605
    },
    {
      "epoch": 519.4778761061947,
      "grad_norm": 7.59375,
      "learning_rate": 7.873444689293036e-06,
      "loss": 1.1132,
      "num_input_tokens_seen": 16412040,
      "step": 29610
    },
    {
      "epoch": 519.566371681416,
      "grad_norm": 7.90625,
      "learning_rate": 7.866294129524548e-06,
      "loss": 1.203,
      "num_input_tokens_seen": 16414584,
      "step": 29615
    },
    {
      "epoch": 519.6548672566372,
      "grad_norm": 7.9375,
      "learning_rate": 7.859146211988811e-06,
      "loss": 0.9287,
      "num_input_tokens_seen": 16417496,
      "step": 29620
    },
    {
      "epoch": 519.7433628318585,
      "grad_norm": 9.0625,
      "learning_rate": 7.852000937788134e-06,
      "loss": 1.0098,
      "num_input_tokens_seen": 16420072,
      "step": 29625
    },
    {
      "epoch": 519.8318584070796,
      "grad_norm": 12.125,
      "learning_rate": 7.844858308024416e-06,
      "loss": 1.0874,
      "num_input_tokens_seen": 16422824,
      "step": 29630
    },
    {
      "epoch": 519.9203539823009,
      "grad_norm": 9.25,
      "learning_rate": 7.837718323799122e-06,
      "loss": 1.0205,
      "num_input_tokens_seen": 16425784,
      "step": 29635
    },
    {
      "epoch": 520.0,
      "grad_norm": 9.375,
      "learning_rate": 7.83058098621334e-06,
      "loss": 1.1264,
      "num_input_tokens_seen": 16428160,
      "step": 29640
    },
    {
      "epoch": 520.0884955752213,
      "grad_norm": 9.0625,
      "learning_rate": 7.823446296367739e-06,
      "loss": 1.1567,
      "num_input_tokens_seen": 16431360,
      "step": 29645
    },
    {
      "epoch": 520.1769911504425,
      "grad_norm": 10.6875,
      "learning_rate": 7.81631425536257e-06,
      "loss": 1.0714,
      "num_input_tokens_seen": 16434240,
      "step": 29650
    },
    {
      "epoch": 520.2654867256637,
      "grad_norm": 5.71875,
      "learning_rate": 7.809184864297689e-06,
      "loss": 0.9961,
      "num_input_tokens_seen": 16436960,
      "step": 29655
    },
    {
      "epoch": 520.3539823008849,
      "grad_norm": 8.5625,
      "learning_rate": 7.802058124272532e-06,
      "loss": 1.0582,
      "num_input_tokens_seen": 16439456,
      "step": 29660
    },
    {
      "epoch": 520.4424778761062,
      "grad_norm": 7.46875,
      "learning_rate": 7.79493403638614e-06,
      "loss": 0.9247,
      "num_input_tokens_seen": 16442288,
      "step": 29665
    },
    {
      "epoch": 520.5309734513274,
      "grad_norm": 10.5625,
      "learning_rate": 7.787812601737132e-06,
      "loss": 1.2497,
      "num_input_tokens_seen": 16444528,
      "step": 29670
    },
    {
      "epoch": 520.6194690265487,
      "grad_norm": 8.5,
      "learning_rate": 7.780693821423715e-06,
      "loss": 1.1168,
      "num_input_tokens_seen": 16447200,
      "step": 29675
    },
    {
      "epoch": 520.70796460177,
      "grad_norm": 8.8125,
      "learning_rate": 7.773577696543705e-06,
      "loss": 1.0485,
      "num_input_tokens_seen": 16450000,
      "step": 29680
    },
    {
      "epoch": 520.7964601769911,
      "grad_norm": 7.9375,
      "learning_rate": 7.7664642281945e-06,
      "loss": 1.0705,
      "num_input_tokens_seen": 16452960,
      "step": 29685
    },
    {
      "epoch": 520.8849557522124,
      "grad_norm": 10.3125,
      "learning_rate": 7.759353417473072e-06,
      "loss": 1.1825,
      "num_input_tokens_seen": 16455744,
      "step": 29690
    },
    {
      "epoch": 520.9734513274336,
      "grad_norm": 10.5,
      "learning_rate": 7.752245265476016e-06,
      "loss": 0.777,
      "num_input_tokens_seen": 16458976,
      "step": 29695
    },
    {
      "epoch": 521.0530973451328,
      "grad_norm": 10.375,
      "learning_rate": 7.745139773299481e-06,
      "loss": 1.0092,
      "num_input_tokens_seen": 16461576,
      "step": 29700
    },
    {
      "epoch": 521.141592920354,
      "grad_norm": 8.1875,
      "learning_rate": 7.738036942039232e-06,
      "loss": 1.0971,
      "num_input_tokens_seen": 16464168,
      "step": 29705
    },
    {
      "epoch": 521.2300884955753,
      "grad_norm": 8.875,
      "learning_rate": 7.73093677279062e-06,
      "loss": 1.0026,
      "num_input_tokens_seen": 16466824,
      "step": 29710
    },
    {
      "epoch": 521.3185840707964,
      "grad_norm": 8.5625,
      "learning_rate": 7.72383926664857e-06,
      "loss": 0.9961,
      "num_input_tokens_seen": 16469736,
      "step": 29715
    },
    {
      "epoch": 521.4070796460177,
      "grad_norm": 6.8125,
      "learning_rate": 7.716744424707606e-06,
      "loss": 1.0796,
      "num_input_tokens_seen": 16472456,
      "step": 29720
    },
    {
      "epoch": 521.4955752212389,
      "grad_norm": 8.6875,
      "learning_rate": 7.709652248061858e-06,
      "loss": 0.8076,
      "num_input_tokens_seen": 16475240,
      "step": 29725
    },
    {
      "epoch": 521.5840707964602,
      "grad_norm": 8.125,
      "learning_rate": 7.702562737805017e-06,
      "loss": 0.9888,
      "num_input_tokens_seen": 16478040,
      "step": 29730
    },
    {
      "epoch": 521.6725663716815,
      "grad_norm": 7.4375,
      "learning_rate": 7.695475895030365e-06,
      "loss": 0.9935,
      "num_input_tokens_seen": 16480776,
      "step": 29735
    },
    {
      "epoch": 521.7610619469026,
      "grad_norm": 6.625,
      "learning_rate": 7.6883917208308e-06,
      "loss": 1.0469,
      "num_input_tokens_seen": 16483192,
      "step": 29740
    },
    {
      "epoch": 521.8495575221239,
      "grad_norm": 8.625,
      "learning_rate": 7.681310216298778e-06,
      "loss": 1.1102,
      "num_input_tokens_seen": 16486296,
      "step": 29745
    },
    {
      "epoch": 521.9380530973451,
      "grad_norm": 7.75,
      "learning_rate": 7.674231382526367e-06,
      "loss": 1.2576,
      "num_input_tokens_seen": 16489416,
      "step": 29750
    },
    {
      "epoch": 522.0176991150443,
      "grad_norm": 8.1875,
      "learning_rate": 7.667155220605198e-06,
      "loss": 1.1691,
      "num_input_tokens_seen": 16491704,
      "step": 29755
    },
    {
      "epoch": 522.1061946902655,
      "grad_norm": 10.0625,
      "learning_rate": 7.660081731626515e-06,
      "loss": 0.9925,
      "num_input_tokens_seen": 16494136,
      "step": 29760
    },
    {
      "epoch": 522.1946902654868,
      "grad_norm": 9.4375,
      "learning_rate": 7.653010916681141e-06,
      "loss": 1.0773,
      "num_input_tokens_seen": 16496792,
      "step": 29765
    },
    {
      "epoch": 522.2831858407079,
      "grad_norm": 12.4375,
      "learning_rate": 7.645942776859472e-06,
      "loss": 1.0014,
      "num_input_tokens_seen": 16499896,
      "step": 29770
    },
    {
      "epoch": 522.3716814159292,
      "grad_norm": 9.6875,
      "learning_rate": 7.63887731325152e-06,
      "loss": 1.1456,
      "num_input_tokens_seen": 16503000,
      "step": 29775
    },
    {
      "epoch": 522.4601769911504,
      "grad_norm": 7.71875,
      "learning_rate": 7.63181452694685e-06,
      "loss": 1.0295,
      "num_input_tokens_seen": 16505544,
      "step": 29780
    },
    {
      "epoch": 522.5486725663717,
      "grad_norm": 9.5,
      "learning_rate": 7.624754419034644e-06,
      "loss": 1.0171,
      "num_input_tokens_seen": 16508040,
      "step": 29785
    },
    {
      "epoch": 522.637168141593,
      "grad_norm": 9.3125,
      "learning_rate": 7.6176969906036645e-06,
      "loss": 1.0267,
      "num_input_tokens_seen": 16510728,
      "step": 29790
    },
    {
      "epoch": 522.7256637168142,
      "grad_norm": 6.875,
      "learning_rate": 7.610642242742242e-06,
      "loss": 1.1462,
      "num_input_tokens_seen": 16513512,
      "step": 29795
    },
    {
      "epoch": 522.8141592920354,
      "grad_norm": 6.1875,
      "learning_rate": 7.603590176538322e-06,
      "loss": 0.8901,
      "num_input_tokens_seen": 16516648,
      "step": 29800
    },
    {
      "epoch": 522.8141592920354,
      "eval_loss": 1.0310944318771362,
      "eval_runtime": 1.0194,
      "eval_samples_per_second": 24.525,
      "eval_steps_per_second": 12.753,
      "num_input_tokens_seen": 16516648,
      "step": 29800
    },
    {
      "epoch": 522.9026548672566,
      "grad_norm": 7.40625,
      "learning_rate": 7.596540793079404e-06,
      "loss": 1.0884,
      "num_input_tokens_seen": 16519720,
      "step": 29805
    },
    {
      "epoch": 522.9911504424779,
      "grad_norm": 8.8125,
      "learning_rate": 7.5894940934526125e-06,
      "loss": 1.1396,
      "num_input_tokens_seen": 16522200,
      "step": 29810
    },
    {
      "epoch": 523.070796460177,
      "grad_norm": 8.5,
      "learning_rate": 7.582450078744621e-06,
      "loss": 1.0178,
      "num_input_tokens_seen": 16524648,
      "step": 29815
    },
    {
      "epoch": 523.1592920353983,
      "grad_norm": 8.5625,
      "learning_rate": 7.575408750041707e-06,
      "loss": 1.1131,
      "num_input_tokens_seen": 16527928,
      "step": 29820
    },
    {
      "epoch": 523.2477876106195,
      "grad_norm": 8.25,
      "learning_rate": 7.568370108429732e-06,
      "loss": 0.9041,
      "num_input_tokens_seen": 16530584,
      "step": 29825
    },
    {
      "epoch": 523.3362831858407,
      "grad_norm": 8.3125,
      "learning_rate": 7.561334154994154e-06,
      "loss": 0.9643,
      "num_input_tokens_seen": 16533272,
      "step": 29830
    },
    {
      "epoch": 523.4247787610619,
      "grad_norm": 7.0625,
      "learning_rate": 7.55430089081999e-06,
      "loss": 1.0433,
      "num_input_tokens_seen": 16536392,
      "step": 29835
    },
    {
      "epoch": 523.5132743362832,
      "grad_norm": 6.75,
      "learning_rate": 7.547270316991864e-06,
      "loss": 1.0456,
      "num_input_tokens_seen": 16539144,
      "step": 29840
    },
    {
      "epoch": 523.6017699115044,
      "grad_norm": 7.625,
      "learning_rate": 7.5402424345939884e-06,
      "loss": 0.9003,
      "num_input_tokens_seen": 16542328,
      "step": 29845
    },
    {
      "epoch": 523.6902654867257,
      "grad_norm": 8.3125,
      "learning_rate": 7.533217244710133e-06,
      "loss": 1.309,
      "num_input_tokens_seen": 16544600,
      "step": 29850
    },
    {
      "epoch": 523.7787610619469,
      "grad_norm": 6.9375,
      "learning_rate": 7.52619474842369e-06,
      "loss": 1.0956,
      "num_input_tokens_seen": 16547592,
      "step": 29855
    },
    {
      "epoch": 523.8672566371681,
      "grad_norm": 11.125,
      "learning_rate": 7.519174946817597e-06,
      "loss": 1.0333,
      "num_input_tokens_seen": 16550248,
      "step": 29860
    },
    {
      "epoch": 523.9557522123894,
      "grad_norm": 7.0,
      "learning_rate": 7.512157840974407e-06,
      "loss": 1.0225,
      "num_input_tokens_seen": 16553336,
      "step": 29865
    },
    {
      "epoch": 524.0353982300885,
      "grad_norm": 7.28125,
      "learning_rate": 7.5051434319762496e-06,
      "loss": 1.403,
      "num_input_tokens_seen": 16555256,
      "step": 29870
    },
    {
      "epoch": 524.1238938053098,
      "grad_norm": 10.5625,
      "learning_rate": 7.498131720904822e-06,
      "loss": 0.998,
      "num_input_tokens_seen": 16557784,
      "step": 29875
    },
    {
      "epoch": 524.212389380531,
      "grad_norm": 10.875,
      "learning_rate": 7.491122708841433e-06,
      "loss": 1.1219,
      "num_input_tokens_seen": 16560568,
      "step": 29880
    },
    {
      "epoch": 524.3008849557522,
      "grad_norm": 7.5625,
      "learning_rate": 7.4841163968669524e-06,
      "loss": 1.0859,
      "num_input_tokens_seen": 16563912,
      "step": 29885
    },
    {
      "epoch": 524.3893805309734,
      "grad_norm": 7.6875,
      "learning_rate": 7.4771127860618355e-06,
      "loss": 0.9291,
      "num_input_tokens_seen": 16566760,
      "step": 29890
    },
    {
      "epoch": 524.4778761061947,
      "grad_norm": 8.5625,
      "learning_rate": 7.470111877506139e-06,
      "loss": 1.1587,
      "num_input_tokens_seen": 16569192,
      "step": 29895
    },
    {
      "epoch": 524.566371681416,
      "grad_norm": 12.0,
      "learning_rate": 7.463113672279479e-06,
      "loss": 1.0012,
      "num_input_tokens_seen": 16571800,
      "step": 29900
    },
    {
      "epoch": 524.6548672566372,
      "grad_norm": 7.875,
      "learning_rate": 7.456118171461071e-06,
      "loss": 1.0362,
      "num_input_tokens_seen": 16574872,
      "step": 29905
    },
    {
      "epoch": 524.7433628318585,
      "grad_norm": 8.3125,
      "learning_rate": 7.449125376129721e-06,
      "loss": 1.2206,
      "num_input_tokens_seen": 16577688,
      "step": 29910
    },
    {
      "epoch": 524.8318584070796,
      "grad_norm": 6.8125,
      "learning_rate": 7.442135287363788e-06,
      "loss": 1.0362,
      "num_input_tokens_seen": 16580808,
      "step": 29915
    },
    {
      "epoch": 524.9203539823009,
      "grad_norm": 9.1875,
      "learning_rate": 7.435147906241247e-06,
      "loss": 1.069,
      "num_input_tokens_seen": 16583624,
      "step": 29920
    },
    {
      "epoch": 525.0,
      "grad_norm": 10.1875,
      "learning_rate": 7.428163233839624e-06,
      "loss": 0.9499,
      "num_input_tokens_seen": 16585880,
      "step": 29925
    },
    {
      "epoch": 525.0884955752213,
      "grad_norm": 8.6875,
      "learning_rate": 7.4211812712360525e-06,
      "loss": 1.1117,
      "num_input_tokens_seen": 16588072,
      "step": 29930
    },
    {
      "epoch": 525.1769911504425,
      "grad_norm": 9.125,
      "learning_rate": 7.4142020195072464e-06,
      "loss": 1.0813,
      "num_input_tokens_seen": 16590344,
      "step": 29935
    },
    {
      "epoch": 525.2654867256637,
      "grad_norm": 8.5,
      "learning_rate": 7.407225479729479e-06,
      "loss": 1.1519,
      "num_input_tokens_seen": 16593336,
      "step": 29940
    },
    {
      "epoch": 525.3539823008849,
      "grad_norm": 7.71875,
      "learning_rate": 7.400251652978632e-06,
      "loss": 0.9296,
      "num_input_tokens_seen": 16596600,
      "step": 29945
    },
    {
      "epoch": 525.4424778761062,
      "grad_norm": 9.1875,
      "learning_rate": 7.393280540330147e-06,
      "loss": 0.9304,
      "num_input_tokens_seen": 16599464,
      "step": 29950
    },
    {
      "epoch": 525.5309734513274,
      "grad_norm": 9.0625,
      "learning_rate": 7.386312142859069e-06,
      "loss": 1.1464,
      "num_input_tokens_seen": 16601976,
      "step": 29955
    },
    {
      "epoch": 525.6194690265487,
      "grad_norm": 8.375,
      "learning_rate": 7.379346461640008e-06,
      "loss": 1.049,
      "num_input_tokens_seen": 16605160,
      "step": 29960
    },
    {
      "epoch": 525.70796460177,
      "grad_norm": 7.875,
      "learning_rate": 7.372383497747149e-06,
      "loss": 1.0698,
      "num_input_tokens_seen": 16607976,
      "step": 29965
    },
    {
      "epoch": 525.7964601769911,
      "grad_norm": 10.6875,
      "learning_rate": 7.3654232522542775e-06,
      "loss": 1.1512,
      "num_input_tokens_seen": 16611288,
      "step": 29970
    },
    {
      "epoch": 525.8849557522124,
      "grad_norm": 8.5625,
      "learning_rate": 7.358465726234756e-06,
      "loss": 1.0037,
      "num_input_tokens_seen": 16614168,
      "step": 29975
    },
    {
      "epoch": 525.9734513274336,
      "grad_norm": 11.75,
      "learning_rate": 7.351510920761512e-06,
      "loss": 0.9882,
      "num_input_tokens_seen": 16617096,
      "step": 29980
    },
    {
      "epoch": 526.0530973451328,
      "grad_norm": 6.375,
      "learning_rate": 7.344558836907067e-06,
      "loss": 0.8127,
      "num_input_tokens_seen": 16619520,
      "step": 29985
    },
    {
      "epoch": 526.141592920354,
      "grad_norm": 7.9375,
      "learning_rate": 7.3376094757435285e-06,
      "loss": 0.8202,
      "num_input_tokens_seen": 16622160,
      "step": 29990
    },
    {
      "epoch": 526.2300884955753,
      "grad_norm": 8.25,
      "learning_rate": 7.330662838342561e-06,
      "loss": 1.1778,
      "num_input_tokens_seen": 16624992,
      "step": 29995
    },
    {
      "epoch": 526.3185840707964,
      "grad_norm": 6.84375,
      "learning_rate": 7.323718925775438e-06,
      "loss": 1.0365,
      "num_input_tokens_seen": 16628144,
      "step": 30000
    },
    {
      "epoch": 526.3185840707964,
      "eval_loss": 1.0310944318771362,
      "eval_runtime": 0.9351,
      "eval_samples_per_second": 26.734,
      "eval_steps_per_second": 13.902,
      "num_input_tokens_seen": 16628144,
      "step": 30000
    },
    {
      "epoch": 526.4070796460177,
      "grad_norm": 8.125,
      "learning_rate": 7.316777739112985e-06,
      "loss": 0.9216,
      "num_input_tokens_seen": 16631152,
      "step": 30005
    },
    {
      "epoch": 526.4955752212389,
      "grad_norm": 8.9375,
      "learning_rate": 7.309839279425626e-06,
      "loss": 1.1568,
      "num_input_tokens_seen": 16634176,
      "step": 30010
    },
    {
      "epoch": 526.5840707964602,
      "grad_norm": 8.4375,
      "learning_rate": 7.302903547783366e-06,
      "loss": 1.1742,
      "num_input_tokens_seen": 16637056,
      "step": 30015
    },
    {
      "epoch": 526.6725663716815,
      "grad_norm": 10.0625,
      "learning_rate": 7.2959705452557644e-06,
      "loss": 0.9975,
      "num_input_tokens_seen": 16639920,
      "step": 30020
    },
    {
      "epoch": 526.7610619469026,
      "grad_norm": 7.1875,
      "learning_rate": 7.289040272911996e-06,
      "loss": 1.0601,
      "num_input_tokens_seen": 16642288,
      "step": 30025
    },
    {
      "epoch": 526.8495575221239,
      "grad_norm": 7.625,
      "learning_rate": 7.282112731820789e-06,
      "loss": 1.0079,
      "num_input_tokens_seen": 16644928,
      "step": 30030
    },
    {
      "epoch": 526.9380530973451,
      "grad_norm": 8.25,
      "learning_rate": 7.275187923050447e-06,
      "loss": 1.1211,
      "num_input_tokens_seen": 16647584,
      "step": 30035
    },
    {
      "epoch": 527.0176991150443,
      "grad_norm": 7.0625,
      "learning_rate": 7.268265847668879e-06,
      "loss": 1.1245,
      "num_input_tokens_seen": 16649752,
      "step": 30040
    },
    {
      "epoch": 527.1061946902655,
      "grad_norm": 7.65625,
      "learning_rate": 7.261346506743538e-06,
      "loss": 0.8873,
      "num_input_tokens_seen": 16652424,
      "step": 30045
    },
    {
      "epoch": 527.1946902654868,
      "grad_norm": 10.4375,
      "learning_rate": 7.254429901341486e-06,
      "loss": 0.9301,
      "num_input_tokens_seen": 16655096,
      "step": 30050
    },
    {
      "epoch": 527.2831858407079,
      "grad_norm": 9.8125,
      "learning_rate": 7.247516032529356e-06,
      "loss": 1.2294,
      "num_input_tokens_seen": 16657864,
      "step": 30055
    },
    {
      "epoch": 527.3716814159292,
      "grad_norm": 8.75,
      "learning_rate": 7.240604901373338e-06,
      "loss": 0.8877,
      "num_input_tokens_seen": 16660648,
      "step": 30060
    },
    {
      "epoch": 527.4601769911504,
      "grad_norm": 12.875,
      "learning_rate": 7.233696508939223e-06,
      "loss": 1.1627,
      "num_input_tokens_seen": 16663704,
      "step": 30065
    },
    {
      "epoch": 527.5486725663717,
      "grad_norm": 7.71875,
      "learning_rate": 7.226790856292376e-06,
      "loss": 0.9399,
      "num_input_tokens_seen": 16666200,
      "step": 30070
    },
    {
      "epoch": 527.637168141593,
      "grad_norm": 8.5625,
      "learning_rate": 7.219887944497727e-06,
      "loss": 1.1311,
      "num_input_tokens_seen": 16669784,
      "step": 30075
    },
    {
      "epoch": 527.7256637168142,
      "grad_norm": 6.53125,
      "learning_rate": 7.2129877746198e-06,
      "loss": 0.989,
      "num_input_tokens_seen": 16673000,
      "step": 30080
    },
    {
      "epoch": 527.8141592920354,
      "grad_norm": 8.5625,
      "learning_rate": 7.20609034772268e-06,
      "loss": 1.144,
      "num_input_tokens_seen": 16675688,
      "step": 30085
    },
    {
      "epoch": 527.9026548672566,
      "grad_norm": 8.625,
      "learning_rate": 7.19919566487004e-06,
      "loss": 1.0559,
      "num_input_tokens_seen": 16677880,
      "step": 30090
    },
    {
      "epoch": 527.9911504424779,
      "grad_norm": 6.78125,
      "learning_rate": 7.192303727125132e-06,
      "loss": 1.2154,
      "num_input_tokens_seen": 16680824,
      "step": 30095
    },
    {
      "epoch": 528.070796460177,
      "grad_norm": 8.375,
      "learning_rate": 7.185414535550777e-06,
      "loss": 1.0992,
      "num_input_tokens_seen": 16682704,
      "step": 30100
    },
    {
      "epoch": 528.1592920353983,
      "grad_norm": 9.9375,
      "learning_rate": 7.178528091209363e-06,
      "loss": 1.0428,
      "num_input_tokens_seen": 16685360,
      "step": 30105
    },
    {
      "epoch": 528.2477876106195,
      "grad_norm": 9.875,
      "learning_rate": 7.171644395162888e-06,
      "loss": 1.0751,
      "num_input_tokens_seen": 16688416,
      "step": 30110
    },
    {
      "epoch": 528.3362831858407,
      "grad_norm": 9.75,
      "learning_rate": 7.164763448472881e-06,
      "loss": 1.0811,
      "num_input_tokens_seen": 16690896,
      "step": 30115
    },
    {
      "epoch": 528.4247787610619,
      "grad_norm": 7.59375,
      "learning_rate": 7.157885252200491e-06,
      "loss": 1.0495,
      "num_input_tokens_seen": 16694256,
      "step": 30120
    },
    {
      "epoch": 528.5132743362832,
      "grad_norm": 14.0,
      "learning_rate": 7.151009807406403e-06,
      "loss": 1.0507,
      "num_input_tokens_seen": 16696720,
      "step": 30125
    },
    {
      "epoch": 528.6017699115044,
      "grad_norm": 8.4375,
      "learning_rate": 7.144137115150909e-06,
      "loss": 0.9706,
      "num_input_tokens_seen": 16699296,
      "step": 30130
    },
    {
      "epoch": 528.6902654867257,
      "grad_norm": 7.4375,
      "learning_rate": 7.1372671764938725e-06,
      "loss": 1.1514,
      "num_input_tokens_seen": 16702016,
      "step": 30135
    },
    {
      "epoch": 528.7787610619469,
      "grad_norm": 6.9375,
      "learning_rate": 7.130399992494705e-06,
      "loss": 0.9227,
      "num_input_tokens_seen": 16704576,
      "step": 30140
    },
    {
      "epoch": 528.8672566371681,
      "grad_norm": 8.3125,
      "learning_rate": 7.123535564212419e-06,
      "loss": 1.0843,
      "num_input_tokens_seen": 16707664,
      "step": 30145
    },
    {
      "epoch": 528.9557522123894,
      "grad_norm": 8.0625,
      "learning_rate": 7.116673892705611e-06,
      "loss": 1.1414,
      "num_input_tokens_seen": 16710416,
      "step": 30150
    },
    {
      "epoch": 529.0353982300885,
      "grad_norm": 8.625,
      "learning_rate": 7.109814979032415e-06,
      "loss": 1.0665,
      "num_input_tokens_seen": 16713216,
      "step": 30155
    },
    {
      "epoch": 529.1238938053098,
      "grad_norm": 7.9375,
      "learning_rate": 7.102958824250577e-06,
      "loss": 1.1399,
      "num_input_tokens_seen": 16716160,
      "step": 30160
    },
    {
      "epoch": 529.212389380531,
      "grad_norm": 10.9375,
      "learning_rate": 7.096105429417393e-06,
      "loss": 1.0924,
      "num_input_tokens_seen": 16718288,
      "step": 30165
    },
    {
      "epoch": 529.3008849557522,
      "grad_norm": 7.40625,
      "learning_rate": 7.0892547955897506e-06,
      "loss": 1.1017,
      "num_input_tokens_seen": 16721344,
      "step": 30170
    },
    {
      "epoch": 529.3893805309734,
      "grad_norm": 9.0625,
      "learning_rate": 7.0824069238241e-06,
      "loss": 1.0282,
      "num_input_tokens_seen": 16724064,
      "step": 30175
    },
    {
      "epoch": 529.4778761061947,
      "grad_norm": 6.9375,
      "learning_rate": 7.075561815176462e-06,
      "loss": 0.998,
      "num_input_tokens_seen": 16726704,
      "step": 30180
    },
    {
      "epoch": 529.566371681416,
      "grad_norm": 11.875,
      "learning_rate": 7.068719470702445e-06,
      "loss": 1.0162,
      "num_input_tokens_seen": 16729536,
      "step": 30185
    },
    {
      "epoch": 529.6548672566372,
      "grad_norm": 9.6875,
      "learning_rate": 7.061879891457229e-06,
      "loss": 0.9811,
      "num_input_tokens_seen": 16732768,
      "step": 30190
    },
    {
      "epoch": 529.7433628318585,
      "grad_norm": 8.875,
      "learning_rate": 7.0550430784955515e-06,
      "loss": 0.8503,
      "num_input_tokens_seen": 16735920,
      "step": 30195
    },
    {
      "epoch": 529.8318584070796,
      "grad_norm": 10.0,
      "learning_rate": 7.048209032871752e-06,
      "loss": 1.0249,
      "num_input_tokens_seen": 16738416,
      "step": 30200
    },
    {
      "epoch": 529.8318584070796,
      "eval_loss": 1.0310944318771362,
      "eval_runtime": 0.9362,
      "eval_samples_per_second": 26.704,
      "eval_steps_per_second": 13.886,
      "num_input_tokens_seen": 16738416,
      "step": 30200
    },
    {
      "epoch": 529.9203539823009,
      "grad_norm": 7.65625,
      "learning_rate": 7.0413777556397055e-06,
      "loss": 1.1126,
      "num_input_tokens_seen": 16741248,
      "step": 30205
    },
    {
      "epoch": 530.0,
      "grad_norm": 7.90625,
      "learning_rate": 7.0345492478528925e-06,
      "loss": 1.0849,
      "num_input_tokens_seen": 16743232,
      "step": 30210
    },
    {
      "epoch": 530.0884955752213,
      "grad_norm": 10.3125,
      "learning_rate": 7.02772351056436e-06,
      "loss": 1.0556,
      "num_input_tokens_seen": 16746000,
      "step": 30215
    },
    {
      "epoch": 530.1769911504425,
      "grad_norm": 7.9375,
      "learning_rate": 7.020900544826709e-06,
      "loss": 0.9644,
      "num_input_tokens_seen": 16749072,
      "step": 30220
    },
    {
      "epoch": 530.2654867256637,
      "grad_norm": 7.6875,
      "learning_rate": 7.014080351692134e-06,
      "loss": 1.0665,
      "num_input_tokens_seen": 16752016,
      "step": 30225
    },
    {
      "epoch": 530.3539823008849,
      "grad_norm": 8.3125,
      "learning_rate": 7.0072629322124024e-06,
      "loss": 1.0599,
      "num_input_tokens_seen": 16754720,
      "step": 30230
    },
    {
      "epoch": 530.4424778761062,
      "grad_norm": 7.4375,
      "learning_rate": 7.000448287438827e-06,
      "loss": 0.8981,
      "num_input_tokens_seen": 16757552,
      "step": 30235
    },
    {
      "epoch": 530.5309734513274,
      "grad_norm": 8.5625,
      "learning_rate": 6.993636418422331e-06,
      "loss": 0.846,
      "num_input_tokens_seen": 16760432,
      "step": 30240
    },
    {
      "epoch": 530.6194690265487,
      "grad_norm": 8.6875,
      "learning_rate": 6.986827326213383e-06,
      "loss": 1.1961,
      "num_input_tokens_seen": 16763216,
      "step": 30245
    },
    {
      "epoch": 530.70796460177,
      "grad_norm": 8.6875,
      "learning_rate": 6.9800210118620205e-06,
      "loss": 1.0361,
      "num_input_tokens_seen": 16765920,
      "step": 30250
    },
    {
      "epoch": 530.7964601769911,
      "grad_norm": 8.6875,
      "learning_rate": 6.973217476417876e-06,
      "loss": 1.0828,
      "num_input_tokens_seen": 16768544,
      "step": 30255
    },
    {
      "epoch": 530.8849557522124,
      "grad_norm": 14.375,
      "learning_rate": 6.96641672093013e-06,
      "loss": 1.0813,
      "num_input_tokens_seen": 16771664,
      "step": 30260
    },
    {
      "epoch": 530.9734513274336,
      "grad_norm": 12.25,
      "learning_rate": 6.95961874644755e-06,
      "loss": 1.3435,
      "num_input_tokens_seen": 16774176,
      "step": 30265
    },
    {
      "epoch": 531.0530973451328,
      "grad_norm": 9.5625,
      "learning_rate": 6.952823554018476e-06,
      "loss": 0.9209,
      "num_input_tokens_seen": 16776376,
      "step": 30270
    },
    {
      "epoch": 531.141592920354,
      "grad_norm": 6.59375,
      "learning_rate": 6.946031144690798e-06,
      "loss": 0.997,
      "num_input_tokens_seen": 16779160,
      "step": 30275
    },
    {
      "epoch": 531.2300884955753,
      "grad_norm": 8.4375,
      "learning_rate": 6.939241519512005e-06,
      "loss": 1.1793,
      "num_input_tokens_seen": 16781928,
      "step": 30280
    },
    {
      "epoch": 531.3185840707964,
      "grad_norm": 8.3125,
      "learning_rate": 6.932454679529129e-06,
      "loss": 1.0524,
      "num_input_tokens_seen": 16785960,
      "step": 30285
    },
    {
      "epoch": 531.4070796460177,
      "grad_norm": 6.28125,
      "learning_rate": 6.925670625788791e-06,
      "loss": 1.024,
      "num_input_tokens_seen": 16788264,
      "step": 30290
    },
    {
      "epoch": 531.4955752212389,
      "grad_norm": 7.84375,
      "learning_rate": 6.918889359337186e-06,
      "loss": 1.0737,
      "num_input_tokens_seen": 16791016,
      "step": 30295
    },
    {
      "epoch": 531.5840707964602,
      "grad_norm": 7.125,
      "learning_rate": 6.912110881220058e-06,
      "loss": 1.0858,
      "num_input_tokens_seen": 16793640,
      "step": 30300
    },
    {
      "epoch": 531.6725663716815,
      "grad_norm": 10.1875,
      "learning_rate": 6.905335192482735e-06,
      "loss": 1.0866,
      "num_input_tokens_seen": 16796056,
      "step": 30305
    },
    {
      "epoch": 531.7610619469026,
      "grad_norm": 7.0625,
      "learning_rate": 6.8985622941701275e-06,
      "loss": 1.0548,
      "num_input_tokens_seen": 16798648,
      "step": 30310
    },
    {
      "epoch": 531.8495575221239,
      "grad_norm": 8.625,
      "learning_rate": 6.89179218732669e-06,
      "loss": 1.004,
      "num_input_tokens_seen": 16801224,
      "step": 30315
    },
    {
      "epoch": 531.9380530973451,
      "grad_norm": 5.78125,
      "learning_rate": 6.8850248729964595e-06,
      "loss": 1.0283,
      "num_input_tokens_seen": 16804552,
      "step": 30320
    },
    {
      "epoch": 532.0176991150443,
      "grad_norm": 9.0625,
      "learning_rate": 6.8782603522230314e-06,
      "loss": 0.9162,
      "num_input_tokens_seen": 16807320,
      "step": 30325
    },
    {
      "epoch": 532.1061946902655,
      "grad_norm": 10.0625,
      "learning_rate": 6.871498626049591e-06,
      "loss": 1.1261,
      "num_input_tokens_seen": 16809752,
      "step": 30330
    },
    {
      "epoch": 532.1946902654868,
      "grad_norm": 11.875,
      "learning_rate": 6.8647396955188875e-06,
      "loss": 0.976,
      "num_input_tokens_seen": 16812584,
      "step": 30335
    },
    {
      "epoch": 532.2831858407079,
      "grad_norm": 14.0,
      "learning_rate": 6.857983561673218e-06,
      "loss": 0.9086,
      "num_input_tokens_seen": 16815704,
      "step": 30340
    },
    {
      "epoch": 532.3716814159292,
      "grad_norm": 7.90625,
      "learning_rate": 6.851230225554467e-06,
      "loss": 1.0304,
      "num_input_tokens_seen": 16818552,
      "step": 30345
    },
    {
      "epoch": 532.4601769911504,
      "grad_norm": 8.9375,
      "learning_rate": 6.8444796882040946e-06,
      "loss": 0.9391,
      "num_input_tokens_seen": 16821528,
      "step": 30350
    },
    {
      "epoch": 532.5486725663717,
      "grad_norm": 7.59375,
      "learning_rate": 6.837731950663106e-06,
      "loss": 0.9382,
      "num_input_tokens_seen": 16824392,
      "step": 30355
    },
    {
      "epoch": 532.637168141593,
      "grad_norm": 9.75,
      "learning_rate": 6.830987013972098e-06,
      "loss": 1.0412,
      "num_input_tokens_seen": 16826856,
      "step": 30360
    },
    {
      "epoch": 532.7256637168142,
      "grad_norm": 6.21875,
      "learning_rate": 6.82424487917121e-06,
      "loss": 1.032,
      "num_input_tokens_seen": 16829736,
      "step": 30365
    },
    {
      "epoch": 532.8141592920354,
      "grad_norm": 9.9375,
      "learning_rate": 6.8175055473001735e-06,
      "loss": 1.2892,
      "num_input_tokens_seen": 16832728,
      "step": 30370
    },
    {
      "epoch": 532.9026548672566,
      "grad_norm": 7.125,
      "learning_rate": 6.8107690193982855e-06,
      "loss": 1.1871,
      "num_input_tokens_seen": 16835096,
      "step": 30375
    },
    {
      "epoch": 532.9911504424779,
      "grad_norm": 6.84375,
      "learning_rate": 6.804035296504385e-06,
      "loss": 1.035,
      "num_input_tokens_seen": 16837848,
      "step": 30380
    },
    {
      "epoch": 533.070796460177,
      "grad_norm": 8.625,
      "learning_rate": 6.797304379656916e-06,
      "loss": 1.1525,
      "num_input_tokens_seen": 16839952,
      "step": 30385
    },
    {
      "epoch": 533.1592920353983,
      "grad_norm": 6.9375,
      "learning_rate": 6.790576269893861e-06,
      "loss": 1.128,
      "num_input_tokens_seen": 16842272,
      "step": 30390
    },
    {
      "epoch": 533.2477876106195,
      "grad_norm": 12.5625,
      "learning_rate": 6.783850968252772e-06,
      "loss": 0.9607,
      "num_input_tokens_seen": 16845504,
      "step": 30395
    },
    {
      "epoch": 533.3362831858407,
      "grad_norm": 6.90625,
      "learning_rate": 6.777128475770789e-06,
      "loss": 1.0122,
      "num_input_tokens_seen": 16848080,
      "step": 30400
    },
    {
      "epoch": 533.3362831858407,
      "eval_loss": 1.0310944318771362,
      "eval_runtime": 0.9398,
      "eval_samples_per_second": 26.602,
      "eval_steps_per_second": 13.833,
      "num_input_tokens_seen": 16848080,
      "step": 30400
    },
    {
      "epoch": 533.4247787610619,
      "grad_norm": 8.625,
      "learning_rate": 6.77040879348459e-06,
      "loss": 0.9997,
      "num_input_tokens_seen": 16851248,
      "step": 30405
    },
    {
      "epoch": 533.5132743362832,
      "grad_norm": 10.5625,
      "learning_rate": 6.763691922430443e-06,
      "loss": 0.9812,
      "num_input_tokens_seen": 16853952,
      "step": 30410
    },
    {
      "epoch": 533.6017699115044,
      "grad_norm": 9.5,
      "learning_rate": 6.756977863644178e-06,
      "loss": 1.0662,
      "num_input_tokens_seen": 16856848,
      "step": 30415
    },
    {
      "epoch": 533.6902654867257,
      "grad_norm": 9.3125,
      "learning_rate": 6.7502666181611804e-06,
      "loss": 1.0824,
      "num_input_tokens_seen": 16859280,
      "step": 30420
    },
    {
      "epoch": 533.7787610619469,
      "grad_norm": 6.25,
      "learning_rate": 6.743558187016405e-06,
      "loss": 1.0073,
      "num_input_tokens_seen": 16862336,
      "step": 30425
    },
    {
      "epoch": 533.8672566371681,
      "grad_norm": 7.875,
      "learning_rate": 6.7368525712443925e-06,
      "loss": 0.9743,
      "num_input_tokens_seen": 16865648,
      "step": 30430
    },
    {
      "epoch": 533.9557522123894,
      "grad_norm": 7.15625,
      "learning_rate": 6.7301497718792155e-06,
      "loss": 1.1082,
      "num_input_tokens_seen": 16868848,
      "step": 30435
    },
    {
      "epoch": 534.0353982300885,
      "grad_norm": 7.90625,
      "learning_rate": 6.723449789954544e-06,
      "loss": 1.2228,
      "num_input_tokens_seen": 16871160,
      "step": 30440
    },
    {
      "epoch": 534.1238938053098,
      "grad_norm": 8.5625,
      "learning_rate": 6.716752626503586e-06,
      "loss": 1.0956,
      "num_input_tokens_seen": 16873608,
      "step": 30445
    },
    {
      "epoch": 534.212389380531,
      "grad_norm": 8.375,
      "learning_rate": 6.710058282559131e-06,
      "loss": 1.1876,
      "num_input_tokens_seen": 16876584,
      "step": 30450
    },
    {
      "epoch": 534.3008849557522,
      "grad_norm": 8.4375,
      "learning_rate": 6.703366759153545e-06,
      "loss": 0.8841,
      "num_input_tokens_seen": 16879320,
      "step": 30455
    },
    {
      "epoch": 534.3893805309734,
      "grad_norm": 11.625,
      "learning_rate": 6.6966780573187335e-06,
      "loss": 1.2088,
      "num_input_tokens_seen": 16881656,
      "step": 30460
    },
    {
      "epoch": 534.4778761061947,
      "grad_norm": 7.34375,
      "learning_rate": 6.689992178086174e-06,
      "loss": 1.0327,
      "num_input_tokens_seen": 16884376,
      "step": 30465
    },
    {
      "epoch": 534.566371681416,
      "grad_norm": 8.1875,
      "learning_rate": 6.683309122486925e-06,
      "loss": 1.0399,
      "num_input_tokens_seen": 16887544,
      "step": 30470
    },
    {
      "epoch": 534.6548672566372,
      "grad_norm": 5.625,
      "learning_rate": 6.676628891551584e-06,
      "loss": 0.8502,
      "num_input_tokens_seen": 16890792,
      "step": 30475
    },
    {
      "epoch": 534.7433628318585,
      "grad_norm": 9.75,
      "learning_rate": 6.6699514863103385e-06,
      "loss": 1.3165,
      "num_input_tokens_seen": 16893592,
      "step": 30480
    },
    {
      "epoch": 534.8318584070796,
      "grad_norm": 8.3125,
      "learning_rate": 6.663276907792921e-06,
      "loss": 0.9712,
      "num_input_tokens_seen": 16896248,
      "step": 30485
    },
    {
      "epoch": 534.9203539823009,
      "grad_norm": 6.28125,
      "learning_rate": 6.656605157028634e-06,
      "loss": 1.0062,
      "num_input_tokens_seen": 16899128,
      "step": 30490
    },
    {
      "epoch": 535.0,
      "grad_norm": 11.0625,
      "learning_rate": 6.649936235046358e-06,
      "loss": 0.8731,
      "num_input_tokens_seen": 16901656,
      "step": 30495
    },
    {
      "epoch": 535.0884955752213,
      "grad_norm": 6.78125,
      "learning_rate": 6.643270142874508e-06,
      "loss": 0.9383,
      "num_input_tokens_seen": 16904600,
      "step": 30500
    },
    {
      "epoch": 535.1769911504425,
      "grad_norm": 8.75,
      "learning_rate": 6.636606881541094e-06,
      "loss": 1.2087,
      "num_input_tokens_seen": 16907128,
      "step": 30505
    },
    {
      "epoch": 535.2654867256637,
      "grad_norm": 8.875,
      "learning_rate": 6.629946452073662e-06,
      "loss": 1.0384,
      "num_input_tokens_seen": 16909592,
      "step": 30510
    },
    {
      "epoch": 535.3539823008849,
      "grad_norm": 8.75,
      "learning_rate": 6.6232888554993375e-06,
      "loss": 1.0935,
      "num_input_tokens_seen": 16912472,
      "step": 30515
    },
    {
      "epoch": 535.4424778761062,
      "grad_norm": 9.6875,
      "learning_rate": 6.616634092844817e-06,
      "loss": 0.9466,
      "num_input_tokens_seen": 16915464,
      "step": 30520
    },
    {
      "epoch": 535.5309734513274,
      "grad_norm": 6.34375,
      "learning_rate": 6.609982165136331e-06,
      "loss": 1.0463,
      "num_input_tokens_seen": 16918056,
      "step": 30525
    },
    {
      "epoch": 535.6194690265487,
      "grad_norm": 6.96875,
      "learning_rate": 6.603333073399706e-06,
      "loss": 0.9868,
      "num_input_tokens_seen": 16920952,
      "step": 30530
    },
    {
      "epoch": 535.70796460177,
      "grad_norm": 13.4375,
      "learning_rate": 6.596686818660308e-06,
      "loss": 1.0563,
      "num_input_tokens_seen": 16924024,
      "step": 30535
    },
    {
      "epoch": 535.7964601769911,
      "grad_norm": 9.0625,
      "learning_rate": 6.590043401943066e-06,
      "loss": 1.0929,
      "num_input_tokens_seen": 16926936,
      "step": 30540
    },
    {
      "epoch": 535.8849557522124,
      "grad_norm": 11.6875,
      "learning_rate": 6.583402824272494e-06,
      "loss": 1.0388,
      "num_input_tokens_seen": 16929656,
      "step": 30545
    },
    {
      "epoch": 535.9734513274336,
      "grad_norm": 6.1875,
      "learning_rate": 6.576765086672634e-06,
      "loss": 1.0684,
      "num_input_tokens_seen": 16932408,
      "step": 30550
    },
    {
      "epoch": 536.0530973451328,
      "grad_norm": 7.9375,
      "learning_rate": 6.57013019016712e-06,
      "loss": 1.0923,
      "num_input_tokens_seen": 16934856,
      "step": 30555
    },
    {
      "epoch": 536.141592920354,
      "grad_norm": 8.125,
      "learning_rate": 6.563498135779142e-06,
      "loss": 0.8367,
      "num_input_tokens_seen": 16937416,
      "step": 30560
    },
    {
      "epoch": 536.2300884955753,
      "grad_norm": 6.875,
      "learning_rate": 6.556868924531431e-06,
      "loss": 0.9823,
      "num_input_tokens_seen": 16940488,
      "step": 30565
    },
    {
      "epoch": 536.3185840707964,
      "grad_norm": 7.59375,
      "learning_rate": 6.550242557446304e-06,
      "loss": 1.1421,
      "num_input_tokens_seen": 16943272,
      "step": 30570
    },
    {
      "epoch": 536.4070796460177,
      "grad_norm": 9.4375,
      "learning_rate": 6.543619035545634e-06,
      "loss": 0.909,
      "num_input_tokens_seen": 16945992,
      "step": 30575
    },
    {
      "epoch": 536.4955752212389,
      "grad_norm": 9.25,
      "learning_rate": 6.53699835985084e-06,
      "loss": 0.995,
      "num_input_tokens_seen": 16949160,
      "step": 30580
    },
    {
      "epoch": 536.5840707964602,
      "grad_norm": 10.3125,
      "learning_rate": 6.530380531382927e-06,
      "loss": 1.1935,
      "num_input_tokens_seen": 16952072,
      "step": 30585
    },
    {
      "epoch": 536.6725663716815,
      "grad_norm": 6.71875,
      "learning_rate": 6.523765551162433e-06,
      "loss": 1.0401,
      "num_input_tokens_seen": 16954776,
      "step": 30590
    },
    {
      "epoch": 536.7610619469026,
      "grad_norm": 7.15625,
      "learning_rate": 6.517153420209476e-06,
      "loss": 0.9779,
      "num_input_tokens_seen": 16957624,
      "step": 30595
    },
    {
      "epoch": 536.8495575221239,
      "grad_norm": 10.375,
      "learning_rate": 6.510544139543739e-06,
      "loss": 1.2131,
      "num_input_tokens_seen": 16960312,
      "step": 30600
    },
    {
      "epoch": 536.8495575221239,
      "eval_loss": 1.0310944318771362,
      "eval_runtime": 0.9371,
      "eval_samples_per_second": 26.679,
      "eval_steps_per_second": 13.873,
      "num_input_tokens_seen": 16960312,
      "step": 30600
    },
    {
      "epoch": 536.9380530973451,
      "grad_norm": 7.84375,
      "learning_rate": 6.503937710184452e-06,
      "loss": 1.2195,
      "num_input_tokens_seen": 16963144,
      "step": 30605
    },
    {
      "epoch": 537.0176991150443,
      "grad_norm": 8.125,
      "learning_rate": 6.4973341331503954e-06,
      "loss": 1.0574,
      "num_input_tokens_seen": 16965568,
      "step": 30610
    },
    {
      "epoch": 537.1061946902655,
      "grad_norm": 7.4375,
      "learning_rate": 6.490733409459942e-06,
      "loss": 1.1621,
      "num_input_tokens_seen": 16968432,
      "step": 30615
    },
    {
      "epoch": 537.1946902654868,
      "grad_norm": 6.78125,
      "learning_rate": 6.484135540130995e-06,
      "loss": 1.0799,
      "num_input_tokens_seen": 16970672,
      "step": 30620
    },
    {
      "epoch": 537.2831858407079,
      "grad_norm": 9.6875,
      "learning_rate": 6.4775405261810364e-06,
      "loss": 1.0293,
      "num_input_tokens_seen": 16973712,
      "step": 30625
    },
    {
      "epoch": 537.3716814159292,
      "grad_norm": 11.5625,
      "learning_rate": 6.470948368627092e-06,
      "loss": 1.0107,
      "num_input_tokens_seen": 16976528,
      "step": 30630
    },
    {
      "epoch": 537.4601769911504,
      "grad_norm": 6.03125,
      "learning_rate": 6.464359068485756e-06,
      "loss": 0.9998,
      "num_input_tokens_seen": 16979184,
      "step": 30635
    },
    {
      "epoch": 537.5486725663717,
      "grad_norm": 8.75,
      "learning_rate": 6.457772626773195e-06,
      "loss": 1.0986,
      "num_input_tokens_seen": 16981520,
      "step": 30640
    },
    {
      "epoch": 537.637168141593,
      "grad_norm": 8.4375,
      "learning_rate": 6.451189044505104e-06,
      "loss": 1.1163,
      "num_input_tokens_seen": 16984336,
      "step": 30645
    },
    {
      "epoch": 537.7256637168142,
      "grad_norm": 8.5,
      "learning_rate": 6.44460832269676e-06,
      "loss": 1.0541,
      "num_input_tokens_seen": 16986896,
      "step": 30650
    },
    {
      "epoch": 537.8141592920354,
      "grad_norm": 7.96875,
      "learning_rate": 6.438030462363001e-06,
      "loss": 0.9757,
      "num_input_tokens_seen": 16989680,
      "step": 30655
    },
    {
      "epoch": 537.9026548672566,
      "grad_norm": 7.40625,
      "learning_rate": 6.431455464518205e-06,
      "loss": 0.9728,
      "num_input_tokens_seen": 16992784,
      "step": 30660
    },
    {
      "epoch": 537.9911504424779,
      "grad_norm": 7.21875,
      "learning_rate": 6.424883330176326e-06,
      "loss": 1.0218,
      "num_input_tokens_seen": 16995632,
      "step": 30665
    },
    {
      "epoch": 538.070796460177,
      "grad_norm": 7.75,
      "learning_rate": 6.418314060350864e-06,
      "loss": 1.0982,
      "num_input_tokens_seen": 16997944,
      "step": 30670
    },
    {
      "epoch": 538.1592920353983,
      "grad_norm": 8.0,
      "learning_rate": 6.4117476560548895e-06,
      "loss": 1.123,
      "num_input_tokens_seen": 17000472,
      "step": 30675
    },
    {
      "epoch": 538.2477876106195,
      "grad_norm": 9.5625,
      "learning_rate": 6.405184118301016e-06,
      "loss": 0.941,
      "num_input_tokens_seen": 17003144,
      "step": 30680
    },
    {
      "epoch": 538.3362831858407,
      "grad_norm": 8.375,
      "learning_rate": 6.398623448101434e-06,
      "loss": 1.0587,
      "num_input_tokens_seen": 17005368,
      "step": 30685
    },
    {
      "epoch": 538.4247787610619,
      "grad_norm": 11.125,
      "learning_rate": 6.392065646467871e-06,
      "loss": 1.2348,
      "num_input_tokens_seen": 17007480,
      "step": 30690
    },
    {
      "epoch": 538.5132743362832,
      "grad_norm": 7.34375,
      "learning_rate": 6.385510714411632e-06,
      "loss": 0.9916,
      "num_input_tokens_seen": 17010552,
      "step": 30695
    },
    {
      "epoch": 538.6017699115044,
      "grad_norm": 7.3125,
      "learning_rate": 6.378958652943559e-06,
      "loss": 0.9849,
      "num_input_tokens_seen": 17013720,
      "step": 30700
    },
    {
      "epoch": 538.6902654867257,
      "grad_norm": 9.25,
      "learning_rate": 6.3724094630740776e-06,
      "loss": 0.9914,
      "num_input_tokens_seen": 17016536,
      "step": 30705
    },
    {
      "epoch": 538.7787610619469,
      "grad_norm": 14.5,
      "learning_rate": 6.365863145813136e-06,
      "loss": 1.064,
      "num_input_tokens_seen": 17019432,
      "step": 30710
    },
    {
      "epoch": 538.8672566371681,
      "grad_norm": 8.75,
      "learning_rate": 6.359319702170269e-06,
      "loss": 0.9772,
      "num_input_tokens_seen": 17022088,
      "step": 30715
    },
    {
      "epoch": 538.9557522123894,
      "grad_norm": 9.75,
      "learning_rate": 6.352779133154566e-06,
      "loss": 1.2397,
      "num_input_tokens_seen": 17025400,
      "step": 30720
    },
    {
      "epoch": 539.0353982300885,
      "grad_norm": 8.75,
      "learning_rate": 6.346241439774648e-06,
      "loss": 0.8141,
      "num_input_tokens_seen": 17028208,
      "step": 30725
    },
    {
      "epoch": 539.1238938053098,
      "grad_norm": 10.0625,
      "learning_rate": 6.339706623038716e-06,
      "loss": 1.1433,
      "num_input_tokens_seen": 17031008,
      "step": 30730
    },
    {
      "epoch": 539.212389380531,
      "grad_norm": 7.875,
      "learning_rate": 6.333174683954532e-06,
      "loss": 0.9645,
      "num_input_tokens_seen": 17033824,
      "step": 30735
    },
    {
      "epoch": 539.3008849557522,
      "grad_norm": 7.28125,
      "learning_rate": 6.326645623529387e-06,
      "loss": 1.0517,
      "num_input_tokens_seen": 17036752,
      "step": 30740
    },
    {
      "epoch": 539.3893805309734,
      "grad_norm": 9.625,
      "learning_rate": 6.320119442770156e-06,
      "loss": 1.0692,
      "num_input_tokens_seen": 17039376,
      "step": 30745
    },
    {
      "epoch": 539.4778761061947,
      "grad_norm": 7.375,
      "learning_rate": 6.313596142683254e-06,
      "loss": 1.2272,
      "num_input_tokens_seen": 17042432,
      "step": 30750
    },
    {
      "epoch": 539.566371681416,
      "grad_norm": 6.28125,
      "learning_rate": 6.307075724274647e-06,
      "loss": 1.1251,
      "num_input_tokens_seen": 17045056,
      "step": 30755
    },
    {
      "epoch": 539.6548672566372,
      "grad_norm": 8.5,
      "learning_rate": 6.300558188549882e-06,
      "loss": 1.1117,
      "num_input_tokens_seen": 17047920,
      "step": 30760
    },
    {
      "epoch": 539.7433628318585,
      "grad_norm": 10.4375,
      "learning_rate": 6.29404353651403e-06,
      "loss": 0.9906,
      "num_input_tokens_seen": 17050336,
      "step": 30765
    },
    {
      "epoch": 539.8318584070796,
      "grad_norm": 7.0,
      "learning_rate": 6.287531769171737e-06,
      "loss": 0.9715,
      "num_input_tokens_seen": 17053440,
      "step": 30770
    },
    {
      "epoch": 539.9203539823009,
      "grad_norm": 8.3125,
      "learning_rate": 6.2810228875272045e-06,
      "loss": 1.1151,
      "num_input_tokens_seen": 17055792,
      "step": 30775
    },
    {
      "epoch": 540.0,
      "grad_norm": 15.5625,
      "learning_rate": 6.274516892584179e-06,
      "loss": 0.916,
      "num_input_tokens_seen": 17058016,
      "step": 30780
    },
    {
      "epoch": 540.0884955752213,
      "grad_norm": 8.4375,
      "learning_rate": 6.268013785345969e-06,
      "loss": 1.1481,
      "num_input_tokens_seen": 17060944,
      "step": 30785
    },
    {
      "epoch": 540.1769911504425,
      "grad_norm": 7.8125,
      "learning_rate": 6.26151356681543e-06,
      "loss": 0.8652,
      "num_input_tokens_seen": 17063952,
      "step": 30790
    },
    {
      "epoch": 540.2654867256637,
      "grad_norm": 8.4375,
      "learning_rate": 6.255016237994981e-06,
      "loss": 0.9102,
      "num_input_tokens_seen": 17067056,
      "step": 30795
    },
    {
      "epoch": 540.3539823008849,
      "grad_norm": 10.25,
      "learning_rate": 6.248521799886603e-06,
      "loss": 0.9375,
      "num_input_tokens_seen": 17069536,
      "step": 30800
    },
    {
      "epoch": 540.3539823008849,
      "eval_loss": 1.0310944318771362,
      "eval_runtime": 0.9409,
      "eval_samples_per_second": 26.571,
      "eval_steps_per_second": 13.817,
      "num_input_tokens_seen": 17069536,
      "step": 30800
    },
    {
      "epoch": 540.4424778761062,
      "grad_norm": 7.9375,
      "learning_rate": 6.242030253491798e-06,
      "loss": 1.1156,
      "num_input_tokens_seen": 17072192,
      "step": 30805
    },
    {
      "epoch": 540.5309734513274,
      "grad_norm": 10.5,
      "learning_rate": 6.235541599811656e-06,
      "loss": 0.9366,
      "num_input_tokens_seen": 17074896,
      "step": 30810
    },
    {
      "epoch": 540.6194690265487,
      "grad_norm": 10.25,
      "learning_rate": 6.229055839846814e-06,
      "loss": 1.1082,
      "num_input_tokens_seen": 17077920,
      "step": 30815
    },
    {
      "epoch": 540.70796460177,
      "grad_norm": 13.125,
      "learning_rate": 6.222572974597455e-06,
      "loss": 1.2201,
      "num_input_tokens_seen": 17080752,
      "step": 30820
    },
    {
      "epoch": 540.7964601769911,
      "grad_norm": 7.03125,
      "learning_rate": 6.216093005063306e-06,
      "loss": 1.0358,
      "num_input_tokens_seen": 17083488,
      "step": 30825
    },
    {
      "epoch": 540.8849557522124,
      "grad_norm": 10.125,
      "learning_rate": 6.209615932243678e-06,
      "loss": 1.1733,
      "num_input_tokens_seen": 17086080,
      "step": 30830
    },
    {
      "epoch": 540.9734513274336,
      "grad_norm": 8.0,
      "learning_rate": 6.203141757137399e-06,
      "loss": 1.0411,
      "num_input_tokens_seen": 17089104,
      "step": 30835
    },
    {
      "epoch": 541.0530973451328,
      "grad_norm": 9.0625,
      "learning_rate": 6.196670480742886e-06,
      "loss": 1.1098,
      "num_input_tokens_seen": 17092256,
      "step": 30840
    },
    {
      "epoch": 541.141592920354,
      "grad_norm": 7.6875,
      "learning_rate": 6.190202104058074e-06,
      "loss": 0.9702,
      "num_input_tokens_seen": 17094928,
      "step": 30845
    },
    {
      "epoch": 541.2300884955753,
      "grad_norm": 8.8125,
      "learning_rate": 6.183736628080475e-06,
      "loss": 1.1538,
      "num_input_tokens_seen": 17097776,
      "step": 30850
    },
    {
      "epoch": 541.3185840707964,
      "grad_norm": 7.875,
      "learning_rate": 6.177274053807155e-06,
      "loss": 0.9666,
      "num_input_tokens_seen": 17100288,
      "step": 30855
    },
    {
      "epoch": 541.4070796460177,
      "grad_norm": 7.78125,
      "learning_rate": 6.170814382234713e-06,
      "loss": 1.0589,
      "num_input_tokens_seen": 17103216,
      "step": 30860
    },
    {
      "epoch": 541.4955752212389,
      "grad_norm": 8.1875,
      "learning_rate": 6.16435761435932e-06,
      "loss": 1.097,
      "num_input_tokens_seen": 17105984,
      "step": 30865
    },
    {
      "epoch": 541.5840707964602,
      "grad_norm": 10.0625,
      "learning_rate": 6.157903751176681e-06,
      "loss": 1.078,
      "num_input_tokens_seen": 17108288,
      "step": 30870
    },
    {
      "epoch": 541.6725663716815,
      "grad_norm": 9.5,
      "learning_rate": 6.151452793682066e-06,
      "loss": 0.9889,
      "num_input_tokens_seen": 17110976,
      "step": 30875
    },
    {
      "epoch": 541.7610619469026,
      "grad_norm": 11.0,
      "learning_rate": 6.145004742870305e-06,
      "loss": 1.0069,
      "num_input_tokens_seen": 17113904,
      "step": 30880
    },
    {
      "epoch": 541.8495575221239,
      "grad_norm": 9.0,
      "learning_rate": 6.138559599735752e-06,
      "loss": 1.1769,
      "num_input_tokens_seen": 17116528,
      "step": 30885
    },
    {
      "epoch": 541.9380530973451,
      "grad_norm": 7.46875,
      "learning_rate": 6.132117365272344e-06,
      "loss": 1.0893,
      "num_input_tokens_seen": 17119472,
      "step": 30890
    },
    {
      "epoch": 542.0176991150443,
      "grad_norm": 6.4375,
      "learning_rate": 6.125678040473545e-06,
      "loss": 0.8764,
      "num_input_tokens_seen": 17121816,
      "step": 30895
    },
    {
      "epoch": 542.1061946902655,
      "grad_norm": 6.53125,
      "learning_rate": 6.1192416263323755e-06,
      "loss": 1.1232,
      "num_input_tokens_seen": 17124424,
      "step": 30900
    },
    {
      "epoch": 542.1946902654868,
      "grad_norm": 8.375,
      "learning_rate": 6.112808123841424e-06,
      "loss": 1.1616,
      "num_input_tokens_seen": 17127016,
      "step": 30905
    },
    {
      "epoch": 542.2831858407079,
      "grad_norm": 8.125,
      "learning_rate": 6.106377533992805e-06,
      "loss": 1.0217,
      "num_input_tokens_seen": 17129384,
      "step": 30910
    },
    {
      "epoch": 542.3716814159292,
      "grad_norm": 14.0625,
      "learning_rate": 6.099949857778204e-06,
      "loss": 1.2654,
      "num_input_tokens_seen": 17131784,
      "step": 30915
    },
    {
      "epoch": 542.4601769911504,
      "grad_norm": 7.28125,
      "learning_rate": 6.093525096188852e-06,
      "loss": 0.9972,
      "num_input_tokens_seen": 17134984,
      "step": 30920
    },
    {
      "epoch": 542.5486725663717,
      "grad_norm": 7.90625,
      "learning_rate": 6.087103250215518e-06,
      "loss": 1.0519,
      "num_input_tokens_seen": 17138312,
      "step": 30925
    },
    {
      "epoch": 542.637168141593,
      "grad_norm": 12.8125,
      "learning_rate": 6.080684320848537e-06,
      "loss": 1.051,
      "num_input_tokens_seen": 17141176,
      "step": 30930
    },
    {
      "epoch": 542.7256637168142,
      "grad_norm": 10.0625,
      "learning_rate": 6.074268309077794e-06,
      "loss": 1.0743,
      "num_input_tokens_seen": 17143976,
      "step": 30935
    },
    {
      "epoch": 542.8141592920354,
      "grad_norm": 8.1875,
      "learning_rate": 6.067855215892709e-06,
      "loss": 1.0348,
      "num_input_tokens_seen": 17146664,
      "step": 30940
    },
    {
      "epoch": 542.9026548672566,
      "grad_norm": 5.84375,
      "learning_rate": 6.061445042282271e-06,
      "loss": 0.8601,
      "num_input_tokens_seen": 17149832,
      "step": 30945
    },
    {
      "epoch": 542.9911504424779,
      "grad_norm": 7.1875,
      "learning_rate": 6.055037789234999e-06,
      "loss": 0.9959,
      "num_input_tokens_seen": 17152792,
      "step": 30950
    },
    {
      "epoch": 543.070796460177,
      "grad_norm": 7.59375,
      "learning_rate": 6.048633457738975e-06,
      "loss": 1.13,
      "num_input_tokens_seen": 17155384,
      "step": 30955
    },
    {
      "epoch": 543.1592920353983,
      "grad_norm": 9.25,
      "learning_rate": 6.042232048781837e-06,
      "loss": 1.1254,
      "num_input_tokens_seen": 17158376,
      "step": 30960
    },
    {
      "epoch": 543.2477876106195,
      "grad_norm": 8.6875,
      "learning_rate": 6.035833563350757e-06,
      "loss": 0.9202,
      "num_input_tokens_seen": 17161096,
      "step": 30965
    },
    {
      "epoch": 543.3362831858407,
      "grad_norm": 7.5625,
      "learning_rate": 6.0294380024324525e-06,
      "loss": 1.0011,
      "num_input_tokens_seen": 17163752,
      "step": 30970
    },
    {
      "epoch": 543.4247787610619,
      "grad_norm": 9.9375,
      "learning_rate": 6.023045367013213e-06,
      "loss": 1.0812,
      "num_input_tokens_seen": 17166888,
      "step": 30975
    },
    {
      "epoch": 543.5132743362832,
      "grad_norm": 8.125,
      "learning_rate": 6.016655658078851e-06,
      "loss": 1.0433,
      "num_input_tokens_seen": 17169976,
      "step": 30980
    },
    {
      "epoch": 543.6017699115044,
      "grad_norm": 7.59375,
      "learning_rate": 6.010268876614753e-06,
      "loss": 0.8625,
      "num_input_tokens_seen": 17172872,
      "step": 30985
    },
    {
      "epoch": 543.6902654867257,
      "grad_norm": 10.3125,
      "learning_rate": 6.0038850236058266e-06,
      "loss": 1.0265,
      "num_input_tokens_seen": 17175304,
      "step": 30990
    },
    {
      "epoch": 543.7787610619469,
      "grad_norm": 9.9375,
      "learning_rate": 5.997504100036549e-06,
      "loss": 1.1074,
      "num_input_tokens_seen": 17178024,
      "step": 30995
    },
    {
      "epoch": 543.8672566371681,
      "grad_norm": 10.6875,
      "learning_rate": 5.991126106890949e-06,
      "loss": 0.8701,
      "num_input_tokens_seen": 17180696,
      "step": 31000
    },
    {
      "epoch": 543.8672566371681,
      "eval_loss": 1.0310944318771362,
      "eval_runtime": 0.9383,
      "eval_samples_per_second": 26.643,
      "eval_steps_per_second": 13.854,
      "num_input_tokens_seen": 17180696,
      "step": 31000
    },
    {
      "epoch": 543.9557522123894,
      "grad_norm": 6.9375,
      "learning_rate": 5.984751045152576e-06,
      "loss": 1.1504,
      "num_input_tokens_seen": 17183368,
      "step": 31005
    },
    {
      "epoch": 544.0353982300885,
      "grad_norm": 6.8125,
      "learning_rate": 5.978378915804553e-06,
      "loss": 1.72,
      "num_input_tokens_seen": 17185848,
      "step": 31010
    },
    {
      "epoch": 544.1238938053098,
      "grad_norm": 10.75,
      "learning_rate": 5.972009719829547e-06,
      "loss": 1.1765,
      "num_input_tokens_seen": 17188584,
      "step": 31015
    },
    {
      "epoch": 544.212389380531,
      "grad_norm": 6.90625,
      "learning_rate": 5.965643458209755e-06,
      "loss": 0.9927,
      "num_input_tokens_seen": 17191592,
      "step": 31020
    },
    {
      "epoch": 544.3008849557522,
      "grad_norm": 8.5,
      "learning_rate": 5.95928013192695e-06,
      "loss": 1.1965,
      "num_input_tokens_seen": 17193992,
      "step": 31025
    },
    {
      "epoch": 544.3893805309734,
      "grad_norm": 8.25,
      "learning_rate": 5.952919741962423e-06,
      "loss": 1.0482,
      "num_input_tokens_seen": 17196472,
      "step": 31030
    },
    {
      "epoch": 544.4778761061947,
      "grad_norm": 8.3125,
      "learning_rate": 5.946562289297042e-06,
      "loss": 1.0731,
      "num_input_tokens_seen": 17199352,
      "step": 31035
    },
    {
      "epoch": 544.566371681416,
      "grad_norm": 7.1875,
      "learning_rate": 5.9402077749111855e-06,
      "loss": 1.0091,
      "num_input_tokens_seen": 17201912,
      "step": 31040
    },
    {
      "epoch": 544.6548672566372,
      "grad_norm": 8.0,
      "learning_rate": 5.933856199784821e-06,
      "loss": 1.061,
      "num_input_tokens_seen": 17205416,
      "step": 31045
    },
    {
      "epoch": 544.7433628318585,
      "grad_norm": 8.875,
      "learning_rate": 5.927507564897419e-06,
      "loss": 1.0099,
      "num_input_tokens_seen": 17208216,
      "step": 31050
    },
    {
      "epoch": 544.8318584070796,
      "grad_norm": 9.25,
      "learning_rate": 5.9211618712280395e-06,
      "loss": 0.9456,
      "num_input_tokens_seen": 17210728,
      "step": 31055
    },
    {
      "epoch": 544.9203539823009,
      "grad_norm": 8.6875,
      "learning_rate": 5.914819119755255e-06,
      "loss": 1.0078,
      "num_input_tokens_seen": 17213624,
      "step": 31060
    },
    {
      "epoch": 545.0,
      "grad_norm": 4.40625,
      "learning_rate": 5.908479311457205e-06,
      "loss": 0.9528,
      "num_input_tokens_seen": 17216256,
      "step": 31065
    },
    {
      "epoch": 545.0884955752213,
      "grad_norm": 10.0,
      "learning_rate": 5.902142447311559e-06,
      "loss": 1.2569,
      "num_input_tokens_seen": 17219056,
      "step": 31070
    },
    {
      "epoch": 545.1769911504425,
      "grad_norm": 9.5625,
      "learning_rate": 5.895808528295546e-06,
      "loss": 1.0019,
      "num_input_tokens_seen": 17222096,
      "step": 31075
    },
    {
      "epoch": 545.2654867256637,
      "grad_norm": 7.71875,
      "learning_rate": 5.889477555385941e-06,
      "loss": 0.9466,
      "num_input_tokens_seen": 17224800,
      "step": 31080
    },
    {
      "epoch": 545.3539823008849,
      "grad_norm": 9.875,
      "learning_rate": 5.883149529559051e-06,
      "loss": 0.9905,
      "num_input_tokens_seen": 17227504,
      "step": 31085
    },
    {
      "epoch": 545.4424778761062,
      "grad_norm": 10.625,
      "learning_rate": 5.876824451790738e-06,
      "loss": 1.05,
      "num_input_tokens_seen": 17230192,
      "step": 31090
    },
    {
      "epoch": 545.5309734513274,
      "grad_norm": 7.9375,
      "learning_rate": 5.87050232305642e-06,
      "loss": 1.2195,
      "num_input_tokens_seen": 17232848,
      "step": 31095
    },
    {
      "epoch": 545.6194690265487,
      "grad_norm": 5.25,
      "learning_rate": 5.864183144331034e-06,
      "loss": 0.9682,
      "num_input_tokens_seen": 17235744,
      "step": 31100
    },
    {
      "epoch": 545.70796460177,
      "grad_norm": 7.78125,
      "learning_rate": 5.857866916589089e-06,
      "loss": 1.047,
      "num_input_tokens_seen": 17238736,
      "step": 31105
    },
    {
      "epoch": 545.7964601769911,
      "grad_norm": 9.1875,
      "learning_rate": 5.8515536408046216e-06,
      "loss": 1.0822,
      "num_input_tokens_seen": 17241600,
      "step": 31110
    },
    {
      "epoch": 545.8849557522124,
      "grad_norm": 7.84375,
      "learning_rate": 5.845243317951208e-06,
      "loss": 0.9802,
      "num_input_tokens_seen": 17244768,
      "step": 31115
    },
    {
      "epoch": 545.9734513274336,
      "grad_norm": 8.75,
      "learning_rate": 5.838935949001997e-06,
      "loss": 1.0027,
      "num_input_tokens_seen": 17247600,
      "step": 31120
    },
    {
      "epoch": 546.0530973451328,
      "grad_norm": 7.5625,
      "learning_rate": 5.8326315349296476e-06,
      "loss": 1.0659,
      "num_input_tokens_seen": 17249768,
      "step": 31125
    },
    {
      "epoch": 546.141592920354,
      "grad_norm": 7.40625,
      "learning_rate": 5.826330076706396e-06,
      "loss": 1.1015,
      "num_input_tokens_seen": 17252632,
      "step": 31130
    },
    {
      "epoch": 546.2300884955753,
      "grad_norm": 8.5,
      "learning_rate": 5.820031575303988e-06,
      "loss": 0.9641,
      "num_input_tokens_seen": 17255800,
      "step": 31135
    },
    {
      "epoch": 546.3185840707964,
      "grad_norm": 7.71875,
      "learning_rate": 5.813736031693745e-06,
      "loss": 0.9798,
      "num_input_tokens_seen": 17258328,
      "step": 31140
    },
    {
      "epoch": 546.4070796460177,
      "grad_norm": 10.5,
      "learning_rate": 5.807443446846522e-06,
      "loss": 1.122,
      "num_input_tokens_seen": 17260872,
      "step": 31145
    },
    {
      "epoch": 546.4955752212389,
      "grad_norm": 7.75,
      "learning_rate": 5.801153821732699e-06,
      "loss": 1.2526,
      "num_input_tokens_seen": 17263752,
      "step": 31150
    },
    {
      "epoch": 546.5840707964602,
      "grad_norm": 6.96875,
      "learning_rate": 5.794867157322229e-06,
      "loss": 1.1587,
      "num_input_tokens_seen": 17266408,
      "step": 31155
    },
    {
      "epoch": 546.6725663716815,
      "grad_norm": 8.5625,
      "learning_rate": 5.788583454584593e-06,
      "loss": 1.0545,
      "num_input_tokens_seen": 17269416,
      "step": 31160
    },
    {
      "epoch": 546.7610619469026,
      "grad_norm": 7.84375,
      "learning_rate": 5.7823027144888075e-06,
      "loss": 0.8529,
      "num_input_tokens_seen": 17272344,
      "step": 31165
    },
    {
      "epoch": 546.8495575221239,
      "grad_norm": 9.1875,
      "learning_rate": 5.776024938003455e-06,
      "loss": 0.9176,
      "num_input_tokens_seen": 17275080,
      "step": 31170
    },
    {
      "epoch": 546.9380530973451,
      "grad_norm": 7.90625,
      "learning_rate": 5.7697501260966345e-06,
      "loss": 1.1055,
      "num_input_tokens_seen": 17277832,
      "step": 31175
    },
    {
      "epoch": 547.0176991150443,
      "grad_norm": 8.8125,
      "learning_rate": 5.7634782797360145e-06,
      "loss": 0.9426,
      "num_input_tokens_seen": 17280952,
      "step": 31180
    },
    {
      "epoch": 547.1061946902655,
      "grad_norm": 6.9375,
      "learning_rate": 5.757209399888777e-06,
      "loss": 0.8461,
      "num_input_tokens_seen": 17283928,
      "step": 31185
    },
    {
      "epoch": 547.1946902654868,
      "grad_norm": 5.40625,
      "learning_rate": 5.750943487521679e-06,
      "loss": 0.9852,
      "num_input_tokens_seen": 17287032,
      "step": 31190
    },
    {
      "epoch": 547.2831858407079,
      "grad_norm": 10.375,
      "learning_rate": 5.744680543600986e-06,
      "loss": 1.1705,
      "num_input_tokens_seen": 17289544,
      "step": 31195
    },
    {
      "epoch": 547.3716814159292,
      "grad_norm": 9.125,
      "learning_rate": 5.738420569092537e-06,
      "loss": 1.0842,
      "num_input_tokens_seen": 17291896,
      "step": 31200
    },
    {
      "epoch": 547.3716814159292,
      "eval_loss": 1.0310944318771362,
      "eval_runtime": 0.9376,
      "eval_samples_per_second": 26.664,
      "eval_steps_per_second": 13.865,
      "num_input_tokens_seen": 17291896,
      "step": 31200
    },
    {
      "epoch": 547.4601769911504,
      "grad_norm": 7.9375,
      "learning_rate": 5.732163564961684e-06,
      "loss": 1.1007,
      "num_input_tokens_seen": 17295032,
      "step": 31205
    },
    {
      "epoch": 547.5486725663717,
      "grad_norm": 9.8125,
      "learning_rate": 5.725909532173354e-06,
      "loss": 1.191,
      "num_input_tokens_seen": 17297896,
      "step": 31210
    },
    {
      "epoch": 547.637168141593,
      "grad_norm": 9.0625,
      "learning_rate": 5.719658471691977e-06,
      "loss": 0.9915,
      "num_input_tokens_seen": 17300744,
      "step": 31215
    },
    {
      "epoch": 547.7256637168142,
      "grad_norm": 7.1875,
      "learning_rate": 5.71341038448156e-06,
      "loss": 1.0813,
      "num_input_tokens_seen": 17303528,
      "step": 31220
    },
    {
      "epoch": 547.8141592920354,
      "grad_norm": 7.46875,
      "learning_rate": 5.707165271505635e-06,
      "loss": 1.0059,
      "num_input_tokens_seen": 17305800,
      "step": 31225
    },
    {
      "epoch": 547.9026548672566,
      "grad_norm": 7.6875,
      "learning_rate": 5.700923133727271e-06,
      "loss": 1.0313,
      "num_input_tokens_seen": 17308344,
      "step": 31230
    },
    {
      "epoch": 547.9911504424779,
      "grad_norm": 7.15625,
      "learning_rate": 5.694683972109083e-06,
      "loss": 1.071,
      "num_input_tokens_seen": 17311448,
      "step": 31235
    },
    {
      "epoch": 548.070796460177,
      "grad_norm": 10.3125,
      "learning_rate": 5.688447787613241e-06,
      "loss": 0.866,
      "num_input_tokens_seen": 17314056,
      "step": 31240
    },
    {
      "epoch": 548.1592920353983,
      "grad_norm": 10.4375,
      "learning_rate": 5.6822145812014285e-06,
      "loss": 1.1417,
      "num_input_tokens_seen": 17316504,
      "step": 31245
    },
    {
      "epoch": 548.2477876106195,
      "grad_norm": 8.625,
      "learning_rate": 5.675984353834896e-06,
      "loss": 1.1516,
      "num_input_tokens_seen": 17318920,
      "step": 31250
    },
    {
      "epoch": 548.3362831858407,
      "grad_norm": 10.25,
      "learning_rate": 5.66975710647441e-06,
      "loss": 1.045,
      "num_input_tokens_seen": 17321736,
      "step": 31255
    },
    {
      "epoch": 548.4247787610619,
      "grad_norm": 7.375,
      "learning_rate": 5.663532840080304e-06,
      "loss": 1.0024,
      "num_input_tokens_seen": 17324840,
      "step": 31260
    },
    {
      "epoch": 548.5132743362832,
      "grad_norm": 9.25,
      "learning_rate": 5.6573115556124325e-06,
      "loss": 1.1144,
      "num_input_tokens_seen": 17327752,
      "step": 31265
    },
    {
      "epoch": 548.6017699115044,
      "grad_norm": 8.4375,
      "learning_rate": 5.651093254030185e-06,
      "loss": 0.9456,
      "num_input_tokens_seen": 17330232,
      "step": 31270
    },
    {
      "epoch": 548.6902654867257,
      "grad_norm": 8.125,
      "learning_rate": 5.644877936292514e-06,
      "loss": 1.1724,
      "num_input_tokens_seen": 17332840,
      "step": 31275
    },
    {
      "epoch": 548.7787610619469,
      "grad_norm": 10.0625,
      "learning_rate": 5.638665603357901e-06,
      "loss": 1.0806,
      "num_input_tokens_seen": 17335192,
      "step": 31280
    },
    {
      "epoch": 548.8672566371681,
      "grad_norm": 6.0625,
      "learning_rate": 5.632456256184357e-06,
      "loss": 0.9307,
      "num_input_tokens_seen": 17338424,
      "step": 31285
    },
    {
      "epoch": 548.9557522123894,
      "grad_norm": 6.84375,
      "learning_rate": 5.626249895729452e-06,
      "loss": 0.9951,
      "num_input_tokens_seen": 17341608,
      "step": 31290
    },
    {
      "epoch": 549.0353982300885,
      "grad_norm": 8.0,
      "learning_rate": 5.620046522950273e-06,
      "loss": 0.9791,
      "num_input_tokens_seen": 17343792,
      "step": 31295
    },
    {
      "epoch": 549.1238938053098,
      "grad_norm": 7.28125,
      "learning_rate": 5.613846138803464e-06,
      "loss": 1.0152,
      "num_input_tokens_seen": 17346336,
      "step": 31300
    },
    {
      "epoch": 549.212389380531,
      "grad_norm": 8.125,
      "learning_rate": 5.607648744245206e-06,
      "loss": 1.0554,
      "num_input_tokens_seen": 17349184,
      "step": 31305
    },
    {
      "epoch": 549.3008849557522,
      "grad_norm": 9.375,
      "learning_rate": 5.601454340231207e-06,
      "loss": 0.9597,
      "num_input_tokens_seen": 17351792,
      "step": 31310
    },
    {
      "epoch": 549.3893805309734,
      "grad_norm": 5.625,
      "learning_rate": 5.595262927716724e-06,
      "loss": 0.9804,
      "num_input_tokens_seen": 17354304,
      "step": 31315
    },
    {
      "epoch": 549.4778761061947,
      "grad_norm": 7.34375,
      "learning_rate": 5.589074507656561e-06,
      "loss": 1.0548,
      "num_input_tokens_seen": 17357216,
      "step": 31320
    },
    {
      "epoch": 549.566371681416,
      "grad_norm": 11.1875,
      "learning_rate": 5.582889081005044e-06,
      "loss": 1.2482,
      "num_input_tokens_seen": 17359888,
      "step": 31325
    },
    {
      "epoch": 549.6548672566372,
      "grad_norm": 9.75,
      "learning_rate": 5.5767066487160316e-06,
      "loss": 1.1786,
      "num_input_tokens_seen": 17362672,
      "step": 31330
    },
    {
      "epoch": 549.7433628318585,
      "grad_norm": 10.5625,
      "learning_rate": 5.570527211742949e-06,
      "loss": 1.1719,
      "num_input_tokens_seen": 17365440,
      "step": 31335
    },
    {
      "epoch": 549.8318584070796,
      "grad_norm": 7.96875,
      "learning_rate": 5.564350771038731e-06,
      "loss": 0.9201,
      "num_input_tokens_seen": 17368096,
      "step": 31340
    },
    {
      "epoch": 549.9203539823009,
      "grad_norm": 7.3125,
      "learning_rate": 5.558177327555875e-06,
      "loss": 1.0012,
      "num_input_tokens_seen": 17371584,
      "step": 31345
    },
    {
      "epoch": 550.0,
      "grad_norm": 8.125,
      "learning_rate": 5.552006882246388e-06,
      "loss": 1.0123,
      "num_input_tokens_seen": 17373936,
      "step": 31350
    },
    {
      "epoch": 550.0884955752213,
      "grad_norm": 8.25,
      "learning_rate": 5.545839436061839e-06,
      "loss": 0.9409,
      "num_input_tokens_seen": 17376656,
      "step": 31355
    },
    {
      "epoch": 550.1769911504425,
      "grad_norm": 6.09375,
      "learning_rate": 5.539674989953331e-06,
      "loss": 0.9624,
      "num_input_tokens_seen": 17379680,
      "step": 31360
    },
    {
      "epoch": 550.2654867256637,
      "grad_norm": 11.4375,
      "learning_rate": 5.533513544871488e-06,
      "loss": 1.1285,
      "num_input_tokens_seen": 17382448,
      "step": 31365
    },
    {
      "epoch": 550.3539823008849,
      "grad_norm": 8.3125,
      "learning_rate": 5.527355101766493e-06,
      "loss": 0.9438,
      "num_input_tokens_seen": 17385248,
      "step": 31370
    },
    {
      "epoch": 550.4424778761062,
      "grad_norm": 6.59375,
      "learning_rate": 5.521199661588044e-06,
      "loss": 0.9562,
      "num_input_tokens_seen": 17388048,
      "step": 31375
    },
    {
      "epoch": 550.5309734513274,
      "grad_norm": 8.25,
      "learning_rate": 5.5150472252853944e-06,
      "loss": 1.009,
      "num_input_tokens_seen": 17391056,
      "step": 31380
    },
    {
      "epoch": 550.6194690265487,
      "grad_norm": 8.125,
      "learning_rate": 5.50889779380733e-06,
      "loss": 1.1538,
      "num_input_tokens_seen": 17393616,
      "step": 31385
    },
    {
      "epoch": 550.70796460177,
      "grad_norm": 10.8125,
      "learning_rate": 5.5027513681021605e-06,
      "loss": 1.2621,
      "num_input_tokens_seen": 17396496,
      "step": 31390
    },
    {
      "epoch": 550.7964601769911,
      "grad_norm": 13.375,
      "learning_rate": 5.4966079491177545e-06,
      "loss": 1.1253,
      "num_input_tokens_seen": 17399280,
      "step": 31395
    },
    {
      "epoch": 550.8849557522124,
      "grad_norm": 7.40625,
      "learning_rate": 5.490467537801491e-06,
      "loss": 1.0571,
      "num_input_tokens_seen": 17402176,
      "step": 31400
    },
    {
      "epoch": 550.8849557522124,
      "eval_loss": 1.0310944318771362,
      "eval_runtime": 0.9365,
      "eval_samples_per_second": 26.694,
      "eval_steps_per_second": 13.881,
      "num_input_tokens_seen": 17402176,
      "step": 31400
    },
    {
      "epoch": 550.9734513274336,
      "grad_norm": 7.625,
      "learning_rate": 5.484330135100313e-06,
      "loss": 0.893,
      "num_input_tokens_seen": 17404896,
      "step": 31405
    },
    {
      "epoch": 551.0530973451328,
      "grad_norm": 9.1875,
      "learning_rate": 5.4781957419606785e-06,
      "loss": 1.3484,
      "num_input_tokens_seen": 17407064,
      "step": 31410
    },
    {
      "epoch": 551.141592920354,
      "grad_norm": 6.9375,
      "learning_rate": 5.472064359328577e-06,
      "loss": 0.9308,
      "num_input_tokens_seen": 17409928,
      "step": 31415
    },
    {
      "epoch": 551.2300884955753,
      "grad_norm": 7.125,
      "learning_rate": 5.4659359881495565e-06,
      "loss": 0.9669,
      "num_input_tokens_seen": 17412792,
      "step": 31420
    },
    {
      "epoch": 551.3185840707964,
      "grad_norm": 14.3125,
      "learning_rate": 5.4598106293686916e-06,
      "loss": 1.1983,
      "num_input_tokens_seen": 17415784,
      "step": 31425
    },
    {
      "epoch": 551.4070796460177,
      "grad_norm": 11.3125,
      "learning_rate": 5.45368828393058e-06,
      "loss": 1.1106,
      "num_input_tokens_seen": 17418584,
      "step": 31430
    },
    {
      "epoch": 551.4955752212389,
      "grad_norm": 7.5,
      "learning_rate": 5.44756895277937e-06,
      "loss": 0.9878,
      "num_input_tokens_seen": 17421864,
      "step": 31435
    },
    {
      "epoch": 551.5840707964602,
      "grad_norm": 6.125,
      "learning_rate": 5.441452636858746e-06,
      "loss": 1.035,
      "num_input_tokens_seen": 17425064,
      "step": 31440
    },
    {
      "epoch": 551.6725663716815,
      "grad_norm": 6.46875,
      "learning_rate": 5.435339337111905e-06,
      "loss": 0.9966,
      "num_input_tokens_seen": 17427704,
      "step": 31445
    },
    {
      "epoch": 551.7610619469026,
      "grad_norm": 6.625,
      "learning_rate": 5.42922905448161e-06,
      "loss": 1.0385,
      "num_input_tokens_seen": 17430376,
      "step": 31450
    },
    {
      "epoch": 551.8495575221239,
      "grad_norm": 8.0625,
      "learning_rate": 5.423121789910129e-06,
      "loss": 0.9861,
      "num_input_tokens_seen": 17432824,
      "step": 31455
    },
    {
      "epoch": 551.9380530973451,
      "grad_norm": 8.1875,
      "learning_rate": 5.417017544339287e-06,
      "loss": 0.941,
      "num_input_tokens_seen": 17435656,
      "step": 31460
    },
    {
      "epoch": 552.0176991150443,
      "grad_norm": 7.84375,
      "learning_rate": 5.410916318710443e-06,
      "loss": 1.2216,
      "num_input_tokens_seen": 17437960,
      "step": 31465
    },
    {
      "epoch": 552.1061946902655,
      "grad_norm": 7.59375,
      "learning_rate": 5.404818113964466e-06,
      "loss": 1.1198,
      "num_input_tokens_seen": 17440696,
      "step": 31470
    },
    {
      "epoch": 552.1946902654868,
      "grad_norm": 9.125,
      "learning_rate": 5.398722931041792e-06,
      "loss": 1.2056,
      "num_input_tokens_seen": 17443592,
      "step": 31475
    },
    {
      "epoch": 552.2831858407079,
      "grad_norm": 7.625,
      "learning_rate": 5.392630770882367e-06,
      "loss": 0.8612,
      "num_input_tokens_seen": 17446504,
      "step": 31480
    },
    {
      "epoch": 552.3716814159292,
      "grad_norm": 9.1875,
      "learning_rate": 5.3865416344256705e-06,
      "loss": 1.0944,
      "num_input_tokens_seen": 17448888,
      "step": 31485
    },
    {
      "epoch": 552.4601769911504,
      "grad_norm": 9.4375,
      "learning_rate": 5.380455522610742e-06,
      "loss": 0.8129,
      "num_input_tokens_seen": 17451464,
      "step": 31490
    },
    {
      "epoch": 552.5486725663717,
      "grad_norm": 9.6875,
      "learning_rate": 5.374372436376116e-06,
      "loss": 0.9892,
      "num_input_tokens_seen": 17453976,
      "step": 31495
    },
    {
      "epoch": 552.637168141593,
      "grad_norm": 7.40625,
      "learning_rate": 5.368292376659895e-06,
      "loss": 1.0934,
      "num_input_tokens_seen": 17457128,
      "step": 31500
    },
    {
      "epoch": 552.7256637168142,
      "grad_norm": 8.8125,
      "learning_rate": 5.362215344399701e-06,
      "loss": 1.1397,
      "num_input_tokens_seen": 17459960,
      "step": 31505
    },
    {
      "epoch": 552.8141592920354,
      "grad_norm": 10.375,
      "learning_rate": 5.356141340532678e-06,
      "loss": 1.0492,
      "num_input_tokens_seen": 17462712,
      "step": 31510
    },
    {
      "epoch": 552.9026548672566,
      "grad_norm": 9.0,
      "learning_rate": 5.350070365995522e-06,
      "loss": 1.1559,
      "num_input_tokens_seen": 17466280,
      "step": 31515
    },
    {
      "epoch": 552.9911504424779,
      "grad_norm": 6.90625,
      "learning_rate": 5.344002421724459e-06,
      "loss": 1.0563,
      "num_input_tokens_seen": 17468696,
      "step": 31520
    },
    {
      "epoch": 553.070796460177,
      "grad_norm": 10.0,
      "learning_rate": 5.337937508655228e-06,
      "loss": 0.9772,
      "num_input_tokens_seen": 17471184,
      "step": 31525
    },
    {
      "epoch": 553.1592920353983,
      "grad_norm": 8.9375,
      "learning_rate": 5.331875627723126e-06,
      "loss": 0.9063,
      "num_input_tokens_seen": 17473776,
      "step": 31530
    },
    {
      "epoch": 553.2477876106195,
      "grad_norm": 10.1875,
      "learning_rate": 5.325816779862963e-06,
      "loss": 1.0275,
      "num_input_tokens_seen": 17477056,
      "step": 31535
    },
    {
      "epoch": 553.3362831858407,
      "grad_norm": 6.78125,
      "learning_rate": 5.319760966009102e-06,
      "loss": 0.9826,
      "num_input_tokens_seen": 17480128,
      "step": 31540
    },
    {
      "epoch": 553.4247787610619,
      "grad_norm": 6.21875,
      "learning_rate": 5.3137081870954096e-06,
      "loss": 0.8605,
      "num_input_tokens_seen": 17483264,
      "step": 31545
    },
    {
      "epoch": 553.5132743362832,
      "grad_norm": 8.5,
      "learning_rate": 5.307658444055313e-06,
      "loss": 1.01,
      "num_input_tokens_seen": 17486160,
      "step": 31550
    },
    {
      "epoch": 553.6017699115044,
      "grad_norm": 12.75,
      "learning_rate": 5.301611737821749e-06,
      "loss": 1.1435,
      "num_input_tokens_seen": 17488608,
      "step": 31555
    },
    {
      "epoch": 553.6902654867257,
      "grad_norm": 8.9375,
      "learning_rate": 5.295568069327206e-06,
      "loss": 1.1671,
      "num_input_tokens_seen": 17491408,
      "step": 31560
    },
    {
      "epoch": 553.7787610619469,
      "grad_norm": 8.0625,
      "learning_rate": 5.289527439503683e-06,
      "loss": 0.9999,
      "num_input_tokens_seen": 17493936,
      "step": 31565
    },
    {
      "epoch": 553.8672566371681,
      "grad_norm": 9.3125,
      "learning_rate": 5.28348984928273e-06,
      "loss": 1.1376,
      "num_input_tokens_seen": 17496592,
      "step": 31570
    },
    {
      "epoch": 553.9557522123894,
      "grad_norm": 9.4375,
      "learning_rate": 5.27745529959541e-06,
      "loss": 1.2002,
      "num_input_tokens_seen": 17499392,
      "step": 31575
    },
    {
      "epoch": 554.0353982300885,
      "grad_norm": 9.0625,
      "learning_rate": 5.271423791372335e-06,
      "loss": 0.919,
      "num_input_tokens_seen": 17501760,
      "step": 31580
    },
    {
      "epoch": 554.1238938053098,
      "grad_norm": 10.0,
      "learning_rate": 5.26539532554364e-06,
      "loss": 1.1886,
      "num_input_tokens_seen": 17504624,
      "step": 31585
    },
    {
      "epoch": 554.212389380531,
      "grad_norm": 8.4375,
      "learning_rate": 5.25936990303898e-06,
      "loss": 1.132,
      "num_input_tokens_seen": 17507328,
      "step": 31590
    },
    {
      "epoch": 554.3008849557522,
      "grad_norm": 7.84375,
      "learning_rate": 5.253347524787555e-06,
      "loss": 1.1851,
      "num_input_tokens_seen": 17510256,
      "step": 31595
    },
    {
      "epoch": 554.3893805309734,
      "grad_norm": 8.25,
      "learning_rate": 5.2473281917181035e-06,
      "loss": 0.9075,
      "num_input_tokens_seen": 17512704,
      "step": 31600
    },
    {
      "epoch": 554.3893805309734,
      "eval_loss": 1.0310944318771362,
      "eval_runtime": 0.9439,
      "eval_samples_per_second": 26.485,
      "eval_steps_per_second": 13.772,
      "num_input_tokens_seen": 17512704,
      "step": 31600
    },
    {
      "epoch": 554.4778761061947,
      "grad_norm": 6.84375,
      "learning_rate": 5.241311904758864e-06,
      "loss": 1.0024,
      "num_input_tokens_seen": 17515408,
      "step": 31605
    },
    {
      "epoch": 554.566371681416,
      "grad_norm": 9.5625,
      "learning_rate": 5.23529866483764e-06,
      "loss": 1.0138,
      "num_input_tokens_seen": 17518336,
      "step": 31610
    },
    {
      "epoch": 554.6548672566372,
      "grad_norm": 9.75,
      "learning_rate": 5.229288472881732e-06,
      "loss": 0.8805,
      "num_input_tokens_seen": 17521072,
      "step": 31615
    },
    {
      "epoch": 554.7433628318585,
      "grad_norm": 6.0625,
      "learning_rate": 5.2232813298180025e-06,
      "loss": 1.0352,
      "num_input_tokens_seen": 17523760,
      "step": 31620
    },
    {
      "epoch": 554.8318584070796,
      "grad_norm": 9.6875,
      "learning_rate": 5.217277236572824e-06,
      "loss": 1.2209,
      "num_input_tokens_seen": 17526512,
      "step": 31625
    },
    {
      "epoch": 554.9203539823009,
      "grad_norm": 8.0,
      "learning_rate": 5.211276194072093e-06,
      "loss": 0.8351,
      "num_input_tokens_seen": 17529760,
      "step": 31630
    },
    {
      "epoch": 555.0,
      "grad_norm": 9.0625,
      "learning_rate": 5.205278203241254e-06,
      "loss": 1.2434,
      "num_input_tokens_seen": 17532192,
      "step": 31635
    },
    {
      "epoch": 555.0884955752213,
      "grad_norm": 7.5625,
      "learning_rate": 5.199283265005278e-06,
      "loss": 0.8446,
      "num_input_tokens_seen": 17535328,
      "step": 31640
    },
    {
      "epoch": 555.1769911504425,
      "grad_norm": 8.6875,
      "learning_rate": 5.193291380288648e-06,
      "loss": 0.9585,
      "num_input_tokens_seen": 17538176,
      "step": 31645
    },
    {
      "epoch": 555.2654867256637,
      "grad_norm": 6.9375,
      "learning_rate": 5.1873025500153995e-06,
      "loss": 1.0482,
      "num_input_tokens_seen": 17541376,
      "step": 31650
    },
    {
      "epoch": 555.3539823008849,
      "grad_norm": 8.3125,
      "learning_rate": 5.181316775109071e-06,
      "loss": 1.0146,
      "num_input_tokens_seen": 17544160,
      "step": 31655
    },
    {
      "epoch": 555.4424778761062,
      "grad_norm": 9.5625,
      "learning_rate": 5.1753340564927564e-06,
      "loss": 1.167,
      "num_input_tokens_seen": 17547088,
      "step": 31660
    },
    {
      "epoch": 555.5309734513274,
      "grad_norm": 7.53125,
      "learning_rate": 5.169354395089068e-06,
      "loss": 1.1331,
      "num_input_tokens_seen": 17550048,
      "step": 31665
    },
    {
      "epoch": 555.6194690265487,
      "grad_norm": 11.125,
      "learning_rate": 5.1633777918201346e-06,
      "loss": 1.1104,
      "num_input_tokens_seen": 17552560,
      "step": 31670
    },
    {
      "epoch": 555.70796460177,
      "grad_norm": 9.5,
      "learning_rate": 5.157404247607625e-06,
      "loss": 1.1167,
      "num_input_tokens_seen": 17555056,
      "step": 31675
    },
    {
      "epoch": 555.7964601769911,
      "grad_norm": 7.53125,
      "learning_rate": 5.1514337633727454e-06,
      "loss": 0.9143,
      "num_input_tokens_seen": 17557648,
      "step": 31680
    },
    {
      "epoch": 555.8849557522124,
      "grad_norm": 8.125,
      "learning_rate": 5.145466340036206e-06,
      "loss": 0.973,
      "num_input_tokens_seen": 17560400,
      "step": 31685
    },
    {
      "epoch": 555.9734513274336,
      "grad_norm": 8.125,
      "learning_rate": 5.139501978518274e-06,
      "loss": 1.1387,
      "num_input_tokens_seen": 17563248,
      "step": 31690
    },
    {
      "epoch": 556.0530973451328,
      "grad_norm": 9.0625,
      "learning_rate": 5.133540679738716e-06,
      "loss": 1.2667,
      "num_input_tokens_seen": 17565200,
      "step": 31695
    },
    {
      "epoch": 556.141592920354,
      "grad_norm": 12.0,
      "learning_rate": 5.127582444616838e-06,
      "loss": 1.0932,
      "num_input_tokens_seen": 17568320,
      "step": 31700
    },
    {
      "epoch": 556.2300884955753,
      "grad_norm": 9.5625,
      "learning_rate": 5.121627274071486e-06,
      "loss": 0.9756,
      "num_input_tokens_seen": 17570880,
      "step": 31705
    },
    {
      "epoch": 556.3185840707964,
      "grad_norm": 9.1875,
      "learning_rate": 5.115675169021009e-06,
      "loss": 1.0134,
      "num_input_tokens_seen": 17573936,
      "step": 31710
    },
    {
      "epoch": 556.4070796460177,
      "grad_norm": 7.1875,
      "learning_rate": 5.1097261303832994e-06,
      "loss": 1.125,
      "num_input_tokens_seen": 17576304,
      "step": 31715
    },
    {
      "epoch": 556.4955752212389,
      "grad_norm": 6.8125,
      "learning_rate": 5.103780159075788e-06,
      "loss": 0.9794,
      "num_input_tokens_seen": 17579376,
      "step": 31720
    },
    {
      "epoch": 556.5840707964602,
      "grad_norm": 7.71875,
      "learning_rate": 5.0978372560154e-06,
      "loss": 1.0749,
      "num_input_tokens_seen": 17581840,
      "step": 31725
    },
    {
      "epoch": 556.6725663716815,
      "grad_norm": 9.1875,
      "learning_rate": 5.091897422118619e-06,
      "loss": 1.0333,
      "num_input_tokens_seen": 17585152,
      "step": 31730
    },
    {
      "epoch": 556.7610619469026,
      "grad_norm": 8.4375,
      "learning_rate": 5.0859606583014305e-06,
      "loss": 1.0167,
      "num_input_tokens_seen": 17587904,
      "step": 31735
    },
    {
      "epoch": 556.8495575221239,
      "grad_norm": 9.25,
      "learning_rate": 5.080026965479365e-06,
      "loss": 1.0584,
      "num_input_tokens_seen": 17590752,
      "step": 31740
    },
    {
      "epoch": 556.9380530973451,
      "grad_norm": 8.0,
      "learning_rate": 5.074096344567475e-06,
      "loss": 1.047,
      "num_input_tokens_seen": 17593792,
      "step": 31745
    },
    {
      "epoch": 557.0176991150443,
      "grad_norm": 7.59375,
      "learning_rate": 5.0681687964803294e-06,
      "loss": 1.0843,
      "num_input_tokens_seen": 17596072,
      "step": 31750
    },
    {
      "epoch": 557.1061946902655,
      "grad_norm": 6.15625,
      "learning_rate": 5.06224432213204e-06,
      "loss": 1.0291,
      "num_input_tokens_seen": 17599032,
      "step": 31755
    },
    {
      "epoch": 557.1946902654868,
      "grad_norm": 6.84375,
      "learning_rate": 5.056322922436224e-06,
      "loss": 1.0467,
      "num_input_tokens_seen": 17601704,
      "step": 31760
    },
    {
      "epoch": 557.2831858407079,
      "grad_norm": 7.75,
      "learning_rate": 5.0504045983060465e-06,
      "loss": 1.0869,
      "num_input_tokens_seen": 17604696,
      "step": 31765
    },
    {
      "epoch": 557.3716814159292,
      "grad_norm": 7.59375,
      "learning_rate": 5.044489350654183e-06,
      "loss": 1.0839,
      "num_input_tokens_seen": 17607176,
      "step": 31770
    },
    {
      "epoch": 557.4601769911504,
      "grad_norm": 8.0,
      "learning_rate": 5.038577180392831e-06,
      "loss": 1.0685,
      "num_input_tokens_seen": 17610440,
      "step": 31775
    },
    {
      "epoch": 557.5486725663717,
      "grad_norm": 6.78125,
      "learning_rate": 5.032668088433729e-06,
      "loss": 1.0134,
      "num_input_tokens_seen": 17613080,
      "step": 31780
    },
    {
      "epoch": 557.637168141593,
      "grad_norm": 7.40625,
      "learning_rate": 5.02676207568814e-06,
      "loss": 1.0978,
      "num_input_tokens_seen": 17616024,
      "step": 31785
    },
    {
      "epoch": 557.7256637168142,
      "grad_norm": 7.8125,
      "learning_rate": 5.02085914306683e-06,
      "loss": 1.0334,
      "num_input_tokens_seen": 17619176,
      "step": 31790
    },
    {
      "epoch": 557.8141592920354,
      "grad_norm": 8.125,
      "learning_rate": 5.014959291480123e-06,
      "loss": 0.9784,
      "num_input_tokens_seen": 17622200,
      "step": 31795
    },
    {
      "epoch": 557.9026548672566,
      "grad_norm": 8.875,
      "learning_rate": 5.009062521837835e-06,
      "loss": 0.9215,
      "num_input_tokens_seen": 17624600,
      "step": 31800
    },
    {
      "epoch": 557.9026548672566,
      "eval_loss": 1.0310944318771362,
      "eval_runtime": 0.9432,
      "eval_samples_per_second": 26.506,
      "eval_steps_per_second": 13.783,
      "num_input_tokens_seen": 17624600,
      "step": 31800
    },
    {
      "epoch": 557.9911504424779,
      "grad_norm": 8.125,
      "learning_rate": 5.003168835049324e-06,
      "loss": 1.1909,
      "num_input_tokens_seen": 17627176,
      "step": 31805
    },
    {
      "epoch": 558.070796460177,
      "grad_norm": 11.8125,
      "learning_rate": 4.997278232023483e-06,
      "loss": 1.2648,
      "num_input_tokens_seen": 17629576,
      "step": 31810
    },
    {
      "epoch": 558.1592920353983,
      "grad_norm": 9.8125,
      "learning_rate": 4.9913907136687036e-06,
      "loss": 0.9978,
      "num_input_tokens_seen": 17632312,
      "step": 31815
    },
    {
      "epoch": 558.2477876106195,
      "grad_norm": 8.875,
      "learning_rate": 4.985506280892918e-06,
      "loss": 0.964,
      "num_input_tokens_seen": 17635736,
      "step": 31820
    },
    {
      "epoch": 558.3362831858407,
      "grad_norm": 7.09375,
      "learning_rate": 4.979624934603589e-06,
      "loss": 0.9632,
      "num_input_tokens_seen": 17638904,
      "step": 31825
    },
    {
      "epoch": 558.4247787610619,
      "grad_norm": 6.84375,
      "learning_rate": 4.97374667570768e-06,
      "loss": 1.0274,
      "num_input_tokens_seen": 17641656,
      "step": 31830
    },
    {
      "epoch": 558.5132743362832,
      "grad_norm": 12.1875,
      "learning_rate": 4.967871505111704e-06,
      "loss": 1.165,
      "num_input_tokens_seen": 17644040,
      "step": 31835
    },
    {
      "epoch": 558.6017699115044,
      "grad_norm": 6.875,
      "learning_rate": 4.961999423721686e-06,
      "loss": 0.9883,
      "num_input_tokens_seen": 17646840,
      "step": 31840
    },
    {
      "epoch": 558.6902654867257,
      "grad_norm": 11.0625,
      "learning_rate": 4.956130432443159e-06,
      "loss": 1.1123,
      "num_input_tokens_seen": 17649496,
      "step": 31845
    },
    {
      "epoch": 558.7787610619469,
      "grad_norm": 8.5625,
      "learning_rate": 4.950264532181215e-06,
      "loss": 1.17,
      "num_input_tokens_seen": 17652520,
      "step": 31850
    },
    {
      "epoch": 558.8672566371681,
      "grad_norm": 7.53125,
      "learning_rate": 4.944401723840433e-06,
      "loss": 0.8805,
      "num_input_tokens_seen": 17655272,
      "step": 31855
    },
    {
      "epoch": 558.9557522123894,
      "grad_norm": 7.84375,
      "learning_rate": 4.938542008324942e-06,
      "loss": 1.1212,
      "num_input_tokens_seen": 17658008,
      "step": 31860
    },
    {
      "epoch": 559.0353982300885,
      "grad_norm": 5.5,
      "learning_rate": 4.9326853865383855e-06,
      "loss": 0.9687,
      "num_input_tokens_seen": 17660312,
      "step": 31865
    },
    {
      "epoch": 559.1238938053098,
      "grad_norm": 8.75,
      "learning_rate": 4.926831859383918e-06,
      "loss": 1.2954,
      "num_input_tokens_seen": 17662584,
      "step": 31870
    },
    {
      "epoch": 559.212389380531,
      "grad_norm": 8.0625,
      "learning_rate": 4.92098142776424e-06,
      "loss": 1.0435,
      "num_input_tokens_seen": 17665240,
      "step": 31875
    },
    {
      "epoch": 559.3008849557522,
      "grad_norm": 9.1875,
      "learning_rate": 4.91513409258155e-06,
      "loss": 1.0067,
      "num_input_tokens_seen": 17667912,
      "step": 31880
    },
    {
      "epoch": 559.3893805309734,
      "grad_norm": 12.125,
      "learning_rate": 4.909289854737581e-06,
      "loss": 1.0315,
      "num_input_tokens_seen": 17670968,
      "step": 31885
    },
    {
      "epoch": 559.4778761061947,
      "grad_norm": 7.15625,
      "learning_rate": 4.903448715133602e-06,
      "loss": 0.9285,
      "num_input_tokens_seen": 17674200,
      "step": 31890
    },
    {
      "epoch": 559.566371681416,
      "grad_norm": 8.1875,
      "learning_rate": 4.897610674670372e-06,
      "loss": 0.986,
      "num_input_tokens_seen": 17677208,
      "step": 31895
    },
    {
      "epoch": 559.6548672566372,
      "grad_norm": 8.1875,
      "learning_rate": 4.8917757342482e-06,
      "loss": 1.0877,
      "num_input_tokens_seen": 17680152,
      "step": 31900
    },
    {
      "epoch": 559.7433628318585,
      "grad_norm": 10.75,
      "learning_rate": 4.885943894766909e-06,
      "loss": 1.0403,
      "num_input_tokens_seen": 17683016,
      "step": 31905
    },
    {
      "epoch": 559.8318584070796,
      "grad_norm": 7.15625,
      "learning_rate": 4.880115157125842e-06,
      "loss": 0.8799,
      "num_input_tokens_seen": 17685928,
      "step": 31910
    },
    {
      "epoch": 559.9203539823009,
      "grad_norm": 7.625,
      "learning_rate": 4.874289522223857e-06,
      "loss": 1.0062,
      "num_input_tokens_seen": 17688408,
      "step": 31915
    },
    {
      "epoch": 560.0,
      "grad_norm": 9.625,
      "learning_rate": 4.868466990959339e-06,
      "loss": 1.4276,
      "num_input_tokens_seen": 17690680,
      "step": 31920
    },
    {
      "epoch": 560.0884955752213,
      "grad_norm": 9.125,
      "learning_rate": 4.8626475642301964e-06,
      "loss": 1.2142,
      "num_input_tokens_seen": 17693160,
      "step": 31925
    },
    {
      "epoch": 560.1769911504425,
      "grad_norm": 8.125,
      "learning_rate": 4.856831242933871e-06,
      "loss": 0.9166,
      "num_input_tokens_seen": 17696664,
      "step": 31930
    },
    {
      "epoch": 560.2654867256637,
      "grad_norm": 11.125,
      "learning_rate": 4.851018027967294e-06,
      "loss": 1.1163,
      "num_input_tokens_seen": 17699176,
      "step": 31935
    },
    {
      "epoch": 560.3539823008849,
      "grad_norm": 8.3125,
      "learning_rate": 4.845207920226946e-06,
      "loss": 1.0742,
      "num_input_tokens_seen": 17701880,
      "step": 31940
    },
    {
      "epoch": 560.4424778761062,
      "grad_norm": 11.625,
      "learning_rate": 4.839400920608825e-06,
      "loss": 0.9244,
      "num_input_tokens_seen": 17705080,
      "step": 31945
    },
    {
      "epoch": 560.5309734513274,
      "grad_norm": 9.0625,
      "learning_rate": 4.83359703000843e-06,
      "loss": 1.0129,
      "num_input_tokens_seen": 17707736,
      "step": 31950
    },
    {
      "epoch": 560.6194690265487,
      "grad_norm": 8.25,
      "learning_rate": 4.827796249320804e-06,
      "loss": 1.2365,
      "num_input_tokens_seen": 17710920,
      "step": 31955
    },
    {
      "epoch": 560.70796460177,
      "grad_norm": 8.25,
      "learning_rate": 4.82199857944049e-06,
      "loss": 0.9888,
      "num_input_tokens_seen": 17713496,
      "step": 31960
    },
    {
      "epoch": 560.7964601769911,
      "grad_norm": 8.6875,
      "learning_rate": 4.8162040212615695e-06,
      "loss": 1.0644,
      "num_input_tokens_seen": 17716024,
      "step": 31965
    },
    {
      "epoch": 560.8849557522124,
      "grad_norm": 6.375,
      "learning_rate": 4.810412575677639e-06,
      "loss": 1.0963,
      "num_input_tokens_seen": 17718632,
      "step": 31970
    },
    {
      "epoch": 560.9734513274336,
      "grad_norm": 8.875,
      "learning_rate": 4.804624243581801e-06,
      "loss": 0.8374,
      "num_input_tokens_seen": 17721432,
      "step": 31975
    },
    {
      "epoch": 561.0530973451328,
      "grad_norm": 10.875,
      "learning_rate": 4.798839025866703e-06,
      "loss": 1.3117,
      "num_input_tokens_seen": 17723504,
      "step": 31980
    },
    {
      "epoch": 561.141592920354,
      "grad_norm": 6.59375,
      "learning_rate": 4.793056923424491e-06,
      "loss": 0.8845,
      "num_input_tokens_seen": 17726144,
      "step": 31985
    },
    {
      "epoch": 561.2300884955753,
      "grad_norm": 7.5,
      "learning_rate": 4.78727793714683e-06,
      "loss": 1.1596,
      "num_input_tokens_seen": 17728640,
      "step": 31990
    },
    {
      "epoch": 561.3185840707964,
      "grad_norm": 8.4375,
      "learning_rate": 4.7815020679249285e-06,
      "loss": 0.9619,
      "num_input_tokens_seen": 17731440,
      "step": 31995
    },
    {
      "epoch": 561.4070796460177,
      "grad_norm": 7.46875,
      "learning_rate": 4.775729316649483e-06,
      "loss": 1.0647,
      "num_input_tokens_seen": 17734208,
      "step": 32000
    },
    {
      "epoch": 561.4070796460177,
      "eval_loss": 1.0310944318771362,
      "eval_runtime": 0.9385,
      "eval_samples_per_second": 26.638,
      "eval_steps_per_second": 13.852,
      "num_input_tokens_seen": 17734208,
      "step": 32000
    },
    {
      "epoch": 561.4955752212389,
      "grad_norm": 11.8125,
      "learning_rate": 4.769959684210728e-06,
      "loss": 0.9958,
      "num_input_tokens_seen": 17737056,
      "step": 32005
    },
    {
      "epoch": 561.5840707964602,
      "grad_norm": 8.1875,
      "learning_rate": 4.764193171498426e-06,
      "loss": 1.1081,
      "num_input_tokens_seen": 17739680,
      "step": 32010
    },
    {
      "epoch": 561.6725663716815,
      "grad_norm": 9.6875,
      "learning_rate": 4.75842977940183e-06,
      "loss": 1.0868,
      "num_input_tokens_seen": 17742704,
      "step": 32015
    },
    {
      "epoch": 561.7610619469026,
      "grad_norm": 6.5,
      "learning_rate": 4.752669508809729e-06,
      "loss": 0.9689,
      "num_input_tokens_seen": 17745136,
      "step": 32020
    },
    {
      "epoch": 561.8495575221239,
      "grad_norm": 9.625,
      "learning_rate": 4.746912360610445e-06,
      "loss": 1.1324,
      "num_input_tokens_seen": 17748368,
      "step": 32025
    },
    {
      "epoch": 561.9380530973451,
      "grad_norm": 8.75,
      "learning_rate": 4.741158335691781e-06,
      "loss": 1.0365,
      "num_input_tokens_seen": 17751360,
      "step": 32030
    },
    {
      "epoch": 562.0176991150443,
      "grad_norm": 7.0625,
      "learning_rate": 4.7354074349410994e-06,
      "loss": 1.0275,
      "num_input_tokens_seen": 17754280,
      "step": 32035
    },
    {
      "epoch": 562.1061946902655,
      "grad_norm": 9.1875,
      "learning_rate": 4.729659659245245e-06,
      "loss": 0.9941,
      "num_input_tokens_seen": 17756760,
      "step": 32040
    },
    {
      "epoch": 562.1946902654868,
      "grad_norm": 7.28125,
      "learning_rate": 4.723915009490601e-06,
      "loss": 0.904,
      "num_input_tokens_seen": 17759368,
      "step": 32045
    },
    {
      "epoch": 562.2831858407079,
      "grad_norm": 7.65625,
      "learning_rate": 4.718173486563077e-06,
      "loss": 0.9371,
      "num_input_tokens_seen": 17762840,
      "step": 32050
    },
    {
      "epoch": 562.3716814159292,
      "grad_norm": 7.34375,
      "learning_rate": 4.71243509134808e-06,
      "loss": 1.004,
      "num_input_tokens_seen": 17765256,
      "step": 32055
    },
    {
      "epoch": 562.4601769911504,
      "grad_norm": 9.9375,
      "learning_rate": 4.706699824730532e-06,
      "loss": 1.1181,
      "num_input_tokens_seen": 17767816,
      "step": 32060
    },
    {
      "epoch": 562.5486725663717,
      "grad_norm": 8.4375,
      "learning_rate": 4.700967687594901e-06,
      "loss": 0.9393,
      "num_input_tokens_seen": 17770376,
      "step": 32065
    },
    {
      "epoch": 562.637168141593,
      "grad_norm": 10.125,
      "learning_rate": 4.69523868082514e-06,
      "loss": 1.2679,
      "num_input_tokens_seen": 17773224,
      "step": 32070
    },
    {
      "epoch": 562.7256637168142,
      "grad_norm": 7.75,
      "learning_rate": 4.689512805304747e-06,
      "loss": 1.1819,
      "num_input_tokens_seen": 17775848,
      "step": 32075
    },
    {
      "epoch": 562.8141592920354,
      "grad_norm": 7.65625,
      "learning_rate": 4.683790061916707e-06,
      "loss": 1.1252,
      "num_input_tokens_seen": 17778744,
      "step": 32080
    },
    {
      "epoch": 562.9026548672566,
      "grad_norm": 7.875,
      "learning_rate": 4.678070451543551e-06,
      "loss": 0.961,
      "num_input_tokens_seen": 17782056,
      "step": 32085
    },
    {
      "epoch": 562.9911504424779,
      "grad_norm": 12.5625,
      "learning_rate": 4.6723539750673204e-06,
      "loss": 1.1237,
      "num_input_tokens_seen": 17784792,
      "step": 32090
    },
    {
      "epoch": 563.070796460177,
      "grad_norm": 8.3125,
      "learning_rate": 4.666640633369551e-06,
      "loss": 1.1288,
      "num_input_tokens_seen": 17787160,
      "step": 32095
    },
    {
      "epoch": 563.1592920353983,
      "grad_norm": 8.75,
      "learning_rate": 4.660930427331323e-06,
      "loss": 1.1569,
      "num_input_tokens_seen": 17789784,
      "step": 32100
    },
    {
      "epoch": 563.2477876106195,
      "grad_norm": 9.1875,
      "learning_rate": 4.6552233578332244e-06,
      "loss": 1.0537,
      "num_input_tokens_seen": 17792136,
      "step": 32105
    },
    {
      "epoch": 563.3362831858407,
      "grad_norm": 7.25,
      "learning_rate": 4.649519425755347e-06,
      "loss": 0.9572,
      "num_input_tokens_seen": 17795048,
      "step": 32110
    },
    {
      "epoch": 563.4247787610619,
      "grad_norm": 9.6875,
      "learning_rate": 4.64381863197732e-06,
      "loss": 1.0531,
      "num_input_tokens_seen": 17797864,
      "step": 32115
    },
    {
      "epoch": 563.5132743362832,
      "grad_norm": 9.5625,
      "learning_rate": 4.638120977378269e-06,
      "loss": 1.0519,
      "num_input_tokens_seen": 17800248,
      "step": 32120
    },
    {
      "epoch": 563.6017699115044,
      "grad_norm": 9.25,
      "learning_rate": 4.632426462836848e-06,
      "loss": 0.9213,
      "num_input_tokens_seen": 17803256,
      "step": 32125
    },
    {
      "epoch": 563.6902654867257,
      "grad_norm": 7.9375,
      "learning_rate": 4.626735089231224e-06,
      "loss": 0.9996,
      "num_input_tokens_seen": 17806648,
      "step": 32130
    },
    {
      "epoch": 563.7787610619469,
      "grad_norm": 7.625,
      "learning_rate": 4.621046857439068e-06,
      "loss": 1.0103,
      "num_input_tokens_seen": 17809448,
      "step": 32135
    },
    {
      "epoch": 563.8672566371681,
      "grad_norm": 9.8125,
      "learning_rate": 4.615361768337587e-06,
      "loss": 1.0029,
      "num_input_tokens_seen": 17812696,
      "step": 32140
    },
    {
      "epoch": 563.9557522123894,
      "grad_norm": 9.6875,
      "learning_rate": 4.6096798228034946e-06,
      "loss": 1.0881,
      "num_input_tokens_seen": 17815784,
      "step": 32145
    },
    {
      "epoch": 564.0353982300885,
      "grad_norm": 8.625,
      "learning_rate": 4.604001021713008e-06,
      "loss": 1.1735,
      "num_input_tokens_seen": 17818264,
      "step": 32150
    },
    {
      "epoch": 564.1238938053098,
      "grad_norm": 7.4375,
      "learning_rate": 4.598325365941883e-06,
      "loss": 1.0888,
      "num_input_tokens_seen": 17821000,
      "step": 32155
    },
    {
      "epoch": 564.212389380531,
      "grad_norm": 9.625,
      "learning_rate": 4.5926528563653645e-06,
      "loss": 1.0869,
      "num_input_tokens_seen": 17823640,
      "step": 32160
    },
    {
      "epoch": 564.3008849557522,
      "grad_norm": 8.1875,
      "learning_rate": 4.5869834938582295e-06,
      "loss": 1.1755,
      "num_input_tokens_seen": 17826488,
      "step": 32165
    },
    {
      "epoch": 564.3893805309734,
      "grad_norm": 7.4375,
      "learning_rate": 4.581317279294772e-06,
      "loss": 0.8618,
      "num_input_tokens_seen": 17829416,
      "step": 32170
    },
    {
      "epoch": 564.4778761061947,
      "grad_norm": 8.875,
      "learning_rate": 4.57565421354878e-06,
      "loss": 1.1334,
      "num_input_tokens_seen": 17832136,
      "step": 32175
    },
    {
      "epoch": 564.566371681416,
      "grad_norm": 8.625,
      "learning_rate": 4.569994297493579e-06,
      "loss": 1.0227,
      "num_input_tokens_seen": 17834632,
      "step": 32180
    },
    {
      "epoch": 564.6548672566372,
      "grad_norm": 7.90625,
      "learning_rate": 4.564337532002002e-06,
      "loss": 1.0874,
      "num_input_tokens_seen": 17837256,
      "step": 32185
    },
    {
      "epoch": 564.7433628318585,
      "grad_norm": 7.5,
      "learning_rate": 4.55868391794638e-06,
      "loss": 0.9047,
      "num_input_tokens_seen": 17840200,
      "step": 32190
    },
    {
      "epoch": 564.8318584070796,
      "grad_norm": 8.5,
      "learning_rate": 4.553033456198588e-06,
      "loss": 1.1079,
      "num_input_tokens_seen": 17842376,
      "step": 32195
    },
    {
      "epoch": 564.9203539823009,
      "grad_norm": 8.4375,
      "learning_rate": 4.54738614762999e-06,
      "loss": 1.1348,
      "num_input_tokens_seen": 17845224,
      "step": 32200
    },
    {
      "epoch": 564.9203539823009,
      "eval_loss": 1.0310944318771362,
      "eval_runtime": 0.9365,
      "eval_samples_per_second": 26.696,
      "eval_steps_per_second": 13.882,
      "num_input_tokens_seen": 17845224,
      "step": 32200
    },
    {
      "epoch": 565.0,
      "grad_norm": 11.3125,
      "learning_rate": 4.541741993111465e-06,
      "loss": 0.9769,
      "num_input_tokens_seen": 17848216,
      "step": 32205
    },
    {
      "epoch": 565.0884955752213,
      "grad_norm": 8.1875,
      "learning_rate": 4.536100993513423e-06,
      "loss": 1.0311,
      "num_input_tokens_seen": 17851208,
      "step": 32210
    },
    {
      "epoch": 565.1769911504425,
      "grad_norm": 8.1875,
      "learning_rate": 4.530463149705768e-06,
      "loss": 1.0837,
      "num_input_tokens_seen": 17853640,
      "step": 32215
    },
    {
      "epoch": 565.2654867256637,
      "grad_norm": 6.3125,
      "learning_rate": 4.524828462557934e-06,
      "loss": 1.1221,
      "num_input_tokens_seen": 17856360,
      "step": 32220
    },
    {
      "epoch": 565.3539823008849,
      "grad_norm": 12.3125,
      "learning_rate": 4.5191969329388625e-06,
      "loss": 1.1625,
      "num_input_tokens_seen": 17859016,
      "step": 32225
    },
    {
      "epoch": 565.4424778761062,
      "grad_norm": 7.0,
      "learning_rate": 4.5135685617169965e-06,
      "loss": 0.98,
      "num_input_tokens_seen": 17861672,
      "step": 32230
    },
    {
      "epoch": 565.5309734513274,
      "grad_norm": 11.3125,
      "learning_rate": 4.507943349760313e-06,
      "loss": 1.1138,
      "num_input_tokens_seen": 17864712,
      "step": 32235
    },
    {
      "epoch": 565.6194690265487,
      "grad_norm": 13.4375,
      "learning_rate": 4.502321297936277e-06,
      "loss": 0.7648,
      "num_input_tokens_seen": 17867672,
      "step": 32240
    },
    {
      "epoch": 565.70796460177,
      "grad_norm": 9.5,
      "learning_rate": 4.496702407111888e-06,
      "loss": 1.0954,
      "num_input_tokens_seen": 17870504,
      "step": 32245
    },
    {
      "epoch": 565.7964601769911,
      "grad_norm": 8.375,
      "learning_rate": 4.491086678153653e-06,
      "loss": 1.0562,
      "num_input_tokens_seen": 17873192,
      "step": 32250
    },
    {
      "epoch": 565.8849557522124,
      "grad_norm": 10.4375,
      "learning_rate": 4.485474111927579e-06,
      "loss": 1.0682,
      "num_input_tokens_seen": 17875992,
      "step": 32255
    },
    {
      "epoch": 565.9734513274336,
      "grad_norm": 7.71875,
      "learning_rate": 4.479864709299197e-06,
      "loss": 1.1009,
      "num_input_tokens_seen": 17878952,
      "step": 32260
    },
    {
      "epoch": 566.0530973451328,
      "grad_norm": 9.5,
      "learning_rate": 4.474258471133555e-06,
      "loss": 1.0172,
      "num_input_tokens_seen": 17881336,
      "step": 32265
    },
    {
      "epoch": 566.141592920354,
      "grad_norm": 9.75,
      "learning_rate": 4.4686553982952014e-06,
      "loss": 1.015,
      "num_input_tokens_seen": 17884392,
      "step": 32270
    },
    {
      "epoch": 566.2300884955753,
      "grad_norm": 8.9375,
      "learning_rate": 4.463055491648191e-06,
      "loss": 1.0087,
      "num_input_tokens_seen": 17886920,
      "step": 32275
    },
    {
      "epoch": 566.3185840707964,
      "grad_norm": 7.9375,
      "learning_rate": 4.457458752056112e-06,
      "loss": 0.9238,
      "num_input_tokens_seen": 17889928,
      "step": 32280
    },
    {
      "epoch": 566.4070796460177,
      "grad_norm": 11.9375,
      "learning_rate": 4.451865180382042e-06,
      "loss": 1.1087,
      "num_input_tokens_seen": 17892968,
      "step": 32285
    },
    {
      "epoch": 566.4955752212389,
      "grad_norm": 8.375,
      "learning_rate": 4.4462747774885936e-06,
      "loss": 1.0228,
      "num_input_tokens_seen": 17895448,
      "step": 32290
    },
    {
      "epoch": 566.5840707964602,
      "grad_norm": 7.78125,
      "learning_rate": 4.440687544237859e-06,
      "loss": 1.0588,
      "num_input_tokens_seen": 17898200,
      "step": 32295
    },
    {
      "epoch": 566.6725663716815,
      "grad_norm": 9.125,
      "learning_rate": 4.435103481491471e-06,
      "loss": 1.0059,
      "num_input_tokens_seen": 17900936,
      "step": 32300
    },
    {
      "epoch": 566.7610619469026,
      "grad_norm": 8.25,
      "learning_rate": 4.429522590110569e-06,
      "loss": 1.0315,
      "num_input_tokens_seen": 17903896,
      "step": 32305
    },
    {
      "epoch": 566.8495575221239,
      "grad_norm": 8.5,
      "learning_rate": 4.423944870955779e-06,
      "loss": 1.0141,
      "num_input_tokens_seen": 17906728,
      "step": 32310
    },
    {
      "epoch": 566.9380530973451,
      "grad_norm": 8.5,
      "learning_rate": 4.418370324887272e-06,
      "loss": 1.1366,
      "num_input_tokens_seen": 17909592,
      "step": 32315
    },
    {
      "epoch": 567.0176991150443,
      "grad_norm": 7.46875,
      "learning_rate": 4.412798952764699e-06,
      "loss": 1.2242,
      "num_input_tokens_seen": 17911816,
      "step": 32320
    },
    {
      "epoch": 567.1061946902655,
      "grad_norm": 8.5625,
      "learning_rate": 4.407230755447245e-06,
      "loss": 1.0404,
      "num_input_tokens_seen": 17914632,
      "step": 32325
    },
    {
      "epoch": 567.1946902654868,
      "grad_norm": 8.1875,
      "learning_rate": 4.401665733793598e-06,
      "loss": 1.1111,
      "num_input_tokens_seen": 17917448,
      "step": 32330
    },
    {
      "epoch": 567.2831858407079,
      "grad_norm": 6.84375,
      "learning_rate": 4.3961038886619425e-06,
      "loss": 1.068,
      "num_input_tokens_seen": 17920248,
      "step": 32335
    },
    {
      "epoch": 567.3716814159292,
      "grad_norm": 8.625,
      "learning_rate": 4.39054522091e-06,
      "loss": 1.16,
      "num_input_tokens_seen": 17923000,
      "step": 32340
    },
    {
      "epoch": 567.4601769911504,
      "grad_norm": 8.8125,
      "learning_rate": 4.384989731394979e-06,
      "loss": 0.986,
      "num_input_tokens_seen": 17925672,
      "step": 32345
    },
    {
      "epoch": 567.5486725663717,
      "grad_norm": 8.5,
      "learning_rate": 4.379437420973598e-06,
      "loss": 0.9985,
      "num_input_tokens_seen": 17928584,
      "step": 32350
    },
    {
      "epoch": 567.637168141593,
      "grad_norm": 7.71875,
      "learning_rate": 4.373888290502107e-06,
      "loss": 1.0681,
      "num_input_tokens_seen": 17930920,
      "step": 32355
    },
    {
      "epoch": 567.7256637168142,
      "grad_norm": 10.5,
      "learning_rate": 4.36834234083624e-06,
      "loss": 1.059,
      "num_input_tokens_seen": 17933672,
      "step": 32360
    },
    {
      "epoch": 567.8141592920354,
      "grad_norm": 8.0625,
      "learning_rate": 4.362799572831258e-06,
      "loss": 0.8689,
      "num_input_tokens_seen": 17936280,
      "step": 32365
    },
    {
      "epoch": 567.9026548672566,
      "grad_norm": 9.25,
      "learning_rate": 4.35725998734193e-06,
      "loss": 1.2258,
      "num_input_tokens_seen": 17940008,
      "step": 32370
    },
    {
      "epoch": 567.9911504424779,
      "grad_norm": 9.8125,
      "learning_rate": 4.3517235852225195e-06,
      "loss": 0.9848,
      "num_input_tokens_seen": 17942904,
      "step": 32375
    },
    {
      "epoch": 568.070796460177,
      "grad_norm": 8.0,
      "learning_rate": 4.346190367326822e-06,
      "loss": 1.0048,
      "num_input_tokens_seen": 17945488,
      "step": 32380
    },
    {
      "epoch": 568.1592920353983,
      "grad_norm": 8.25,
      "learning_rate": 4.340660334508115e-06,
      "loss": 0.9752,
      "num_input_tokens_seen": 17948192,
      "step": 32385
    },
    {
      "epoch": 568.2477876106195,
      "grad_norm": 10.0625,
      "learning_rate": 4.335133487619206e-06,
      "loss": 1.0632,
      "num_input_tokens_seen": 17951216,
      "step": 32390
    },
    {
      "epoch": 568.3362831858407,
      "grad_norm": 7.6875,
      "learning_rate": 4.329609827512409e-06,
      "loss": 1.0239,
      "num_input_tokens_seen": 17953760,
      "step": 32395
    },
    {
      "epoch": 568.4247787610619,
      "grad_norm": 8.4375,
      "learning_rate": 4.324089355039531e-06,
      "loss": 1.1134,
      "num_input_tokens_seen": 17956288,
      "step": 32400
    },
    {
      "epoch": 568.4247787610619,
      "eval_loss": 1.0310944318771362,
      "eval_runtime": 0.9386,
      "eval_samples_per_second": 26.636,
      "eval_steps_per_second": 13.851,
      "num_input_tokens_seen": 17956288,
      "step": 32400
    },
    {
      "epoch": 568.5132743362832,
      "grad_norm": 6.625,
      "learning_rate": 4.3185720710519075e-06,
      "loss": 0.9194,
      "num_input_tokens_seen": 17958768,
      "step": 32405
    },
    {
      "epoch": 568.6017699115044,
      "grad_norm": 9.4375,
      "learning_rate": 4.3130579764003724e-06,
      "loss": 0.9397,
      "num_input_tokens_seen": 17961936,
      "step": 32410
    },
    {
      "epoch": 568.6902654867257,
      "grad_norm": 10.625,
      "learning_rate": 4.307547071935267e-06,
      "loss": 1.298,
      "num_input_tokens_seen": 17964496,
      "step": 32415
    },
    {
      "epoch": 568.7787610619469,
      "grad_norm": 10.3125,
      "learning_rate": 4.302039358506435e-06,
      "loss": 1.2178,
      "num_input_tokens_seen": 17967456,
      "step": 32420
    },
    {
      "epoch": 568.8672566371681,
      "grad_norm": 7.5,
      "learning_rate": 4.296534836963245e-06,
      "loss": 0.9479,
      "num_input_tokens_seen": 17970240,
      "step": 32425
    },
    {
      "epoch": 568.9557522123894,
      "grad_norm": 7.84375,
      "learning_rate": 4.291033508154555e-06,
      "loss": 1.1188,
      "num_input_tokens_seen": 17973040,
      "step": 32430
    },
    {
      "epoch": 569.0353982300885,
      "grad_norm": 5.90625,
      "learning_rate": 4.285535372928748e-06,
      "loss": 1.1095,
      "num_input_tokens_seen": 17975216,
      "step": 32435
    },
    {
      "epoch": 569.1238938053098,
      "grad_norm": 8.125,
      "learning_rate": 4.280040432133695e-06,
      "loss": 1.0019,
      "num_input_tokens_seen": 17978336,
      "step": 32440
    },
    {
      "epoch": 569.212389380531,
      "grad_norm": 9.1875,
      "learning_rate": 4.274548686616789e-06,
      "loss": 1.2352,
      "num_input_tokens_seen": 17980672,
      "step": 32445
    },
    {
      "epoch": 569.3008849557522,
      "grad_norm": 12.5625,
      "learning_rate": 4.2690601372249364e-06,
      "loss": 0.9788,
      "num_input_tokens_seen": 17983712,
      "step": 32450
    },
    {
      "epoch": 569.3893805309734,
      "grad_norm": 7.3125,
      "learning_rate": 4.263574784804525e-06,
      "loss": 0.9491,
      "num_input_tokens_seen": 17986272,
      "step": 32455
    },
    {
      "epoch": 569.4778761061947,
      "grad_norm": 8.375,
      "learning_rate": 4.258092630201479e-06,
      "loss": 1.0045,
      "num_input_tokens_seen": 17989152,
      "step": 32460
    },
    {
      "epoch": 569.566371681416,
      "grad_norm": 6.15625,
      "learning_rate": 4.252613674261202e-06,
      "loss": 0.9323,
      "num_input_tokens_seen": 17992048,
      "step": 32465
    },
    {
      "epoch": 569.6548672566372,
      "grad_norm": 8.5,
      "learning_rate": 4.2471379178286224e-06,
      "loss": 1.142,
      "num_input_tokens_seen": 17994928,
      "step": 32470
    },
    {
      "epoch": 569.7433628318585,
      "grad_norm": 6.125,
      "learning_rate": 4.241665361748181e-06,
      "loss": 1.0469,
      "num_input_tokens_seen": 17998240,
      "step": 32475
    },
    {
      "epoch": 569.8318584070796,
      "grad_norm": 6.8125,
      "learning_rate": 4.2361960068637994e-06,
      "loss": 1.0398,
      "num_input_tokens_seen": 18000864,
      "step": 32480
    },
    {
      "epoch": 569.9203539823009,
      "grad_norm": 11.6875,
      "learning_rate": 4.230729854018933e-06,
      "loss": 1.2882,
      "num_input_tokens_seen": 18003328,
      "step": 32485
    },
    {
      "epoch": 570.0,
      "grad_norm": 4.0625,
      "learning_rate": 4.225266904056521e-06,
      "loss": 0.8029,
      "num_input_tokens_seen": 18005936,
      "step": 32490
    },
    {
      "epoch": 570.0884955752213,
      "grad_norm": 9.25,
      "learning_rate": 4.21980715781903e-06,
      "loss": 1.1435,
      "num_input_tokens_seen": 18008688,
      "step": 32495
    },
    {
      "epoch": 570.1769911504425,
      "grad_norm": 7.65625,
      "learning_rate": 4.214350616148416e-06,
      "loss": 1.1206,
      "num_input_tokens_seen": 18011408,
      "step": 32500
    },
    {
      "epoch": 570.2654867256637,
      "grad_norm": 6.9375,
      "learning_rate": 4.20889727988614e-06,
      "loss": 1.122,
      "num_input_tokens_seen": 18014400,
      "step": 32505
    },
    {
      "epoch": 570.3539823008849,
      "grad_norm": 9.6875,
      "learning_rate": 4.20344714987318e-06,
      "loss": 1.0982,
      "num_input_tokens_seen": 18017136,
      "step": 32510
    },
    {
      "epoch": 570.4424778761062,
      "grad_norm": 7.625,
      "learning_rate": 4.198000226950022e-06,
      "loss": 1.0062,
      "num_input_tokens_seen": 18019680,
      "step": 32515
    },
    {
      "epoch": 570.5309734513274,
      "grad_norm": 7.0625,
      "learning_rate": 4.192556511956635e-06,
      "loss": 0.9973,
      "num_input_tokens_seen": 18022384,
      "step": 32520
    },
    {
      "epoch": 570.6194690265487,
      "grad_norm": 8.4375,
      "learning_rate": 4.18711600573252e-06,
      "loss": 1.2099,
      "num_input_tokens_seen": 18024800,
      "step": 32525
    },
    {
      "epoch": 570.70796460177,
      "grad_norm": 8.9375,
      "learning_rate": 4.181678709116671e-06,
      "loss": 0.9434,
      "num_input_tokens_seen": 18027936,
      "step": 32530
    },
    {
      "epoch": 570.7964601769911,
      "grad_norm": 7.0,
      "learning_rate": 4.1762446229475785e-06,
      "loss": 0.8805,
      "num_input_tokens_seen": 18030944,
      "step": 32535
    },
    {
      "epoch": 570.8849557522124,
      "grad_norm": 8.6875,
      "learning_rate": 4.17081374806326e-06,
      "loss": 0.9993,
      "num_input_tokens_seen": 18033872,
      "step": 32540
    },
    {
      "epoch": 570.9734513274336,
      "grad_norm": 9.125,
      "learning_rate": 4.165386085301212e-06,
      "loss": 0.9419,
      "num_input_tokens_seen": 18037040,
      "step": 32545
    },
    {
      "epoch": 571.0530973451328,
      "grad_norm": 8.625,
      "learning_rate": 4.1599616354984525e-06,
      "loss": 1.169,
      "num_input_tokens_seen": 18039696,
      "step": 32550
    },
    {
      "epoch": 571.141592920354,
      "grad_norm": 9.5625,
      "learning_rate": 4.154540399491508e-06,
      "loss": 1.2334,
      "num_input_tokens_seen": 18042096,
      "step": 32555
    },
    {
      "epoch": 571.2300884955753,
      "grad_norm": 8.125,
      "learning_rate": 4.149122378116394e-06,
      "loss": 0.76,
      "num_input_tokens_seen": 18045184,
      "step": 32560
    },
    {
      "epoch": 571.3185840707964,
      "grad_norm": 9.75,
      "learning_rate": 4.14370757220863e-06,
      "loss": 1.2146,
      "num_input_tokens_seen": 18047728,
      "step": 32565
    },
    {
      "epoch": 571.4070796460177,
      "grad_norm": 9.0,
      "learning_rate": 4.138295982603263e-06,
      "loss": 1.1152,
      "num_input_tokens_seen": 18050320,
      "step": 32570
    },
    {
      "epoch": 571.4955752212389,
      "grad_norm": 6.75,
      "learning_rate": 4.132887610134814e-06,
      "loss": 1.0184,
      "num_input_tokens_seen": 18052912,
      "step": 32575
    },
    {
      "epoch": 571.5840707964602,
      "grad_norm": 8.8125,
      "learning_rate": 4.127482455637335e-06,
      "loss": 1.0755,
      "num_input_tokens_seen": 18055760,
      "step": 32580
    },
    {
      "epoch": 571.6725663716815,
      "grad_norm": 8.6875,
      "learning_rate": 4.1220805199443545e-06,
      "loss": 1.059,
      "num_input_tokens_seen": 18058816,
      "step": 32585
    },
    {
      "epoch": 571.7610619469026,
      "grad_norm": 9.25,
      "learning_rate": 4.116681803888925e-06,
      "loss": 1.2061,
      "num_input_tokens_seen": 18061376,
      "step": 32590
    },
    {
      "epoch": 571.8495575221239,
      "grad_norm": 7.21875,
      "learning_rate": 4.111286308303605e-06,
      "loss": 0.878,
      "num_input_tokens_seen": 18063872,
      "step": 32595
    },
    {
      "epoch": 571.9380530973451,
      "grad_norm": 6.6875,
      "learning_rate": 4.105894034020433e-06,
      "loss": 1.0489,
      "num_input_tokens_seen": 18066176,
      "step": 32600
    },
    {
      "epoch": 571.9380530973451,
      "eval_loss": 1.0310944318771362,
      "eval_runtime": 0.9376,
      "eval_samples_per_second": 26.663,
      "eval_steps_per_second": 13.865,
      "num_input_tokens_seen": 18066176,
      "step": 32600
    },
    {
      "epoch": 572.0176991150443,
      "grad_norm": 6.78125,
      "learning_rate": 4.100504981870975e-06,
      "loss": 0.9346,
      "num_input_tokens_seen": 18069256,
      "step": 32605
    },
    {
      "epoch": 572.1061946902655,
      "grad_norm": 6.40625,
      "learning_rate": 4.0951191526862915e-06,
      "loss": 0.9679,
      "num_input_tokens_seen": 18072344,
      "step": 32610
    },
    {
      "epoch": 572.1946902654868,
      "grad_norm": 8.1875,
      "learning_rate": 4.089736547296938e-06,
      "loss": 1.0441,
      "num_input_tokens_seen": 18075000,
      "step": 32615
    },
    {
      "epoch": 572.2831858407079,
      "grad_norm": 6.5625,
      "learning_rate": 4.08435716653299e-06,
      "loss": 1.0753,
      "num_input_tokens_seen": 18077624,
      "step": 32620
    },
    {
      "epoch": 572.3716814159292,
      "grad_norm": 8.5,
      "learning_rate": 4.0789810112240005e-06,
      "loss": 1.1312,
      "num_input_tokens_seen": 18080568,
      "step": 32625
    },
    {
      "epoch": 572.4601769911504,
      "grad_norm": 13.125,
      "learning_rate": 4.073608082199057e-06,
      "loss": 1.0143,
      "num_input_tokens_seen": 18083304,
      "step": 32630
    },
    {
      "epoch": 572.5486725663717,
      "grad_norm": 8.25,
      "learning_rate": 4.068238380286718e-06,
      "loss": 1.0696,
      "num_input_tokens_seen": 18086296,
      "step": 32635
    },
    {
      "epoch": 572.637168141593,
      "grad_norm": 8.125,
      "learning_rate": 4.062871906315072e-06,
      "loss": 0.9834,
      "num_input_tokens_seen": 18088744,
      "step": 32640
    },
    {
      "epoch": 572.7256637168142,
      "grad_norm": 12.4375,
      "learning_rate": 4.057508661111686e-06,
      "loss": 1.194,
      "num_input_tokens_seen": 18091768,
      "step": 32645
    },
    {
      "epoch": 572.8141592920354,
      "grad_norm": 7.34375,
      "learning_rate": 4.052148645503648e-06,
      "loss": 1.018,
      "num_input_tokens_seen": 18094504,
      "step": 32650
    },
    {
      "epoch": 572.9026548672566,
      "grad_norm": 7.34375,
      "learning_rate": 4.046791860317531e-06,
      "loss": 0.8979,
      "num_input_tokens_seen": 18097352,
      "step": 32655
    },
    {
      "epoch": 572.9911504424779,
      "grad_norm": 9.0,
      "learning_rate": 4.041438306379431e-06,
      "loss": 1.1533,
      "num_input_tokens_seen": 18100328,
      "step": 32660
    },
    {
      "epoch": 573.070796460177,
      "grad_norm": 8.0625,
      "learning_rate": 4.036087984514916e-06,
      "loss": 1.0558,
      "num_input_tokens_seen": 18102664,
      "step": 32665
    },
    {
      "epoch": 573.1592920353983,
      "grad_norm": 9.8125,
      "learning_rate": 4.030740895549084e-06,
      "loss": 1.144,
      "num_input_tokens_seen": 18105384,
      "step": 32670
    },
    {
      "epoch": 573.2477876106195,
      "grad_norm": 9.75,
      "learning_rate": 4.025397040306531e-06,
      "loss": 1.0159,
      "num_input_tokens_seen": 18108568,
      "step": 32675
    },
    {
      "epoch": 573.3362831858407,
      "grad_norm": 9.25,
      "learning_rate": 4.0200564196113285e-06,
      "loss": 1.138,
      "num_input_tokens_seen": 18111096,
      "step": 32680
    },
    {
      "epoch": 573.4247787610619,
      "grad_norm": 10.625,
      "learning_rate": 4.014719034287079e-06,
      "loss": 1.0201,
      "num_input_tokens_seen": 18113960,
      "step": 32685
    },
    {
      "epoch": 573.5132743362832,
      "grad_norm": 8.0,
      "learning_rate": 4.0093848851568775e-06,
      "loss": 1.1879,
      "num_input_tokens_seen": 18117160,
      "step": 32690
    },
    {
      "epoch": 573.6017699115044,
      "grad_norm": 8.375,
      "learning_rate": 4.004053973043304e-06,
      "loss": 1.0317,
      "num_input_tokens_seen": 18119576,
      "step": 32695
    },
    {
      "epoch": 573.6902654867257,
      "grad_norm": 7.59375,
      "learning_rate": 3.998726298768465e-06,
      "loss": 1.043,
      "num_input_tokens_seen": 18121800,
      "step": 32700
    },
    {
      "epoch": 573.7787610619469,
      "grad_norm": 8.3125,
      "learning_rate": 3.99340186315395e-06,
      "loss": 1.0551,
      "num_input_tokens_seen": 18125048,
      "step": 32705
    },
    {
      "epoch": 573.8672566371681,
      "grad_norm": 8.0625,
      "learning_rate": 3.988080667020849e-06,
      "loss": 0.9608,
      "num_input_tokens_seen": 18128360,
      "step": 32710
    },
    {
      "epoch": 573.9557522123894,
      "grad_norm": 8.75,
      "learning_rate": 3.982762711189766e-06,
      "loss": 0.7974,
      "num_input_tokens_seen": 18130840,
      "step": 32715
    },
    {
      "epoch": 574.0353982300885,
      "grad_norm": 6.84375,
      "learning_rate": 3.977447996480785e-06,
      "loss": 1.1275,
      "num_input_tokens_seen": 18132992,
      "step": 32720
    },
    {
      "epoch": 574.1238938053098,
      "grad_norm": 8.0,
      "learning_rate": 3.97213652371351e-06,
      "loss": 1.1098,
      "num_input_tokens_seen": 18136064,
      "step": 32725
    },
    {
      "epoch": 574.212389380531,
      "grad_norm": 9.6875,
      "learning_rate": 3.966828293707042e-06,
      "loss": 1.161,
      "num_input_tokens_seen": 18139200,
      "step": 32730
    },
    {
      "epoch": 574.3008849557522,
      "grad_norm": 8.9375,
      "learning_rate": 3.961523307279963e-06,
      "loss": 1.0752,
      "num_input_tokens_seen": 18141776,
      "step": 32735
    },
    {
      "epoch": 574.3893805309734,
      "grad_norm": 8.25,
      "learning_rate": 3.956221565250382e-06,
      "loss": 1.0318,
      "num_input_tokens_seen": 18145104,
      "step": 32740
    },
    {
      "epoch": 574.4778761061947,
      "grad_norm": 8.8125,
      "learning_rate": 3.950923068435883e-06,
      "loss": 1.0101,
      "num_input_tokens_seen": 18147888,
      "step": 32745
    },
    {
      "epoch": 574.566371681416,
      "grad_norm": 8.8125,
      "learning_rate": 3.945627817653566e-06,
      "loss": 1.0015,
      "num_input_tokens_seen": 18150768,
      "step": 32750
    },
    {
      "epoch": 574.6548672566372,
      "grad_norm": 6.71875,
      "learning_rate": 3.9403358137200335e-06,
      "loss": 1.1267,
      "num_input_tokens_seen": 18153312,
      "step": 32755
    },
    {
      "epoch": 574.7433628318585,
      "grad_norm": 12.8125,
      "learning_rate": 3.9350470574513605e-06,
      "loss": 0.9311,
      "num_input_tokens_seen": 18156224,
      "step": 32760
    },
    {
      "epoch": 574.8318584070796,
      "grad_norm": 7.78125,
      "learning_rate": 3.9297615496631525e-06,
      "loss": 1.0454,
      "num_input_tokens_seen": 18158832,
      "step": 32765
    },
    {
      "epoch": 574.9203539823009,
      "grad_norm": 6.8125,
      "learning_rate": 3.924479291170505e-06,
      "loss": 1.0706,
      "num_input_tokens_seen": 18161520,
      "step": 32770
    },
    {
      "epoch": 575.0,
      "grad_norm": 3.640625,
      "learning_rate": 3.919200282788002e-06,
      "loss": 0.8576,
      "num_input_tokens_seen": 18163744,
      "step": 32775
    },
    {
      "epoch": 575.0884955752213,
      "grad_norm": 7.90625,
      "learning_rate": 3.913924525329726e-06,
      "loss": 1.0601,
      "num_input_tokens_seen": 18166336,
      "step": 32780
    },
    {
      "epoch": 575.1769911504425,
      "grad_norm": 8.4375,
      "learning_rate": 3.908652019609279e-06,
      "loss": 1.0664,
      "num_input_tokens_seen": 18169104,
      "step": 32785
    },
    {
      "epoch": 575.2654867256637,
      "grad_norm": 8.5625,
      "learning_rate": 3.9033827664397364e-06,
      "loss": 0.9992,
      "num_input_tokens_seen": 18172128,
      "step": 32790
    },
    {
      "epoch": 575.3539823008849,
      "grad_norm": 8.1875,
      "learning_rate": 3.898116766633694e-06,
      "loss": 1.1606,
      "num_input_tokens_seen": 18174704,
      "step": 32795
    },
    {
      "epoch": 575.4424778761062,
      "grad_norm": 8.5,
      "learning_rate": 3.8928540210032225e-06,
      "loss": 1.0341,
      "num_input_tokens_seen": 18177520,
      "step": 32800
    },
    {
      "epoch": 575.4424778761062,
      "eval_loss": 1.0310944318771362,
      "eval_runtime": 0.9372,
      "eval_samples_per_second": 26.675,
      "eval_steps_per_second": 13.871,
      "num_input_tokens_seen": 18177520,
      "step": 32800
    },
    {
      "epoch": 575.5309734513274,
      "grad_norm": 7.15625,
      "learning_rate": 3.887594530359909e-06,
      "loss": 1.1444,
      "num_input_tokens_seen": 18180560,
      "step": 32805
    },
    {
      "epoch": 575.6194690265487,
      "grad_norm": 9.0,
      "learning_rate": 3.88233829551484e-06,
      "loss": 1.0999,
      "num_input_tokens_seen": 18183440,
      "step": 32810
    },
    {
      "epoch": 575.70796460177,
      "grad_norm": 8.25,
      "learning_rate": 3.877085317278581e-06,
      "loss": 0.8837,
      "num_input_tokens_seen": 18186112,
      "step": 32815
    },
    {
      "epoch": 575.7964601769911,
      "grad_norm": 12.1875,
      "learning_rate": 3.87183559646122e-06,
      "loss": 1.1859,
      "num_input_tokens_seen": 18188416,
      "step": 32820
    },
    {
      "epoch": 575.8849557522124,
      "grad_norm": 8.6875,
      "learning_rate": 3.866589133872317e-06,
      "loss": 0.9896,
      "num_input_tokens_seen": 18191600,
      "step": 32825
    },
    {
      "epoch": 575.9734513274336,
      "grad_norm": 7.84375,
      "learning_rate": 3.861345930320948e-06,
      "loss": 0.9502,
      "num_input_tokens_seen": 18194976,
      "step": 32830
    },
    {
      "epoch": 576.0530973451328,
      "grad_norm": 6.625,
      "learning_rate": 3.856105986615688e-06,
      "loss": 0.8856,
      "num_input_tokens_seen": 18196944,
      "step": 32835
    },
    {
      "epoch": 576.141592920354,
      "grad_norm": 12.125,
      "learning_rate": 3.850869303564589e-06,
      "loss": 1.2106,
      "num_input_tokens_seen": 18199616,
      "step": 32840
    },
    {
      "epoch": 576.2300884955753,
      "grad_norm": 7.1875,
      "learning_rate": 3.845635881975226e-06,
      "loss": 1.0291,
      "num_input_tokens_seen": 18201968,
      "step": 32845
    },
    {
      "epoch": 576.3185840707964,
      "grad_norm": 8.625,
      "learning_rate": 3.840405722654647e-06,
      "loss": 1.0139,
      "num_input_tokens_seen": 18205616,
      "step": 32850
    },
    {
      "epoch": 576.4070796460177,
      "grad_norm": 8.25,
      "learning_rate": 3.835178826409419e-06,
      "loss": 1.1219,
      "num_input_tokens_seen": 18208912,
      "step": 32855
    },
    {
      "epoch": 576.4955752212389,
      "grad_norm": 9.0625,
      "learning_rate": 3.8299551940455895e-06,
      "loss": 1.1379,
      "num_input_tokens_seen": 18211392,
      "step": 32860
    },
    {
      "epoch": 576.5840707964602,
      "grad_norm": 8.4375,
      "learning_rate": 3.824734826368703e-06,
      "loss": 0.9574,
      "num_input_tokens_seen": 18213808,
      "step": 32865
    },
    {
      "epoch": 576.6725663716815,
      "grad_norm": 9.8125,
      "learning_rate": 3.819517724183813e-06,
      "loss": 0.9253,
      "num_input_tokens_seen": 18216704,
      "step": 32870
    },
    {
      "epoch": 576.7610619469026,
      "grad_norm": 7.28125,
      "learning_rate": 3.8143038882954648e-06,
      "loss": 0.9763,
      "num_input_tokens_seen": 18219600,
      "step": 32875
    },
    {
      "epoch": 576.8495575221239,
      "grad_norm": 7.90625,
      "learning_rate": 3.8090933195076867e-06,
      "loss": 0.9381,
      "num_input_tokens_seen": 18222784,
      "step": 32880
    },
    {
      "epoch": 576.9380530973451,
      "grad_norm": 8.25,
      "learning_rate": 3.8038860186240198e-06,
      "loss": 0.9763,
      "num_input_tokens_seen": 18225408,
      "step": 32885
    },
    {
      "epoch": 577.0176991150443,
      "grad_norm": 6.34375,
      "learning_rate": 3.7986819864475026e-06,
      "loss": 1.3178,
      "num_input_tokens_seen": 18227512,
      "step": 32890
    },
    {
      "epoch": 577.1061946902655,
      "grad_norm": 6.21875,
      "learning_rate": 3.793481223780651e-06,
      "loss": 0.8746,
      "num_input_tokens_seen": 18230008,
      "step": 32895
    },
    {
      "epoch": 577.1946902654868,
      "grad_norm": 9.25,
      "learning_rate": 3.788283731425496e-06,
      "loss": 1.0682,
      "num_input_tokens_seen": 18233240,
      "step": 32900
    },
    {
      "epoch": 577.2831858407079,
      "grad_norm": 6.6875,
      "learning_rate": 3.7830895101835488e-06,
      "loss": 0.9168,
      "num_input_tokens_seen": 18235976,
      "step": 32905
    },
    {
      "epoch": 577.3716814159292,
      "grad_norm": 6.90625,
      "learning_rate": 3.7778985608558274e-06,
      "loss": 1.1077,
      "num_input_tokens_seen": 18238216,
      "step": 32910
    },
    {
      "epoch": 577.4601769911504,
      "grad_norm": 8.0,
      "learning_rate": 3.7727108842428443e-06,
      "loss": 1.0852,
      "num_input_tokens_seen": 18241032,
      "step": 32915
    },
    {
      "epoch": 577.5486725663717,
      "grad_norm": 8.75,
      "learning_rate": 3.7675264811446065e-06,
      "loss": 1.1001,
      "num_input_tokens_seen": 18244040,
      "step": 32920
    },
    {
      "epoch": 577.637168141593,
      "grad_norm": 8.5,
      "learning_rate": 3.7623453523605994e-06,
      "loss": 1.1278,
      "num_input_tokens_seen": 18246904,
      "step": 32925
    },
    {
      "epoch": 577.7256637168142,
      "grad_norm": 10.25,
      "learning_rate": 3.757167498689834e-06,
      "loss": 1.2109,
      "num_input_tokens_seen": 18250312,
      "step": 32930
    },
    {
      "epoch": 577.8141592920354,
      "grad_norm": 8.5625,
      "learning_rate": 3.7519929209307914e-06,
      "loss": 1.0884,
      "num_input_tokens_seen": 18253240,
      "step": 32935
    },
    {
      "epoch": 577.9026548672566,
      "grad_norm": 8.1875,
      "learning_rate": 3.746821619881463e-06,
      "loss": 1.0125,
      "num_input_tokens_seen": 18255992,
      "step": 32940
    },
    {
      "epoch": 577.9911504424779,
      "grad_norm": 8.375,
      "learning_rate": 3.74165359633932e-06,
      "loss": 1.0,
      "num_input_tokens_seen": 18258840,
      "step": 32945
    },
    {
      "epoch": 578.070796460177,
      "grad_norm": 7.125,
      "learning_rate": 3.736488851101341e-06,
      "loss": 0.9028,
      "num_input_tokens_seen": 18261208,
      "step": 32950
    },
    {
      "epoch": 578.1592920353983,
      "grad_norm": 7.59375,
      "learning_rate": 3.7313273849640035e-06,
      "loss": 1.1801,
      "num_input_tokens_seen": 18263768,
      "step": 32955
    },
    {
      "epoch": 578.2477876106195,
      "grad_norm": 7.9375,
      "learning_rate": 3.7261691987232533e-06,
      "loss": 1.0323,
      "num_input_tokens_seen": 18266536,
      "step": 32960
    },
    {
      "epoch": 578.3362831858407,
      "grad_norm": 7.78125,
      "learning_rate": 3.7210142931745575e-06,
      "loss": 0.8529,
      "num_input_tokens_seen": 18269880,
      "step": 32965
    },
    {
      "epoch": 578.4247787610619,
      "grad_norm": 8.875,
      "learning_rate": 3.7158626691128712e-06,
      "loss": 1.0983,
      "num_input_tokens_seen": 18272712,
      "step": 32970
    },
    {
      "epoch": 578.5132743362832,
      "grad_norm": 10.25,
      "learning_rate": 3.710714327332629e-06,
      "loss": 1.1363,
      "num_input_tokens_seen": 18275096,
      "step": 32975
    },
    {
      "epoch": 578.6017699115044,
      "grad_norm": 6.90625,
      "learning_rate": 3.7055692686277815e-06,
      "loss": 1.0426,
      "num_input_tokens_seen": 18277656,
      "step": 32980
    },
    {
      "epoch": 578.6902654867257,
      "grad_norm": 9.1875,
      "learning_rate": 3.70042749379175e-06,
      "loss": 1.2552,
      "num_input_tokens_seen": 18280568,
      "step": 32985
    },
    {
      "epoch": 578.7787610619469,
      "grad_norm": 5.1875,
      "learning_rate": 3.6952890036174693e-06,
      "loss": 0.8458,
      "num_input_tokens_seen": 18283720,
      "step": 32990
    },
    {
      "epoch": 578.8672566371681,
      "grad_norm": 6.84375,
      "learning_rate": 3.690153798897353e-06,
      "loss": 1.064,
      "num_input_tokens_seen": 18286264,
      "step": 32995
    },
    {
      "epoch": 578.9557522123894,
      "grad_norm": 12.4375,
      "learning_rate": 3.6850218804233225e-06,
      "loss": 1.2816,
      "num_input_tokens_seen": 18289064,
      "step": 33000
    },
    {
      "epoch": 578.9557522123894,
      "eval_loss": 1.0310944318771362,
      "eval_runtime": 0.9381,
      "eval_samples_per_second": 26.651,
      "eval_steps_per_second": 13.858,
      "num_input_tokens_seen": 18289064,
      "step": 33000
    },
    {
      "epoch": 579.0353982300885,
      "grad_norm": 5.71875,
      "learning_rate": 3.679893248986779e-06,
      "loss": 0.9358,
      "num_input_tokens_seen": 18291328,
      "step": 33005
    },
    {
      "epoch": 579.1238938053098,
      "grad_norm": 8.375,
      "learning_rate": 3.6747679053786147e-06,
      "loss": 1.0211,
      "num_input_tokens_seen": 18294256,
      "step": 33010
    },
    {
      "epoch": 579.212389380531,
      "grad_norm": 8.4375,
      "learning_rate": 3.669645850389228e-06,
      "loss": 1.2921,
      "num_input_tokens_seen": 18296704,
      "step": 33015
    },
    {
      "epoch": 579.3008849557522,
      "grad_norm": 10.1875,
      "learning_rate": 3.664527084808514e-06,
      "loss": 1.0789,
      "num_input_tokens_seen": 18299616,
      "step": 33020
    },
    {
      "epoch": 579.3893805309734,
      "grad_norm": 7.09375,
      "learning_rate": 3.6594116094258337e-06,
      "loss": 1.0307,
      "num_input_tokens_seen": 18302576,
      "step": 33025
    },
    {
      "epoch": 579.4778761061947,
      "grad_norm": 7.4375,
      "learning_rate": 3.6542994250300665e-06,
      "loss": 0.9835,
      "num_input_tokens_seen": 18304896,
      "step": 33030
    },
    {
      "epoch": 579.566371681416,
      "grad_norm": 8.875,
      "learning_rate": 3.6491905324095825e-06,
      "loss": 1.0205,
      "num_input_tokens_seen": 18307728,
      "step": 33035
    },
    {
      "epoch": 579.6548672566372,
      "grad_norm": 8.0,
      "learning_rate": 3.644084932352221e-06,
      "loss": 1.0759,
      "num_input_tokens_seen": 18310288,
      "step": 33040
    },
    {
      "epoch": 579.7433628318585,
      "grad_norm": 7.84375,
      "learning_rate": 3.6389826256453457e-06,
      "loss": 0.8324,
      "num_input_tokens_seen": 18313440,
      "step": 33045
    },
    {
      "epoch": 579.8318584070796,
      "grad_norm": 7.96875,
      "learning_rate": 3.633883613075781e-06,
      "loss": 1.1338,
      "num_input_tokens_seen": 18316080,
      "step": 33050
    },
    {
      "epoch": 579.9203539823009,
      "grad_norm": 6.8125,
      "learning_rate": 3.6287878954298693e-06,
      "loss": 1.1327,
      "num_input_tokens_seen": 18319152,
      "step": 33055
    },
    {
      "epoch": 580.0,
      "grad_norm": 9.0,
      "learning_rate": 3.6236954734934354e-06,
      "loss": 1.1487,
      "num_input_tokens_seen": 18321640,
      "step": 33060
    },
    {
      "epoch": 580.0884955752213,
      "grad_norm": 7.6875,
      "learning_rate": 3.618606348051784e-06,
      "loss": 0.9349,
      "num_input_tokens_seen": 18324056,
      "step": 33065
    },
    {
      "epoch": 580.1769911504425,
      "grad_norm": 6.8125,
      "learning_rate": 3.6135205198897376e-06,
      "loss": 0.9561,
      "num_input_tokens_seen": 18326808,
      "step": 33070
    },
    {
      "epoch": 580.2654867256637,
      "grad_norm": 7.78125,
      "learning_rate": 3.6084379897915854e-06,
      "loss": 1.1434,
      "num_input_tokens_seen": 18329192,
      "step": 33075
    },
    {
      "epoch": 580.3539823008849,
      "grad_norm": 9.6875,
      "learning_rate": 3.6033587585411115e-06,
      "loss": 0.9748,
      "num_input_tokens_seen": 18332360,
      "step": 33080
    },
    {
      "epoch": 580.4424778761062,
      "grad_norm": 7.40625,
      "learning_rate": 3.5982828269216117e-06,
      "loss": 1.0253,
      "num_input_tokens_seen": 18335080,
      "step": 33085
    },
    {
      "epoch": 580.5309734513274,
      "grad_norm": 7.875,
      "learning_rate": 3.593210195715843e-06,
      "loss": 1.1301,
      "num_input_tokens_seen": 18337576,
      "step": 33090
    },
    {
      "epoch": 580.6194690265487,
      "grad_norm": 9.0,
      "learning_rate": 3.5881408657060773e-06,
      "loss": 1.1881,
      "num_input_tokens_seen": 18340584,
      "step": 33095
    },
    {
      "epoch": 580.70796460177,
      "grad_norm": 6.21875,
      "learning_rate": 3.583074837674075e-06,
      "loss": 0.804,
      "num_input_tokens_seen": 18343384,
      "step": 33100
    },
    {
      "epoch": 580.7964601769911,
      "grad_norm": 8.5,
      "learning_rate": 3.578012112401069e-06,
      "loss": 1.0113,
      "num_input_tokens_seen": 18346248,
      "step": 33105
    },
    {
      "epoch": 580.8849557522124,
      "grad_norm": 9.4375,
      "learning_rate": 3.5729526906677996e-06,
      "loss": 1.1089,
      "num_input_tokens_seen": 18349016,
      "step": 33110
    },
    {
      "epoch": 580.9734513274336,
      "grad_norm": 16.625,
      "learning_rate": 3.5678965732545007e-06,
      "loss": 1.1801,
      "num_input_tokens_seen": 18351912,
      "step": 33115
    },
    {
      "epoch": 581.0530973451328,
      "grad_norm": 8.6875,
      "learning_rate": 3.562843760940876e-06,
      "loss": 1.2034,
      "num_input_tokens_seen": 18354064,
      "step": 33120
    },
    {
      "epoch": 581.141592920354,
      "grad_norm": 9.625,
      "learning_rate": 3.5577942545061473e-06,
      "loss": 0.9206,
      "num_input_tokens_seen": 18356880,
      "step": 33125
    },
    {
      "epoch": 581.2300884955753,
      "grad_norm": 9.3125,
      "learning_rate": 3.5527480547289967e-06,
      "loss": 0.9849,
      "num_input_tokens_seen": 18359648,
      "step": 33130
    },
    {
      "epoch": 581.3185840707964,
      "grad_norm": 7.53125,
      "learning_rate": 3.547705162387624e-06,
      "loss": 1.0978,
      "num_input_tokens_seen": 18362288,
      "step": 33135
    },
    {
      "epoch": 581.4070796460177,
      "grad_norm": 7.5,
      "learning_rate": 3.542665578259699e-06,
      "loss": 1.1361,
      "num_input_tokens_seen": 18364960,
      "step": 33140
    },
    {
      "epoch": 581.4955752212389,
      "grad_norm": 9.6875,
      "learning_rate": 3.5376293031223945e-06,
      "loss": 1.0252,
      "num_input_tokens_seen": 18367488,
      "step": 33145
    },
    {
      "epoch": 581.5840707964602,
      "grad_norm": 9.375,
      "learning_rate": 3.5325963377523614e-06,
      "loss": 0.8836,
      "num_input_tokens_seen": 18370176,
      "step": 33150
    },
    {
      "epoch": 581.6725663716815,
      "grad_norm": 11.4375,
      "learning_rate": 3.5275666829257536e-06,
      "loss": 1.2549,
      "num_input_tokens_seen": 18373344,
      "step": 33155
    },
    {
      "epoch": 581.7610619469026,
      "grad_norm": 11.125,
      "learning_rate": 3.5225403394181955e-06,
      "loss": 1.0795,
      "num_input_tokens_seen": 18375888,
      "step": 33160
    },
    {
      "epoch": 581.8495575221239,
      "grad_norm": 7.5625,
      "learning_rate": 3.517517308004828e-06,
      "loss": 1.0479,
      "num_input_tokens_seen": 18379152,
      "step": 33165
    },
    {
      "epoch": 581.9380530973451,
      "grad_norm": 9.0625,
      "learning_rate": 3.512497589460251e-06,
      "loss": 1.0277,
      "num_input_tokens_seen": 18381824,
      "step": 33170
    },
    {
      "epoch": 582.0176991150443,
      "grad_norm": 6.625,
      "learning_rate": 3.5074811845585727e-06,
      "loss": 0.8659,
      "num_input_tokens_seen": 18384648,
      "step": 33175
    },
    {
      "epoch": 582.1061946902655,
      "grad_norm": 10.0,
      "learning_rate": 3.5024680940733937e-06,
      "loss": 1.0421,
      "num_input_tokens_seen": 18387320,
      "step": 33180
    },
    {
      "epoch": 582.1946902654868,
      "grad_norm": 9.5625,
      "learning_rate": 3.4974583187777852e-06,
      "loss": 1.0771,
      "num_input_tokens_seen": 18390376,
      "step": 33185
    },
    {
      "epoch": 582.2831858407079,
      "grad_norm": 7.3125,
      "learning_rate": 3.4924518594443204e-06,
      "loss": 1.2595,
      "num_input_tokens_seen": 18392984,
      "step": 33190
    },
    {
      "epoch": 582.3716814159292,
      "grad_norm": 6.53125,
      "learning_rate": 3.4874487168450682e-06,
      "loss": 0.9765,
      "num_input_tokens_seen": 18396008,
      "step": 33195
    },
    {
      "epoch": 582.4601769911504,
      "grad_norm": 7.96875,
      "learning_rate": 3.482448891751558e-06,
      "loss": 1.219,
      "num_input_tokens_seen": 18398888,
      "step": 33200
    },
    {
      "epoch": 582.4601769911504,
      "eval_loss": 1.0310944318771362,
      "eval_runtime": 0.9365,
      "eval_samples_per_second": 26.696,
      "eval_steps_per_second": 13.882,
      "num_input_tokens_seen": 18398888,
      "step": 33200
    },
    {
      "epoch": 582.5486725663717,
      "grad_norm": 6.71875,
      "learning_rate": 3.477452384934843e-06,
      "loss": 0.9783,
      "num_input_tokens_seen": 18401624,
      "step": 33205
    },
    {
      "epoch": 582.637168141593,
      "grad_norm": 8.8125,
      "learning_rate": 3.472459197165434e-06,
      "loss": 0.985,
      "num_input_tokens_seen": 18404264,
      "step": 33210
    },
    {
      "epoch": 582.7256637168142,
      "grad_norm": 8.0625,
      "learning_rate": 3.4674693292133518e-06,
      "loss": 1.1111,
      "num_input_tokens_seen": 18407112,
      "step": 33215
    },
    {
      "epoch": 582.8141592920354,
      "grad_norm": 8.875,
      "learning_rate": 3.4624827818480977e-06,
      "loss": 1.0479,
      "num_input_tokens_seen": 18409928,
      "step": 33220
    },
    {
      "epoch": 582.9026548672566,
      "grad_norm": 9.875,
      "learning_rate": 3.4574995558386474e-06,
      "loss": 1.1333,
      "num_input_tokens_seen": 18412440,
      "step": 33225
    },
    {
      "epoch": 582.9911504424779,
      "grad_norm": 9.3125,
      "learning_rate": 3.452519651953487e-06,
      "loss": 0.8602,
      "num_input_tokens_seen": 18415256,
      "step": 33230
    },
    {
      "epoch": 583.070796460177,
      "grad_norm": 7.1875,
      "learning_rate": 3.447543070960585e-06,
      "loss": 0.825,
      "num_input_tokens_seen": 18417584,
      "step": 33235
    },
    {
      "epoch": 583.1592920353983,
      "grad_norm": 7.875,
      "learning_rate": 3.4425698136273778e-06,
      "loss": 1.159,
      "num_input_tokens_seen": 18420016,
      "step": 33240
    },
    {
      "epoch": 583.2477876106195,
      "grad_norm": 6.90625,
      "learning_rate": 3.437599880720821e-06,
      "loss": 1.0945,
      "num_input_tokens_seen": 18422944,
      "step": 33245
    },
    {
      "epoch": 583.3362831858407,
      "grad_norm": 8.1875,
      "learning_rate": 3.4326332730073267e-06,
      "loss": 1.0829,
      "num_input_tokens_seen": 18425504,
      "step": 33250
    },
    {
      "epoch": 583.4247787610619,
      "grad_norm": 7.0,
      "learning_rate": 3.427669991252813e-06,
      "loss": 0.9507,
      "num_input_tokens_seen": 18428736,
      "step": 33255
    },
    {
      "epoch": 583.5132743362832,
      "grad_norm": 6.875,
      "learning_rate": 3.42271003622269e-06,
      "loss": 1.0951,
      "num_input_tokens_seen": 18431392,
      "step": 33260
    },
    {
      "epoch": 583.6017699115044,
      "grad_norm": 8.3125,
      "learning_rate": 3.4177534086818286e-06,
      "loss": 1.1623,
      "num_input_tokens_seen": 18433712,
      "step": 33265
    },
    {
      "epoch": 583.6902654867257,
      "grad_norm": 7.46875,
      "learning_rate": 3.412800109394612e-06,
      "loss": 0.9776,
      "num_input_tokens_seen": 18436704,
      "step": 33270
    },
    {
      "epoch": 583.7787610619469,
      "grad_norm": 8.0625,
      "learning_rate": 3.4078501391249044e-06,
      "loss": 0.9166,
      "num_input_tokens_seen": 18439536,
      "step": 33275
    },
    {
      "epoch": 583.8672566371681,
      "grad_norm": 6.65625,
      "learning_rate": 3.4029034986360453e-06,
      "loss": 1.1116,
      "num_input_tokens_seen": 18442688,
      "step": 33280
    },
    {
      "epoch": 583.9557522123894,
      "grad_norm": 7.75,
      "learning_rate": 3.397960188690877e-06,
      "loss": 1.0957,
      "num_input_tokens_seen": 18445776,
      "step": 33285
    },
    {
      "epoch": 584.0353982300885,
      "grad_norm": 9.4375,
      "learning_rate": 3.393020210051717e-06,
      "loss": 0.8478,
      "num_input_tokens_seen": 18448192,
      "step": 33290
    },
    {
      "epoch": 584.1238938053098,
      "grad_norm": 11.9375,
      "learning_rate": 3.3880835634803655e-06,
      "loss": 0.9166,
      "num_input_tokens_seen": 18450928,
      "step": 33295
    },
    {
      "epoch": 584.212389380531,
      "grad_norm": 7.84375,
      "learning_rate": 3.383150249738126e-06,
      "loss": 1.123,
      "num_input_tokens_seen": 18453744,
      "step": 33300
    },
    {
      "epoch": 584.3008849557522,
      "grad_norm": 8.125,
      "learning_rate": 3.3782202695857663e-06,
      "loss": 1.1668,
      "num_input_tokens_seen": 18456224,
      "step": 33305
    },
    {
      "epoch": 584.3893805309734,
      "grad_norm": 7.5625,
      "learning_rate": 3.373293623783558e-06,
      "loss": 1.0239,
      "num_input_tokens_seen": 18459008,
      "step": 33310
    },
    {
      "epoch": 584.4778761061947,
      "grad_norm": 9.5625,
      "learning_rate": 3.368370313091257e-06,
      "loss": 1.0988,
      "num_input_tokens_seen": 18461744,
      "step": 33315
    },
    {
      "epoch": 584.566371681416,
      "grad_norm": 8.375,
      "learning_rate": 3.363450338268087e-06,
      "loss": 1.0297,
      "num_input_tokens_seen": 18464800,
      "step": 33320
    },
    {
      "epoch": 584.6548672566372,
      "grad_norm": 7.34375,
      "learning_rate": 3.358533700072783e-06,
      "loss": 0.9296,
      "num_input_tokens_seen": 18467744,
      "step": 33325
    },
    {
      "epoch": 584.7433628318585,
      "grad_norm": 5.53125,
      "learning_rate": 3.3536203992635377e-06,
      "loss": 1.0063,
      "num_input_tokens_seen": 18470672,
      "step": 33330
    },
    {
      "epoch": 584.8318584070796,
      "grad_norm": 9.3125,
      "learning_rate": 3.348710436598057e-06,
      "loss": 1.2041,
      "num_input_tokens_seen": 18472832,
      "step": 33335
    },
    {
      "epoch": 584.9203539823009,
      "grad_norm": 8.375,
      "learning_rate": 3.3438038128335155e-06,
      "loss": 1.0427,
      "num_input_tokens_seen": 18475904,
      "step": 33340
    },
    {
      "epoch": 585.0,
      "grad_norm": 11.4375,
      "learning_rate": 3.338900528726571e-06,
      "loss": 0.9649,
      "num_input_tokens_seen": 18478360,
      "step": 33345
    },
    {
      "epoch": 585.0884955752213,
      "grad_norm": 7.28125,
      "learning_rate": 3.3340005850333812e-06,
      "loss": 1.1031,
      "num_input_tokens_seen": 18480984,
      "step": 33350
    },
    {
      "epoch": 585.1769911504425,
      "grad_norm": 6.96875,
      "learning_rate": 3.329103982509568e-06,
      "loss": 0.8419,
      "num_input_tokens_seen": 18483704,
      "step": 33355
    },
    {
      "epoch": 585.2654867256637,
      "grad_norm": 7.75,
      "learning_rate": 3.324210721910259e-06,
      "loss": 1.1359,
      "num_input_tokens_seen": 18486360,
      "step": 33360
    },
    {
      "epoch": 585.3539823008849,
      "grad_norm": 9.0625,
      "learning_rate": 3.319320803990053e-06,
      "loss": 0.8787,
      "num_input_tokens_seen": 18488936,
      "step": 33365
    },
    {
      "epoch": 585.4424778761062,
      "grad_norm": 10.25,
      "learning_rate": 3.3144342295030274e-06,
      "loss": 1.1732,
      "num_input_tokens_seen": 18491784,
      "step": 33370
    },
    {
      "epoch": 585.5309734513274,
      "grad_norm": 9.4375,
      "learning_rate": 3.309550999202765e-06,
      "loss": 1.1294,
      "num_input_tokens_seen": 18494280,
      "step": 33375
    },
    {
      "epoch": 585.6194690265487,
      "grad_norm": 8.5,
      "learning_rate": 3.3046711138423197e-06,
      "loss": 1.0919,
      "num_input_tokens_seen": 18497624,
      "step": 33380
    },
    {
      "epoch": 585.70796460177,
      "grad_norm": 8.5,
      "learning_rate": 3.2997945741742255e-06,
      "loss": 1.0146,
      "num_input_tokens_seen": 18500872,
      "step": 33385
    },
    {
      "epoch": 585.7964601769911,
      "grad_norm": 6.3125,
      "learning_rate": 3.2949213809505082e-06,
      "loss": 1.0216,
      "num_input_tokens_seen": 18504216,
      "step": 33390
    },
    {
      "epoch": 585.8849557522124,
      "grad_norm": 9.6875,
      "learning_rate": 3.2900515349226834e-06,
      "loss": 1.016,
      "num_input_tokens_seen": 18506872,
      "step": 33395
    },
    {
      "epoch": 585.9734513274336,
      "grad_norm": 10.6875,
      "learning_rate": 3.285185036841731e-06,
      "loss": 1.1362,
      "num_input_tokens_seen": 18509416,
      "step": 33400
    },
    {
      "epoch": 585.9734513274336,
      "eval_loss": 1.0310944318771362,
      "eval_runtime": 0.9346,
      "eval_samples_per_second": 26.75,
      "eval_steps_per_second": 13.91,
      "num_input_tokens_seen": 18509416,
      "step": 33400
    },
    {
      "epoch": 586.0530973451328,
      "grad_norm": 8.5625,
      "learning_rate": 3.2803218874581377e-06,
      "loss": 1.0396,
      "num_input_tokens_seen": 18511720,
      "step": 33405
    },
    {
      "epoch": 586.141592920354,
      "grad_norm": 8.5625,
      "learning_rate": 3.2754620875218494e-06,
      "loss": 1.0584,
      "num_input_tokens_seen": 18514328,
      "step": 33410
    },
    {
      "epoch": 586.2300884955753,
      "grad_norm": 7.96875,
      "learning_rate": 3.2706056377823146e-06,
      "loss": 0.959,
      "num_input_tokens_seen": 18517400,
      "step": 33415
    },
    {
      "epoch": 586.3185840707964,
      "grad_norm": 10.5,
      "learning_rate": 3.2657525389884647e-06,
      "loss": 0.8193,
      "num_input_tokens_seen": 18519880,
      "step": 33420
    },
    {
      "epoch": 586.4070796460177,
      "grad_norm": 7.78125,
      "learning_rate": 3.260902791888698e-06,
      "loss": 1.2246,
      "num_input_tokens_seen": 18523000,
      "step": 33425
    },
    {
      "epoch": 586.4955752212389,
      "grad_norm": 8.6875,
      "learning_rate": 3.2560563972309166e-06,
      "loss": 1.2274,
      "num_input_tokens_seen": 18525544,
      "step": 33430
    },
    {
      "epoch": 586.5840707964602,
      "grad_norm": 7.6875,
      "learning_rate": 3.251213355762489e-06,
      "loss": 0.9208,
      "num_input_tokens_seen": 18528696,
      "step": 33435
    },
    {
      "epoch": 586.6725663716815,
      "grad_norm": 9.375,
      "learning_rate": 3.2463736682302707e-06,
      "loss": 0.9752,
      "num_input_tokens_seen": 18531448,
      "step": 33440
    },
    {
      "epoch": 586.7610619469026,
      "grad_norm": 12.8125,
      "learning_rate": 3.2415373353806124e-06,
      "loss": 1.1917,
      "num_input_tokens_seen": 18534456,
      "step": 33445
    },
    {
      "epoch": 586.8495575221239,
      "grad_norm": 9.1875,
      "learning_rate": 3.236704357959322e-06,
      "loss": 0.9536,
      "num_input_tokens_seen": 18537096,
      "step": 33450
    },
    {
      "epoch": 586.9380530973451,
      "grad_norm": 8.8125,
      "learning_rate": 3.2318747367117154e-06,
      "loss": 1.1069,
      "num_input_tokens_seen": 18540184,
      "step": 33455
    },
    {
      "epoch": 587.0176991150443,
      "grad_norm": 7.53125,
      "learning_rate": 3.227048472382585e-06,
      "loss": 1.0473,
      "num_input_tokens_seen": 18542616,
      "step": 33460
    },
    {
      "epoch": 587.1061946902655,
      "grad_norm": 7.375,
      "learning_rate": 3.2222255657161915e-06,
      "loss": 1.1182,
      "num_input_tokens_seen": 18545096,
      "step": 33465
    },
    {
      "epoch": 587.1946902654868,
      "grad_norm": 10.1875,
      "learning_rate": 3.2174060174562924e-06,
      "loss": 1.0835,
      "num_input_tokens_seen": 18547688,
      "step": 33470
    },
    {
      "epoch": 587.2831858407079,
      "grad_norm": 13.125,
      "learning_rate": 3.2125898283461298e-06,
      "loss": 1.1543,
      "num_input_tokens_seen": 18551448,
      "step": 33475
    },
    {
      "epoch": 587.3716814159292,
      "grad_norm": 7.9375,
      "learning_rate": 3.207776999128406e-06,
      "loss": 1.1138,
      "num_input_tokens_seen": 18553864,
      "step": 33480
    },
    {
      "epoch": 587.4601769911504,
      "grad_norm": 10.6875,
      "learning_rate": 3.202967530545331e-06,
      "loss": 1.0213,
      "num_input_tokens_seen": 18556376,
      "step": 33485
    },
    {
      "epoch": 587.5486725663717,
      "grad_norm": 6.5625,
      "learning_rate": 3.1981614233385778e-06,
      "loss": 0.9989,
      "num_input_tokens_seen": 18559224,
      "step": 33490
    },
    {
      "epoch": 587.637168141593,
      "grad_norm": 6.9375,
      "learning_rate": 3.1933586782493115e-06,
      "loss": 0.9513,
      "num_input_tokens_seen": 18562216,
      "step": 33495
    },
    {
      "epoch": 587.7256637168142,
      "grad_norm": 6.5625,
      "learning_rate": 3.188559296018184e-06,
      "loss": 1.0256,
      "num_input_tokens_seen": 18565080,
      "step": 33500
    },
    {
      "epoch": 587.8141592920354,
      "grad_norm": 10.75,
      "learning_rate": 3.1837632773853098e-06,
      "loss": 1.0999,
      "num_input_tokens_seen": 18568056,
      "step": 33505
    },
    {
      "epoch": 587.9026548672566,
      "grad_norm": 8.6875,
      "learning_rate": 3.178970623090294e-06,
      "loss": 1.0635,
      "num_input_tokens_seen": 18571016,
      "step": 33510
    },
    {
      "epoch": 587.9911504424779,
      "grad_norm": 7.40625,
      "learning_rate": 3.174181333872234e-06,
      "loss": 0.9312,
      "num_input_tokens_seen": 18573560,
      "step": 33515
    },
    {
      "epoch": 588.070796460177,
      "grad_norm": 8.625,
      "learning_rate": 3.169395410469686e-06,
      "loss": 1.1119,
      "num_input_tokens_seen": 18575736,
      "step": 33520
    },
    {
      "epoch": 588.1592920353983,
      "grad_norm": 9.5,
      "learning_rate": 3.164612853620713e-06,
      "loss": 1.0454,
      "num_input_tokens_seen": 18578488,
      "step": 33525
    },
    {
      "epoch": 588.2477876106195,
      "grad_norm": 8.6875,
      "learning_rate": 3.1598336640628333e-06,
      "loss": 1.0102,
      "num_input_tokens_seen": 18581528,
      "step": 33530
    },
    {
      "epoch": 588.3362831858407,
      "grad_norm": 8.5625,
      "learning_rate": 3.155057842533063e-06,
      "loss": 0.8605,
      "num_input_tokens_seen": 18584520,
      "step": 33535
    },
    {
      "epoch": 588.4247787610619,
      "grad_norm": 7.875,
      "learning_rate": 3.1502853897678984e-06,
      "loss": 1.0794,
      "num_input_tokens_seen": 18587672,
      "step": 33540
    },
    {
      "epoch": 588.5132743362832,
      "grad_norm": 10.5,
      "learning_rate": 3.1455163065033017e-06,
      "loss": 1.0889,
      "num_input_tokens_seen": 18590600,
      "step": 33545
    },
    {
      "epoch": 588.6017699115044,
      "grad_norm": 8.4375,
      "learning_rate": 3.140750593474734e-06,
      "loss": 1.0517,
      "num_input_tokens_seen": 18593512,
      "step": 33550
    },
    {
      "epoch": 588.6902654867257,
      "grad_norm": 7.9375,
      "learning_rate": 3.1359882514171294e-06,
      "loss": 0.9949,
      "num_input_tokens_seen": 18596296,
      "step": 33555
    },
    {
      "epoch": 588.7787610619469,
      "grad_norm": 10.3125,
      "learning_rate": 3.1312292810648903e-06,
      "loss": 1.0228,
      "num_input_tokens_seen": 18598744,
      "step": 33560
    },
    {
      "epoch": 588.8672566371681,
      "grad_norm": 8.9375,
      "learning_rate": 3.1264736831519204e-06,
      "loss": 1.1258,
      "num_input_tokens_seen": 18601704,
      "step": 33565
    },
    {
      "epoch": 588.9557522123894,
      "grad_norm": 6.71875,
      "learning_rate": 3.1217214584115863e-06,
      "loss": 1.0923,
      "num_input_tokens_seen": 18604200,
      "step": 33570
    },
    {
      "epoch": 589.0353982300885,
      "grad_norm": 7.875,
      "learning_rate": 3.116972607576746e-06,
      "loss": 0.9448,
      "num_input_tokens_seen": 18606368,
      "step": 33575
    },
    {
      "epoch": 589.1238938053098,
      "grad_norm": 7.3125,
      "learning_rate": 3.1122271313797303e-06,
      "loss": 1.1433,
      "num_input_tokens_seen": 18609264,
      "step": 33580
    },
    {
      "epoch": 589.212389380531,
      "grad_norm": 7.71875,
      "learning_rate": 3.107485030552343e-06,
      "loss": 1.0701,
      "num_input_tokens_seen": 18611920,
      "step": 33585
    },
    {
      "epoch": 589.3008849557522,
      "grad_norm": 8.3125,
      "learning_rate": 3.1027463058258848e-06,
      "loss": 0.8449,
      "num_input_tokens_seen": 18614704,
      "step": 33590
    },
    {
      "epoch": 589.3893805309734,
      "grad_norm": 9.125,
      "learning_rate": 3.0980109579311273e-06,
      "loss": 1.3552,
      "num_input_tokens_seen": 18617264,
      "step": 33595
    },
    {
      "epoch": 589.4778761061947,
      "grad_norm": 10.1875,
      "learning_rate": 3.093278987598314e-06,
      "loss": 1.0518,
      "num_input_tokens_seen": 18620544,
      "step": 33600
    },
    {
      "epoch": 589.4778761061947,
      "eval_loss": 1.0310944318771362,
      "eval_runtime": 0.9352,
      "eval_samples_per_second": 26.733,
      "eval_steps_per_second": 13.901,
      "num_input_tokens_seen": 18620544,
      "step": 33600
    },
    {
      "epoch": 589.566371681416,
      "grad_norm": 8.625,
      "learning_rate": 3.0885503955571826e-06,
      "loss": 1.257,
      "num_input_tokens_seen": 18623232,
      "step": 33605
    },
    {
      "epoch": 589.6548672566372,
      "grad_norm": 6.8125,
      "learning_rate": 3.0838251825369313e-06,
      "loss": 0.985,
      "num_input_tokens_seen": 18625856,
      "step": 33610
    },
    {
      "epoch": 589.7433628318585,
      "grad_norm": 8.25,
      "learning_rate": 3.0791033492662517e-06,
      "loss": 0.976,
      "num_input_tokens_seen": 18628512,
      "step": 33615
    },
    {
      "epoch": 589.8318584070796,
      "grad_norm": 9.625,
      "learning_rate": 3.0743848964733203e-06,
      "loss": 0.9109,
      "num_input_tokens_seen": 18631648,
      "step": 33620
    },
    {
      "epoch": 589.9203539823009,
      "grad_norm": 8.0625,
      "learning_rate": 3.0696698248857625e-06,
      "loss": 1.0323,
      "num_input_tokens_seen": 18634880,
      "step": 33625
    },
    {
      "epoch": 590.0,
      "grad_norm": 6.34375,
      "learning_rate": 3.0649581352307192e-06,
      "loss": 0.9675,
      "num_input_tokens_seen": 18636872,
      "step": 33630
    },
    {
      "epoch": 590.0884955752213,
      "grad_norm": 8.9375,
      "learning_rate": 3.060249828234776e-06,
      "loss": 0.9474,
      "num_input_tokens_seen": 18639944,
      "step": 33635
    },
    {
      "epoch": 590.1769911504425,
      "grad_norm": 13.5,
      "learning_rate": 3.055544904624025e-06,
      "loss": 1.167,
      "num_input_tokens_seen": 18642808,
      "step": 33640
    },
    {
      "epoch": 590.2654867256637,
      "grad_norm": 7.625,
      "learning_rate": 3.050843365124026e-06,
      "loss": 1.1092,
      "num_input_tokens_seen": 18645720,
      "step": 33645
    },
    {
      "epoch": 590.3539823008849,
      "grad_norm": 8.25,
      "learning_rate": 3.0461452104598083e-06,
      "loss": 0.9506,
      "num_input_tokens_seen": 18648408,
      "step": 33650
    },
    {
      "epoch": 590.4424778761062,
      "grad_norm": 10.375,
      "learning_rate": 3.0414504413558836e-06,
      "loss": 1.1363,
      "num_input_tokens_seen": 18651096,
      "step": 33655
    },
    {
      "epoch": 590.5309734513274,
      "grad_norm": 9.6875,
      "learning_rate": 3.0367590585362564e-06,
      "loss": 0.9235,
      "num_input_tokens_seen": 18653976,
      "step": 33660
    },
    {
      "epoch": 590.6194690265487,
      "grad_norm": 7.78125,
      "learning_rate": 3.0320710627243813e-06,
      "loss": 1.0363,
      "num_input_tokens_seen": 18656984,
      "step": 33665
    },
    {
      "epoch": 590.70796460177,
      "grad_norm": 6.84375,
      "learning_rate": 3.027386454643222e-06,
      "loss": 1.1368,
      "num_input_tokens_seen": 18659624,
      "step": 33670
    },
    {
      "epoch": 590.7964601769911,
      "grad_norm": 8.6875,
      "learning_rate": 3.0227052350151914e-06,
      "loss": 0.8857,
      "num_input_tokens_seen": 18662200,
      "step": 33675
    },
    {
      "epoch": 590.8849557522124,
      "grad_norm": 9.625,
      "learning_rate": 3.0180274045621957e-06,
      "loss": 1.1813,
      "num_input_tokens_seen": 18664888,
      "step": 33680
    },
    {
      "epoch": 590.9734513274336,
      "grad_norm": 12.0625,
      "learning_rate": 3.013352964005625e-06,
      "loss": 0.9884,
      "num_input_tokens_seen": 18667976,
      "step": 33685
    },
    {
      "epoch": 591.0530973451328,
      "grad_norm": 8.25,
      "learning_rate": 3.0086819140663218e-06,
      "loss": 1.0333,
      "num_input_tokens_seen": 18670376,
      "step": 33690
    },
    {
      "epoch": 591.141592920354,
      "grad_norm": 7.28125,
      "learning_rate": 3.0040142554646265e-06,
      "loss": 1.0691,
      "num_input_tokens_seen": 18672936,
      "step": 33695
    },
    {
      "epoch": 591.2300884955753,
      "grad_norm": 8.6875,
      "learning_rate": 2.999349988920361e-06,
      "loss": 0.9446,
      "num_input_tokens_seen": 18675784,
      "step": 33700
    },
    {
      "epoch": 591.3185840707964,
      "grad_norm": 5.4375,
      "learning_rate": 2.994689115152796e-06,
      "loss": 0.8345,
      "num_input_tokens_seen": 18678680,
      "step": 33705
    },
    {
      "epoch": 591.4070796460177,
      "grad_norm": 9.4375,
      "learning_rate": 2.9900316348807105e-06,
      "loss": 1.2113,
      "num_input_tokens_seen": 18681144,
      "step": 33710
    },
    {
      "epoch": 591.4955752212389,
      "grad_norm": 7.8125,
      "learning_rate": 2.985377548822338e-06,
      "loss": 1.2488,
      "num_input_tokens_seen": 18683720,
      "step": 33715
    },
    {
      "epoch": 591.5840707964602,
      "grad_norm": 7.78125,
      "learning_rate": 2.980726857695404e-06,
      "loss": 1.0136,
      "num_input_tokens_seen": 18686664,
      "step": 33720
    },
    {
      "epoch": 591.6725663716815,
      "grad_norm": 6.5625,
      "learning_rate": 2.9760795622171017e-06,
      "loss": 0.9138,
      "num_input_tokens_seen": 18689688,
      "step": 33725
    },
    {
      "epoch": 591.7610619469026,
      "grad_norm": 6.96875,
      "learning_rate": 2.971435663104094e-06,
      "loss": 1.0565,
      "num_input_tokens_seen": 18692648,
      "step": 33730
    },
    {
      "epoch": 591.8495575221239,
      "grad_norm": 7.96875,
      "learning_rate": 2.9667951610725385e-06,
      "loss": 1.1142,
      "num_input_tokens_seen": 18695528,
      "step": 33735
    },
    {
      "epoch": 591.9380530973451,
      "grad_norm": 9.1875,
      "learning_rate": 2.9621580568380575e-06,
      "loss": 1.097,
      "num_input_tokens_seen": 18698840,
      "step": 33740
    },
    {
      "epoch": 592.0176991150443,
      "grad_norm": 6.21875,
      "learning_rate": 2.9575243511157453e-06,
      "loss": 1.114,
      "num_input_tokens_seen": 18701360,
      "step": 33745
    },
    {
      "epoch": 592.1061946902655,
      "grad_norm": 6.40625,
      "learning_rate": 2.952894044620186e-06,
      "loss": 1.0232,
      "num_input_tokens_seen": 18704064,
      "step": 33750
    },
    {
      "epoch": 592.1946902654868,
      "grad_norm": 7.5625,
      "learning_rate": 2.948267138065419e-06,
      "loss": 0.8597,
      "num_input_tokens_seen": 18706928,
      "step": 33755
    },
    {
      "epoch": 592.2831858407079,
      "grad_norm": 10.5625,
      "learning_rate": 2.943643632164983e-06,
      "loss": 1.2849,
      "num_input_tokens_seen": 18709744,
      "step": 33760
    },
    {
      "epoch": 592.3716814159292,
      "grad_norm": 9.8125,
      "learning_rate": 2.939023527631879e-06,
      "loss": 1.1463,
      "num_input_tokens_seen": 18712336,
      "step": 33765
    },
    {
      "epoch": 592.4601769911504,
      "grad_norm": 8.375,
      "learning_rate": 2.934406825178576e-06,
      "loss": 1.0259,
      "num_input_tokens_seen": 18715008,
      "step": 33770
    },
    {
      "epoch": 592.5486725663717,
      "grad_norm": 8.8125,
      "learning_rate": 2.9297935255170357e-06,
      "loss": 1.1146,
      "num_input_tokens_seen": 18717728,
      "step": 33775
    },
    {
      "epoch": 592.637168141593,
      "grad_norm": 6.875,
      "learning_rate": 2.925183629358691e-06,
      "loss": 0.9918,
      "num_input_tokens_seen": 18720352,
      "step": 33780
    },
    {
      "epoch": 592.7256637168142,
      "grad_norm": 6.96875,
      "learning_rate": 2.9205771374144346e-06,
      "loss": 1.0647,
      "num_input_tokens_seen": 18723600,
      "step": 33785
    },
    {
      "epoch": 592.8141592920354,
      "grad_norm": 9.125,
      "learning_rate": 2.915974050394657e-06,
      "loss": 1.0164,
      "num_input_tokens_seen": 18726224,
      "step": 33790
    },
    {
      "epoch": 592.9026548672566,
      "grad_norm": 8.4375,
      "learning_rate": 2.9113743690092067e-06,
      "loss": 1.0498,
      "num_input_tokens_seen": 18728816,
      "step": 33795
    },
    {
      "epoch": 592.9911504424779,
      "grad_norm": 8.1875,
      "learning_rate": 2.906778093967402e-06,
      "loss": 1.0665,
      "num_input_tokens_seen": 18731712,
      "step": 33800
    },
    {
      "epoch": 592.9911504424779,
      "eval_loss": 1.0310944318771362,
      "eval_runtime": 0.9355,
      "eval_samples_per_second": 26.724,
      "eval_steps_per_second": 13.897,
      "num_input_tokens_seen": 18731712,
      "step": 33800
    },
    {
      "epoch": 593.070796460177,
      "grad_norm": 6.34375,
      "learning_rate": 2.9021852259780656e-06,
      "loss": 0.9112,
      "num_input_tokens_seen": 18733984,
      "step": 33805
    },
    {
      "epoch": 593.1592920353983,
      "grad_norm": 8.625,
      "learning_rate": 2.8975957657494583e-06,
      "loss": 1.0243,
      "num_input_tokens_seen": 18737056,
      "step": 33810
    },
    {
      "epoch": 593.2477876106195,
      "grad_norm": 10.125,
      "learning_rate": 2.8930097139893417e-06,
      "loss": 1.2514,
      "num_input_tokens_seen": 18739632,
      "step": 33815
    },
    {
      "epoch": 593.3362831858407,
      "grad_norm": 10.375,
      "learning_rate": 2.888427071404945e-06,
      "loss": 1.2794,
      "num_input_tokens_seen": 18742384,
      "step": 33820
    },
    {
      "epoch": 593.4247787610619,
      "grad_norm": 8.3125,
      "learning_rate": 2.8838478387029606e-06,
      "loss": 1.1427,
      "num_input_tokens_seen": 18744960,
      "step": 33825
    },
    {
      "epoch": 593.5132743362832,
      "grad_norm": 8.5625,
      "learning_rate": 2.8792720165895737e-06,
      "loss": 1.0872,
      "num_input_tokens_seen": 18747968,
      "step": 33830
    },
    {
      "epoch": 593.6017699115044,
      "grad_norm": 7.1875,
      "learning_rate": 2.874699605770423e-06,
      "loss": 0.9471,
      "num_input_tokens_seen": 18750800,
      "step": 33835
    },
    {
      "epoch": 593.6902654867257,
      "grad_norm": 7.40625,
      "learning_rate": 2.8701306069506383e-06,
      "loss": 0.9479,
      "num_input_tokens_seen": 18753520,
      "step": 33840
    },
    {
      "epoch": 593.7787610619469,
      "grad_norm": 8.3125,
      "learning_rate": 2.8655650208348178e-06,
      "loss": 0.9708,
      "num_input_tokens_seen": 18756784,
      "step": 33845
    },
    {
      "epoch": 593.8672566371681,
      "grad_norm": 7.375,
      "learning_rate": 2.8610028481270257e-06,
      "loss": 0.9675,
      "num_input_tokens_seen": 18759200,
      "step": 33850
    },
    {
      "epoch": 593.9557522123894,
      "grad_norm": 8.375,
      "learning_rate": 2.856444089530813e-06,
      "loss": 1.0364,
      "num_input_tokens_seen": 18762080,
      "step": 33855
    },
    {
      "epoch": 594.0353982300885,
      "grad_norm": 10.0,
      "learning_rate": 2.8518887457491955e-06,
      "loss": 1.3939,
      "num_input_tokens_seen": 18764248,
      "step": 33860
    },
    {
      "epoch": 594.1238938053098,
      "grad_norm": 8.1875,
      "learning_rate": 2.8473368174846666e-06,
      "loss": 1.0504,
      "num_input_tokens_seen": 18766808,
      "step": 33865
    },
    {
      "epoch": 594.212389380531,
      "grad_norm": 7.8125,
      "learning_rate": 2.842788305439184e-06,
      "loss": 1.0923,
      "num_input_tokens_seen": 18769176,
      "step": 33870
    },
    {
      "epoch": 594.3008849557522,
      "grad_norm": 7.96875,
      "learning_rate": 2.8382432103141925e-06,
      "loss": 1.0511,
      "num_input_tokens_seen": 18771672,
      "step": 33875
    },
    {
      "epoch": 594.3893805309734,
      "grad_norm": 8.125,
      "learning_rate": 2.833701532810598e-06,
      "loss": 0.8539,
      "num_input_tokens_seen": 18774392,
      "step": 33880
    },
    {
      "epoch": 594.4778761061947,
      "grad_norm": 7.84375,
      "learning_rate": 2.8291632736287877e-06,
      "loss": 0.9953,
      "num_input_tokens_seen": 18777176,
      "step": 33885
    },
    {
      "epoch": 594.566371681416,
      "grad_norm": 7.28125,
      "learning_rate": 2.824628433468615e-06,
      "loss": 0.9092,
      "num_input_tokens_seen": 18779848,
      "step": 33890
    },
    {
      "epoch": 594.6548672566372,
      "grad_norm": 9.0625,
      "learning_rate": 2.8200970130294073e-06,
      "loss": 1.098,
      "num_input_tokens_seen": 18783368,
      "step": 33895
    },
    {
      "epoch": 594.7433628318585,
      "grad_norm": 7.96875,
      "learning_rate": 2.8155690130099775e-06,
      "loss": 1.0651,
      "num_input_tokens_seen": 18786648,
      "step": 33900
    },
    {
      "epoch": 594.8318584070796,
      "grad_norm": 8.1875,
      "learning_rate": 2.8110444341085895e-06,
      "loss": 1.2461,
      "num_input_tokens_seen": 18789496,
      "step": 33905
    },
    {
      "epoch": 594.9203539823009,
      "grad_norm": 7.3125,
      "learning_rate": 2.806523277022996e-06,
      "loss": 0.9856,
      "num_input_tokens_seen": 18792952,
      "step": 33910
    },
    {
      "epoch": 595.0,
      "grad_norm": 16.625,
      "learning_rate": 2.802005542450409e-06,
      "loss": 1.0568,
      "num_input_tokens_seen": 18795064,
      "step": 33915
    },
    {
      "epoch": 595.0884955752213,
      "grad_norm": 7.5,
      "learning_rate": 2.797491231087526e-06,
      "loss": 1.0372,
      "num_input_tokens_seen": 18797528,
      "step": 33920
    },
    {
      "epoch": 595.1769911504425,
      "grad_norm": 6.8125,
      "learning_rate": 2.7929803436305137e-06,
      "loss": 1.0328,
      "num_input_tokens_seen": 18800216,
      "step": 33925
    },
    {
      "epoch": 595.2654867256637,
      "grad_norm": 6.46875,
      "learning_rate": 2.788472880774998e-06,
      "loss": 1.0075,
      "num_input_tokens_seen": 18802904,
      "step": 33930
    },
    {
      "epoch": 595.3539823008849,
      "grad_norm": 9.3125,
      "learning_rate": 2.7839688432160977e-06,
      "loss": 1.0911,
      "num_input_tokens_seen": 18805752,
      "step": 33935
    },
    {
      "epoch": 595.4424778761062,
      "grad_norm": 10.3125,
      "learning_rate": 2.779468231648383e-06,
      "loss": 1.004,
      "num_input_tokens_seen": 18808552,
      "step": 33940
    },
    {
      "epoch": 595.5309734513274,
      "grad_norm": 7.78125,
      "learning_rate": 2.774971046765906e-06,
      "loss": 0.9655,
      "num_input_tokens_seen": 18811288,
      "step": 33945
    },
    {
      "epoch": 595.6194690265487,
      "grad_norm": 7.90625,
      "learning_rate": 2.770477289262194e-06,
      "loss": 0.9391,
      "num_input_tokens_seen": 18814216,
      "step": 33950
    },
    {
      "epoch": 595.70796460177,
      "grad_norm": 6.78125,
      "learning_rate": 2.765986959830233e-06,
      "loss": 1.1003,
      "num_input_tokens_seen": 18817288,
      "step": 33955
    },
    {
      "epoch": 595.7964601769911,
      "grad_norm": 7.6875,
      "learning_rate": 2.761500059162492e-06,
      "loss": 1.1265,
      "num_input_tokens_seen": 18819832,
      "step": 33960
    },
    {
      "epoch": 595.8849557522124,
      "grad_norm": 7.0625,
      "learning_rate": 2.757016587950914e-06,
      "loss": 1.1118,
      "num_input_tokens_seen": 18822680,
      "step": 33965
    },
    {
      "epoch": 595.9734513274336,
      "grad_norm": 9.875,
      "learning_rate": 2.752536546886897e-06,
      "loss": 1.0768,
      "num_input_tokens_seen": 18825720,
      "step": 33970
    },
    {
      "epoch": 596.0530973451328,
      "grad_norm": 8.9375,
      "learning_rate": 2.7480599366613234e-06,
      "loss": 1.1695,
      "num_input_tokens_seen": 18828008,
      "step": 33975
    },
    {
      "epoch": 596.141592920354,
      "grad_norm": 7.46875,
      "learning_rate": 2.7435867579645473e-06,
      "loss": 1.1964,
      "num_input_tokens_seen": 18830776,
      "step": 33980
    },
    {
      "epoch": 596.2300884955753,
      "grad_norm": 6.5625,
      "learning_rate": 2.739117011486378e-06,
      "loss": 0.9863,
      "num_input_tokens_seen": 18833240,
      "step": 33985
    },
    {
      "epoch": 596.3185840707964,
      "grad_norm": 6.5625,
      "learning_rate": 2.7346506979161216e-06,
      "loss": 0.9579,
      "num_input_tokens_seen": 18836440,
      "step": 33990
    },
    {
      "epoch": 596.4070796460177,
      "grad_norm": 11.125,
      "learning_rate": 2.7301878179425227e-06,
      "loss": 1.0663,
      "num_input_tokens_seen": 18838728,
      "step": 33995
    },
    {
      "epoch": 596.4955752212389,
      "grad_norm": 8.75,
      "learning_rate": 2.7257283722538244e-06,
      "loss": 0.991,
      "num_input_tokens_seen": 18841128,
      "step": 34000
    },
    {
      "epoch": 596.4955752212389,
      "eval_loss": 1.0310944318771362,
      "eval_runtime": 0.9392,
      "eval_samples_per_second": 26.617,
      "eval_steps_per_second": 13.841,
      "num_input_tokens_seen": 18841128,
      "step": 34000
    },
    {
      "epoch": 596.5840707964602,
      "grad_norm": 7.65625,
      "learning_rate": 2.7212723615377326e-06,
      "loss": 0.9265,
      "num_input_tokens_seen": 18843624,
      "step": 34005
    },
    {
      "epoch": 596.6725663716815,
      "grad_norm": 11.125,
      "learning_rate": 2.7168197864814145e-06,
      "loss": 1.0168,
      "num_input_tokens_seen": 18847016,
      "step": 34010
    },
    {
      "epoch": 596.7610619469026,
      "grad_norm": 8.5625,
      "learning_rate": 2.712370647771509e-06,
      "loss": 1.1195,
      "num_input_tokens_seen": 18849896,
      "step": 34015
    },
    {
      "epoch": 596.8495575221239,
      "grad_norm": 6.71875,
      "learning_rate": 2.707924946094137e-06,
      "loss": 1.1021,
      "num_input_tokens_seen": 18852824,
      "step": 34020
    },
    {
      "epoch": 596.9380530973451,
      "grad_norm": 9.375,
      "learning_rate": 2.7034826821348723e-06,
      "loss": 0.9913,
      "num_input_tokens_seen": 18856008,
      "step": 34025
    },
    {
      "epoch": 597.0176991150443,
      "grad_norm": 8.375,
      "learning_rate": 2.6990438565787786e-06,
      "loss": 1.2051,
      "num_input_tokens_seen": 18858296,
      "step": 34030
    },
    {
      "epoch": 597.1061946902655,
      "grad_norm": 10.75,
      "learning_rate": 2.6946084701103714e-06,
      "loss": 0.9712,
      "num_input_tokens_seen": 18860952,
      "step": 34035
    },
    {
      "epoch": 597.1946902654868,
      "grad_norm": 11.125,
      "learning_rate": 2.6901765234136428e-06,
      "loss": 0.9943,
      "num_input_tokens_seen": 18863576,
      "step": 34040
    },
    {
      "epoch": 597.2831858407079,
      "grad_norm": 11.375,
      "learning_rate": 2.685748017172063e-06,
      "loss": 0.8753,
      "num_input_tokens_seen": 18866712,
      "step": 34045
    },
    {
      "epoch": 597.3716814159292,
      "grad_norm": 12.125,
      "learning_rate": 2.681322952068549e-06,
      "loss": 0.8815,
      "num_input_tokens_seen": 18869336,
      "step": 34050
    },
    {
      "epoch": 597.4601769911504,
      "grad_norm": 7.5625,
      "learning_rate": 2.6769013287855137e-06,
      "loss": 1.2072,
      "num_input_tokens_seen": 18872504,
      "step": 34055
    },
    {
      "epoch": 597.5486725663717,
      "grad_norm": 8.125,
      "learning_rate": 2.6724831480048286e-06,
      "loss": 1.2074,
      "num_input_tokens_seen": 18875192,
      "step": 34060
    },
    {
      "epoch": 597.637168141593,
      "grad_norm": 10.8125,
      "learning_rate": 2.66806841040782e-06,
      "loss": 1.0617,
      "num_input_tokens_seen": 18877944,
      "step": 34065
    },
    {
      "epoch": 597.7256637168142,
      "grad_norm": 7.6875,
      "learning_rate": 2.6636571166753083e-06,
      "loss": 1.0744,
      "num_input_tokens_seen": 18880824,
      "step": 34070
    },
    {
      "epoch": 597.8141592920354,
      "grad_norm": 10.3125,
      "learning_rate": 2.6592492674875598e-06,
      "loss": 1.0175,
      "num_input_tokens_seen": 18883384,
      "step": 34075
    },
    {
      "epoch": 597.9026548672566,
      "grad_norm": 10.4375,
      "learning_rate": 2.6548448635243305e-06,
      "loss": 1.1849,
      "num_input_tokens_seen": 18886024,
      "step": 34080
    },
    {
      "epoch": 597.9911504424779,
      "grad_norm": 7.09375,
      "learning_rate": 2.650443905464828e-06,
      "loss": 0.972,
      "num_input_tokens_seen": 18889032,
      "step": 34085
    },
    {
      "epoch": 598.070796460177,
      "grad_norm": 9.0,
      "learning_rate": 2.646046393987739e-06,
      "loss": 1.0246,
      "num_input_tokens_seen": 18891136,
      "step": 34090
    },
    {
      "epoch": 598.1592920353983,
      "grad_norm": 8.25,
      "learning_rate": 2.64165232977121e-06,
      "loss": 1.0258,
      "num_input_tokens_seen": 18893664,
      "step": 34095
    },
    {
      "epoch": 598.2477876106195,
      "grad_norm": 9.125,
      "learning_rate": 2.6372617134928695e-06,
      "loss": 1.0778,
      "num_input_tokens_seen": 18897056,
      "step": 34100
    },
    {
      "epoch": 598.3362831858407,
      "grad_norm": 8.5625,
      "learning_rate": 2.6328745458297943e-06,
      "loss": 0.9805,
      "num_input_tokens_seen": 18899728,
      "step": 34105
    },
    {
      "epoch": 598.4247787610619,
      "grad_norm": 8.875,
      "learning_rate": 2.6284908274585546e-06,
      "loss": 0.9103,
      "num_input_tokens_seen": 18902928,
      "step": 34110
    },
    {
      "epoch": 598.5132743362832,
      "grad_norm": 6.875,
      "learning_rate": 2.6241105590551595e-06,
      "loss": 1.005,
      "num_input_tokens_seen": 18905840,
      "step": 34115
    },
    {
      "epoch": 598.6017699115044,
      "grad_norm": 6.96875,
      "learning_rate": 2.6197337412951105e-06,
      "loss": 1.0491,
      "num_input_tokens_seen": 18908784,
      "step": 34120
    },
    {
      "epoch": 598.6902654867257,
      "grad_norm": 6.375,
      "learning_rate": 2.6153603748533705e-06,
      "loss": 1.1421,
      "num_input_tokens_seen": 18911312,
      "step": 34125
    },
    {
      "epoch": 598.7787610619469,
      "grad_norm": 9.375,
      "learning_rate": 2.6109904604043585e-06,
      "loss": 1.212,
      "num_input_tokens_seen": 18914208,
      "step": 34130
    },
    {
      "epoch": 598.8672566371681,
      "grad_norm": 10.875,
      "learning_rate": 2.6066239986219765e-06,
      "loss": 1.0871,
      "num_input_tokens_seen": 18917104,
      "step": 34135
    },
    {
      "epoch": 598.9557522123894,
      "grad_norm": 8.5,
      "learning_rate": 2.602260990179592e-06,
      "loss": 0.9731,
      "num_input_tokens_seen": 18919824,
      "step": 34140
    },
    {
      "epoch": 599.0353982300885,
      "grad_norm": 8.4375,
      "learning_rate": 2.5979014357500248e-06,
      "loss": 1.2437,
      "num_input_tokens_seen": 18921896,
      "step": 34145
    },
    {
      "epoch": 599.1238938053098,
      "grad_norm": 7.53125,
      "learning_rate": 2.5935453360055844e-06,
      "loss": 0.9589,
      "num_input_tokens_seen": 18924520,
      "step": 34150
    },
    {
      "epoch": 599.212389380531,
      "grad_norm": 8.8125,
      "learning_rate": 2.5891926916180283e-06,
      "loss": 0.8753,
      "num_input_tokens_seen": 18927192,
      "step": 34155
    },
    {
      "epoch": 599.3008849557522,
      "grad_norm": 7.65625,
      "learning_rate": 2.5848435032585883e-06,
      "loss": 0.9231,
      "num_input_tokens_seen": 18930024,
      "step": 34160
    },
    {
      "epoch": 599.3893805309734,
      "grad_norm": 7.0625,
      "learning_rate": 2.58049777159797e-06,
      "loss": 1.1927,
      "num_input_tokens_seen": 18932728,
      "step": 34165
    },
    {
      "epoch": 599.4778761061947,
      "grad_norm": 9.5625,
      "learning_rate": 2.576155497306332e-06,
      "loss": 1.1833,
      "num_input_tokens_seen": 18935480,
      "step": 34170
    },
    {
      "epoch": 599.566371681416,
      "grad_norm": 6.09375,
      "learning_rate": 2.57181668105331e-06,
      "loss": 0.911,
      "num_input_tokens_seen": 18938088,
      "step": 34175
    },
    {
      "epoch": 599.6548672566372,
      "grad_norm": 8.6875,
      "learning_rate": 2.567481323508014e-06,
      "loss": 1.2733,
      "num_input_tokens_seen": 18940888,
      "step": 34180
    },
    {
      "epoch": 599.7433628318585,
      "grad_norm": 9.5,
      "learning_rate": 2.5631494253389954e-06,
      "loss": 0.9279,
      "num_input_tokens_seen": 18944088,
      "step": 34185
    },
    {
      "epoch": 599.8318584070796,
      "grad_norm": 7.125,
      "learning_rate": 2.5588209872142997e-06,
      "loss": 1.0867,
      "num_input_tokens_seen": 18946984,
      "step": 34190
    },
    {
      "epoch": 599.9203539823009,
      "grad_norm": 12.5625,
      "learning_rate": 2.5544960098014186e-06,
      "loss": 1.115,
      "num_input_tokens_seen": 18949960,
      "step": 34195
    },
    {
      "epoch": 600.0,
      "grad_norm": 6.34375,
      "learning_rate": 2.550174493767318e-06,
      "loss": 0.9105,
      "num_input_tokens_seen": 18952336,
      "step": 34200
    },
    {
      "epoch": 600.0,
      "eval_loss": 1.0310944318771362,
      "eval_runtime": 0.9324,
      "eval_samples_per_second": 26.812,
      "eval_steps_per_second": 13.942,
      "num_input_tokens_seen": 18952336,
      "step": 34200
    },
    {
      "epoch": 600.0884955752213,
      "grad_norm": 8.1875,
      "learning_rate": 2.545856439778438e-06,
      "loss": 1.0783,
      "num_input_tokens_seen": 18955664,
      "step": 34205
    },
    {
      "epoch": 600.1769911504425,
      "grad_norm": 8.625,
      "learning_rate": 2.541541848500667e-06,
      "loss": 1.0526,
      "num_input_tokens_seen": 18958784,
      "step": 34210
    },
    {
      "epoch": 600.2654867256637,
      "grad_norm": 8.75,
      "learning_rate": 2.5372307205993733e-06,
      "loss": 1.0849,
      "num_input_tokens_seen": 18961328,
      "step": 34215
    },
    {
      "epoch": 600.3539823008849,
      "grad_norm": 8.6875,
      "learning_rate": 2.5329230567393917e-06,
      "loss": 1.0549,
      "num_input_tokens_seen": 18963856,
      "step": 34220
    },
    {
      "epoch": 600.4424778761062,
      "grad_norm": 8.25,
      "learning_rate": 2.5286188575850164e-06,
      "loss": 0.9767,
      "num_input_tokens_seen": 18967312,
      "step": 34225
    },
    {
      "epoch": 600.5309734513274,
      "grad_norm": 8.8125,
      "learning_rate": 2.5243181237999984e-06,
      "loss": 1.016,
      "num_input_tokens_seen": 18970112,
      "step": 34230
    },
    {
      "epoch": 600.6194690265487,
      "grad_norm": 7.34375,
      "learning_rate": 2.520020856047578e-06,
      "loss": 1.1515,
      "num_input_tokens_seen": 18972944,
      "step": 34235
    },
    {
      "epoch": 600.70796460177,
      "grad_norm": 5.9375,
      "learning_rate": 2.515727054990438e-06,
      "loss": 1.0196,
      "num_input_tokens_seen": 18975696,
      "step": 34240
    },
    {
      "epoch": 600.7964601769911,
      "grad_norm": 8.125,
      "learning_rate": 2.511436721290747e-06,
      "loss": 0.9709,
      "num_input_tokens_seen": 18978384,
      "step": 34245
    },
    {
      "epoch": 600.8849557522124,
      "grad_norm": 7.125,
      "learning_rate": 2.5071498556101164e-06,
      "loss": 1.0355,
      "num_input_tokens_seen": 18980672,
      "step": 34250
    },
    {
      "epoch": 600.9734513274336,
      "grad_norm": 9.5625,
      "learning_rate": 2.5028664586096485e-06,
      "loss": 1.0553,
      "num_input_tokens_seen": 18983280,
      "step": 34255
    },
    {
      "epoch": 601.0530973451328,
      "grad_norm": 6.4375,
      "learning_rate": 2.498586530949881e-06,
      "loss": 1.0927,
      "num_input_tokens_seen": 18985576,
      "step": 34260
    },
    {
      "epoch": 601.141592920354,
      "grad_norm": 11.4375,
      "learning_rate": 2.4943100732908427e-06,
      "loss": 0.996,
      "num_input_tokens_seen": 18988152,
      "step": 34265
    },
    {
      "epoch": 601.2300884955753,
      "grad_norm": 8.125,
      "learning_rate": 2.4900370862920188e-06,
      "loss": 1.1721,
      "num_input_tokens_seen": 18991048,
      "step": 34270
    },
    {
      "epoch": 601.3185840707964,
      "grad_norm": 7.09375,
      "learning_rate": 2.4857675706123518e-06,
      "loss": 0.9924,
      "num_input_tokens_seen": 18993832,
      "step": 34275
    },
    {
      "epoch": 601.4070796460177,
      "grad_norm": 9.9375,
      "learning_rate": 2.4815015269102543e-06,
      "loss": 1.1157,
      "num_input_tokens_seen": 18996616,
      "step": 34280
    },
    {
      "epoch": 601.4955752212389,
      "grad_norm": 9.125,
      "learning_rate": 2.477238955843611e-06,
      "loss": 1.0823,
      "num_input_tokens_seen": 18999528,
      "step": 34285
    },
    {
      "epoch": 601.5840707964602,
      "grad_norm": 8.625,
      "learning_rate": 2.4729798580697573e-06,
      "loss": 1.0436,
      "num_input_tokens_seen": 19002984,
      "step": 34290
    },
    {
      "epoch": 601.6725663716815,
      "grad_norm": 8.625,
      "learning_rate": 2.4687242342455034e-06,
      "loss": 1.153,
      "num_input_tokens_seen": 19005272,
      "step": 34295
    },
    {
      "epoch": 601.7610619469026,
      "grad_norm": 8.4375,
      "learning_rate": 2.4644720850271196e-06,
      "loss": 1.141,
      "num_input_tokens_seen": 19008008,
      "step": 34300
    },
    {
      "epoch": 601.8495575221239,
      "grad_norm": 6.4375,
      "learning_rate": 2.4602234110703364e-06,
      "loss": 0.9447,
      "num_input_tokens_seen": 19010760,
      "step": 34305
    },
    {
      "epoch": 601.9380530973451,
      "grad_norm": 8.9375,
      "learning_rate": 2.4559782130303576e-06,
      "loss": 0.9395,
      "num_input_tokens_seen": 19013672,
      "step": 34310
    },
    {
      "epoch": 602.0176991150443,
      "grad_norm": 7.9375,
      "learning_rate": 2.451736491561843e-06,
      "loss": 0.954,
      "num_input_tokens_seen": 19016048,
      "step": 34315
    },
    {
      "epoch": 602.1061946902655,
      "grad_norm": 7.40625,
      "learning_rate": 2.4474982473189163e-06,
      "loss": 1.0193,
      "num_input_tokens_seen": 19018496,
      "step": 34320
    },
    {
      "epoch": 602.1946902654868,
      "grad_norm": 8.0625,
      "learning_rate": 2.4432634809551796e-06,
      "loss": 1.0267,
      "num_input_tokens_seen": 19021488,
      "step": 34325
    },
    {
      "epoch": 602.2831858407079,
      "grad_norm": 7.875,
      "learning_rate": 2.439032193123675e-06,
      "loss": 1.1742,
      "num_input_tokens_seen": 19024272,
      "step": 34330
    },
    {
      "epoch": 602.3716814159292,
      "grad_norm": 8.125,
      "learning_rate": 2.4348043844769297e-06,
      "loss": 1.0027,
      "num_input_tokens_seen": 19027088,
      "step": 34335
    },
    {
      "epoch": 602.4601769911504,
      "grad_norm": 9.8125,
      "learning_rate": 2.4305800556669146e-06,
      "loss": 1.0802,
      "num_input_tokens_seen": 19029536,
      "step": 34340
    },
    {
      "epoch": 602.5486725663717,
      "grad_norm": 7.5,
      "learning_rate": 2.426359207345083e-06,
      "loss": 1.0555,
      "num_input_tokens_seen": 19032064,
      "step": 34345
    },
    {
      "epoch": 602.637168141593,
      "grad_norm": 6.25,
      "learning_rate": 2.4221418401623396e-06,
      "loss": 0.9882,
      "num_input_tokens_seen": 19035136,
      "step": 34350
    },
    {
      "epoch": 602.7256637168142,
      "grad_norm": 7.25,
      "learning_rate": 2.4179279547690557e-06,
      "loss": 1.1412,
      "num_input_tokens_seen": 19038016,
      "step": 34355
    },
    {
      "epoch": 602.8141592920354,
      "grad_norm": 7.96875,
      "learning_rate": 2.413717551815062e-06,
      "loss": 1.0583,
      "num_input_tokens_seen": 19040832,
      "step": 34360
    },
    {
      "epoch": 602.9026548672566,
      "grad_norm": 6.6875,
      "learning_rate": 2.409510631949666e-06,
      "loss": 0.9721,
      "num_input_tokens_seen": 19044240,
      "step": 34365
    },
    {
      "epoch": 602.9911504424779,
      "grad_norm": 7.59375,
      "learning_rate": 2.405307195821618e-06,
      "loss": 0.9857,
      "num_input_tokens_seen": 19046640,
      "step": 34370
    },
    {
      "epoch": 603.070796460177,
      "grad_norm": 7.96875,
      "learning_rate": 2.4011072440791372e-06,
      "loss": 1.463,
      "num_input_tokens_seen": 19048584,
      "step": 34375
    },
    {
      "epoch": 603.1592920353983,
      "grad_norm": 6.4375,
      "learning_rate": 2.3969107773699233e-06,
      "loss": 1.0217,
      "num_input_tokens_seen": 19051240,
      "step": 34380
    },
    {
      "epoch": 603.2477876106195,
      "grad_norm": 9.4375,
      "learning_rate": 2.3927177963411096e-06,
      "loss": 0.9198,
      "num_input_tokens_seen": 19054152,
      "step": 34385
    },
    {
      "epoch": 603.3362831858407,
      "grad_norm": 8.8125,
      "learning_rate": 2.3885283016393144e-06,
      "loss": 1.1533,
      "num_input_tokens_seen": 19057112,
      "step": 34390
    },
    {
      "epoch": 603.4247787610619,
      "grad_norm": 6.53125,
      "learning_rate": 2.3843422939106076e-06,
      "loss": 0.8824,
      "num_input_tokens_seen": 19060216,
      "step": 34395
    },
    {
      "epoch": 603.5132743362832,
      "grad_norm": 8.25,
      "learning_rate": 2.380159773800525e-06,
      "loss": 0.9019,
      "num_input_tokens_seen": 19063208,
      "step": 34400
    },
    {
      "epoch": 603.5132743362832,
      "eval_loss": 1.0310944318771362,
      "eval_runtime": 0.9369,
      "eval_samples_per_second": 26.685,
      "eval_steps_per_second": 13.876,
      "num_input_tokens_seen": 19063208,
      "step": 34400
    },
    {
      "epoch": 603.6017699115044,
      "grad_norm": 9.375,
      "learning_rate": 2.3759807419540675e-06,
      "loss": 0.9827,
      "num_input_tokens_seen": 19065992,
      "step": 34405
    },
    {
      "epoch": 603.6902654867257,
      "grad_norm": 8.1875,
      "learning_rate": 2.3718051990156835e-06,
      "loss": 1.0913,
      "num_input_tokens_seen": 19068520,
      "step": 34410
    },
    {
      "epoch": 603.7787610619469,
      "grad_norm": 7.84375,
      "learning_rate": 2.367633145629311e-06,
      "loss": 1.1885,
      "num_input_tokens_seen": 19071192,
      "step": 34415
    },
    {
      "epoch": 603.8672566371681,
      "grad_norm": 9.1875,
      "learning_rate": 2.363464582438316e-06,
      "loss": 1.2651,
      "num_input_tokens_seen": 19074152,
      "step": 34420
    },
    {
      "epoch": 603.9557522123894,
      "grad_norm": 8.3125,
      "learning_rate": 2.3592995100855526e-06,
      "loss": 1.0827,
      "num_input_tokens_seen": 19077112,
      "step": 34425
    },
    {
      "epoch": 604.0353982300885,
      "grad_norm": 9.75,
      "learning_rate": 2.3551379292133273e-06,
      "loss": 0.9651,
      "num_input_tokens_seen": 19079744,
      "step": 34430
    },
    {
      "epoch": 604.1238938053098,
      "grad_norm": 8.3125,
      "learning_rate": 2.3509798404634047e-06,
      "loss": 1.022,
      "num_input_tokens_seen": 19082768,
      "step": 34435
    },
    {
      "epoch": 604.212389380531,
      "grad_norm": 6.5625,
      "learning_rate": 2.346825244477019e-06,
      "loss": 0.9066,
      "num_input_tokens_seen": 19086032,
      "step": 34440
    },
    {
      "epoch": 604.3008849557522,
      "grad_norm": 6.90625,
      "learning_rate": 2.3426741418948545e-06,
      "loss": 0.9763,
      "num_input_tokens_seen": 19088656,
      "step": 34445
    },
    {
      "epoch": 604.3893805309734,
      "grad_norm": 7.0625,
      "learning_rate": 2.3385265333570715e-06,
      "loss": 0.9136,
      "num_input_tokens_seen": 19090896,
      "step": 34450
    },
    {
      "epoch": 604.4778761061947,
      "grad_norm": 8.5625,
      "learning_rate": 2.334382419503278e-06,
      "loss": 0.9835,
      "num_input_tokens_seen": 19093680,
      "step": 34455
    },
    {
      "epoch": 604.566371681416,
      "grad_norm": 8.1875,
      "learning_rate": 2.3302418009725465e-06,
      "loss": 1.1562,
      "num_input_tokens_seen": 19096240,
      "step": 34460
    },
    {
      "epoch": 604.6548672566372,
      "grad_norm": 9.3125,
      "learning_rate": 2.326104678403415e-06,
      "loss": 1.1129,
      "num_input_tokens_seen": 19099152,
      "step": 34465
    },
    {
      "epoch": 604.7433628318585,
      "grad_norm": 8.8125,
      "learning_rate": 2.321971052433883e-06,
      "loss": 1.0596,
      "num_input_tokens_seen": 19102576,
      "step": 34470
    },
    {
      "epoch": 604.8318584070796,
      "grad_norm": 8.625,
      "learning_rate": 2.3178409237014004e-06,
      "loss": 1.2913,
      "num_input_tokens_seen": 19105040,
      "step": 34475
    },
    {
      "epoch": 604.9203539823009,
      "grad_norm": 7.03125,
      "learning_rate": 2.313714292842889e-06,
      "loss": 1.0021,
      "num_input_tokens_seen": 19107632,
      "step": 34480
    },
    {
      "epoch": 605.0,
      "grad_norm": 11.0625,
      "learning_rate": 2.309591160494734e-06,
      "loss": 0.9788,
      "num_input_tokens_seen": 19110056,
      "step": 34485
    },
    {
      "epoch": 605.0884955752213,
      "grad_norm": 8.6875,
      "learning_rate": 2.305471527292763e-06,
      "loss": 1.2068,
      "num_input_tokens_seen": 19112840,
      "step": 34490
    },
    {
      "epoch": 605.1769911504425,
      "grad_norm": 8.5625,
      "learning_rate": 2.3013553938722817e-06,
      "loss": 0.9444,
      "num_input_tokens_seen": 19115512,
      "step": 34495
    },
    {
      "epoch": 605.2654867256637,
      "grad_norm": 8.75,
      "learning_rate": 2.297242760868043e-06,
      "loss": 1.2178,
      "num_input_tokens_seen": 19118184,
      "step": 34500
    },
    {
      "epoch": 605.3539823008849,
      "grad_norm": 13.0,
      "learning_rate": 2.2931336289142735e-06,
      "loss": 0.9862,
      "num_input_tokens_seen": 19120584,
      "step": 34505
    },
    {
      "epoch": 605.4424778761062,
      "grad_norm": 8.5,
      "learning_rate": 2.289027998644655e-06,
      "loss": 0.9657,
      "num_input_tokens_seen": 19123352,
      "step": 34510
    },
    {
      "epoch": 605.5309734513274,
      "grad_norm": 7.15625,
      "learning_rate": 2.2849258706923228e-06,
      "loss": 0.9629,
      "num_input_tokens_seen": 19126584,
      "step": 34515
    },
    {
      "epoch": 605.6194690265487,
      "grad_norm": 14.5625,
      "learning_rate": 2.2808272456898705e-06,
      "loss": 1.08,
      "num_input_tokens_seen": 19129832,
      "step": 34520
    },
    {
      "epoch": 605.70796460177,
      "grad_norm": 7.9375,
      "learning_rate": 2.2767321242693707e-06,
      "loss": 1.1757,
      "num_input_tokens_seen": 19132664,
      "step": 34525
    },
    {
      "epoch": 605.7964601769911,
      "grad_norm": 7.9375,
      "learning_rate": 2.272640507062329e-06,
      "loss": 1.0135,
      "num_input_tokens_seen": 19135288,
      "step": 34530
    },
    {
      "epoch": 605.8849557522124,
      "grad_norm": 8.4375,
      "learning_rate": 2.2685523946997382e-06,
      "loss": 0.9609,
      "num_input_tokens_seen": 19138408,
      "step": 34535
    },
    {
      "epoch": 605.9734513274336,
      "grad_norm": 8.0,
      "learning_rate": 2.2644677878120245e-06,
      "loss": 1.0634,
      "num_input_tokens_seen": 19141016,
      "step": 34540
    },
    {
      "epoch": 606.0530973451328,
      "grad_norm": 8.5,
      "learning_rate": 2.2603866870290897e-06,
      "loss": 0.8499,
      "num_input_tokens_seen": 19143248,
      "step": 34545
    },
    {
      "epoch": 606.141592920354,
      "grad_norm": 10.0,
      "learning_rate": 2.256309092980294e-06,
      "loss": 1.1463,
      "num_input_tokens_seen": 19145712,
      "step": 34550
    },
    {
      "epoch": 606.2300884955753,
      "grad_norm": 12.375,
      "learning_rate": 2.252235006294448e-06,
      "loss": 0.9961,
      "num_input_tokens_seen": 19148704,
      "step": 34555
    },
    {
      "epoch": 606.3185840707964,
      "grad_norm": 8.625,
      "learning_rate": 2.2481644275998333e-06,
      "loss": 0.8113,
      "num_input_tokens_seen": 19152032,
      "step": 34560
    },
    {
      "epoch": 606.4070796460177,
      "grad_norm": 9.625,
      "learning_rate": 2.2440973575241832e-06,
      "loss": 1.2796,
      "num_input_tokens_seen": 19155136,
      "step": 34565
    },
    {
      "epoch": 606.4955752212389,
      "grad_norm": 6.46875,
      "learning_rate": 2.240033796694685e-06,
      "loss": 1.0327,
      "num_input_tokens_seen": 19157584,
      "step": 34570
    },
    {
      "epoch": 606.5840707964602,
      "grad_norm": 6.9375,
      "learning_rate": 2.235973745737999e-06,
      "loss": 1.0347,
      "num_input_tokens_seen": 19160448,
      "step": 34575
    },
    {
      "epoch": 606.6725663716815,
      "grad_norm": 8.1875,
      "learning_rate": 2.2319172052802263e-06,
      "loss": 1.0897,
      "num_input_tokens_seen": 19162944,
      "step": 34580
    },
    {
      "epoch": 606.7610619469026,
      "grad_norm": 7.0625,
      "learning_rate": 2.2278641759469477e-06,
      "loss": 1.0937,
      "num_input_tokens_seen": 19165712,
      "step": 34585
    },
    {
      "epoch": 606.8495575221239,
      "grad_norm": 8.5,
      "learning_rate": 2.2238146583631825e-06,
      "loss": 1.0893,
      "num_input_tokens_seen": 19169184,
      "step": 34590
    },
    {
      "epoch": 606.9380530973451,
      "grad_norm": 5.875,
      "learning_rate": 2.2197686531534256e-06,
      "loss": 1.0278,
      "num_input_tokens_seen": 19171760,
      "step": 34595
    },
    {
      "epoch": 607.0176991150443,
      "grad_norm": 10.125,
      "learning_rate": 2.2157261609416087e-06,
      "loss": 1.1367,
      "num_input_tokens_seen": 19173736,
      "step": 34600
    },
    {
      "epoch": 607.0176991150443,
      "eval_loss": 1.0310944318771362,
      "eval_runtime": 0.936,
      "eval_samples_per_second": 26.71,
      "eval_steps_per_second": 13.889,
      "num_input_tokens_seen": 19173736,
      "step": 34600
    },
    {
      "epoch": 607.1061946902655,
      "grad_norm": 8.8125,
      "learning_rate": 2.211687182351149e-06,
      "loss": 1.0679,
      "num_input_tokens_seen": 19176232,
      "step": 34605
    },
    {
      "epoch": 607.1946902654868,
      "grad_norm": 8.1875,
      "learning_rate": 2.2076517180048993e-06,
      "loss": 1.0352,
      "num_input_tokens_seen": 19179016,
      "step": 34610
    },
    {
      "epoch": 607.2831858407079,
      "grad_norm": 11.125,
      "learning_rate": 2.2036197685251834e-06,
      "loss": 0.9892,
      "num_input_tokens_seen": 19182008,
      "step": 34615
    },
    {
      "epoch": 607.3716814159292,
      "grad_norm": 8.5625,
      "learning_rate": 2.199591334533771e-06,
      "loss": 0.9997,
      "num_input_tokens_seen": 19184632,
      "step": 34620
    },
    {
      "epoch": 607.4601769911504,
      "grad_norm": 10.3125,
      "learning_rate": 2.1955664166519036e-06,
      "loss": 1.0595,
      "num_input_tokens_seen": 19187160,
      "step": 34625
    },
    {
      "epoch": 607.5486725663717,
      "grad_norm": 9.3125,
      "learning_rate": 2.1915450155002793e-06,
      "loss": 1.1746,
      "num_input_tokens_seen": 19189992,
      "step": 34630
    },
    {
      "epoch": 607.637168141593,
      "grad_norm": 10.625,
      "learning_rate": 2.187527131699038e-06,
      "loss": 1.0576,
      "num_input_tokens_seen": 19193256,
      "step": 34635
    },
    {
      "epoch": 607.7256637168142,
      "grad_norm": 7.875,
      "learning_rate": 2.18351276586779e-06,
      "loss": 1.0407,
      "num_input_tokens_seen": 19195912,
      "step": 34640
    },
    {
      "epoch": 607.8141592920354,
      "grad_norm": 7.71875,
      "learning_rate": 2.1795019186256092e-06,
      "loss": 0.8777,
      "num_input_tokens_seen": 19198664,
      "step": 34645
    },
    {
      "epoch": 607.9026548672566,
      "grad_norm": 8.875,
      "learning_rate": 2.1754945905910094e-06,
      "loss": 1.0893,
      "num_input_tokens_seen": 19201800,
      "step": 34650
    },
    {
      "epoch": 607.9911504424779,
      "grad_norm": 7.09375,
      "learning_rate": 2.171490782381977e-06,
      "loss": 1.1282,
      "num_input_tokens_seen": 19204888,
      "step": 34655
    },
    {
      "epoch": 608.070796460177,
      "grad_norm": 5.6875,
      "learning_rate": 2.1674904946159425e-06,
      "loss": 0.9859,
      "num_input_tokens_seen": 19207128,
      "step": 34660
    },
    {
      "epoch": 608.1592920353983,
      "grad_norm": 14.6875,
      "learning_rate": 2.16349372790981e-06,
      "loss": 0.9165,
      "num_input_tokens_seen": 19209784,
      "step": 34665
    },
    {
      "epoch": 608.2477876106195,
      "grad_norm": 7.90625,
      "learning_rate": 2.159500482879928e-06,
      "loss": 1.004,
      "num_input_tokens_seen": 19213112,
      "step": 34670
    },
    {
      "epoch": 608.3362831858407,
      "grad_norm": 8.0,
      "learning_rate": 2.155510760142096e-06,
      "loss": 1.1222,
      "num_input_tokens_seen": 19216136,
      "step": 34675
    },
    {
      "epoch": 608.4247787610619,
      "grad_norm": 8.8125,
      "learning_rate": 2.151524560311588e-06,
      "loss": 0.9517,
      "num_input_tokens_seen": 19219368,
      "step": 34680
    },
    {
      "epoch": 608.5132743362832,
      "grad_norm": 6.875,
      "learning_rate": 2.147541884003129e-06,
      "loss": 0.9271,
      "num_input_tokens_seen": 19221800,
      "step": 34685
    },
    {
      "epoch": 608.6017699115044,
      "grad_norm": 10.75,
      "learning_rate": 2.1435627318308895e-06,
      "loss": 1.2728,
      "num_input_tokens_seen": 19224648,
      "step": 34690
    },
    {
      "epoch": 608.6902654867257,
      "grad_norm": 9.9375,
      "learning_rate": 2.139587104408511e-06,
      "loss": 1.1563,
      "num_input_tokens_seen": 19226984,
      "step": 34695
    },
    {
      "epoch": 608.7787610619469,
      "grad_norm": 8.0,
      "learning_rate": 2.1356150023490783e-06,
      "loss": 1.1116,
      "num_input_tokens_seen": 19229464,
      "step": 34700
    },
    {
      "epoch": 608.8672566371681,
      "grad_norm": 8.4375,
      "learning_rate": 2.1316464262651464e-06,
      "loss": 1.0022,
      "num_input_tokens_seen": 19232392,
      "step": 34705
    },
    {
      "epoch": 608.9557522123894,
      "grad_norm": 8.375,
      "learning_rate": 2.1276813767687224e-06,
      "loss": 1.1509,
      "num_input_tokens_seen": 19235176,
      "step": 34710
    },
    {
      "epoch": 609.0353982300885,
      "grad_norm": 11.625,
      "learning_rate": 2.123719854471254e-06,
      "loss": 0.8775,
      "num_input_tokens_seen": 19237368,
      "step": 34715
    },
    {
      "epoch": 609.1238938053098,
      "grad_norm": 10.0625,
      "learning_rate": 2.119761859983668e-06,
      "loss": 1.1675,
      "num_input_tokens_seen": 19240456,
      "step": 34720
    },
    {
      "epoch": 609.212389380531,
      "grad_norm": 7.84375,
      "learning_rate": 2.1158073939163386e-06,
      "loss": 1.0421,
      "num_input_tokens_seen": 19243416,
      "step": 34725
    },
    {
      "epoch": 609.3008849557522,
      "grad_norm": 9.8125,
      "learning_rate": 2.111856456879088e-06,
      "loss": 1.0542,
      "num_input_tokens_seen": 19246504,
      "step": 34730
    },
    {
      "epoch": 609.3893805309734,
      "grad_norm": 7.125,
      "learning_rate": 2.1079090494811993e-06,
      "loss": 0.8328,
      "num_input_tokens_seen": 19249592,
      "step": 34735
    },
    {
      "epoch": 609.4778761061947,
      "grad_norm": 9.9375,
      "learning_rate": 2.103965172331418e-06,
      "loss": 1.0251,
      "num_input_tokens_seen": 19252856,
      "step": 34740
    },
    {
      "epoch": 609.566371681416,
      "grad_norm": 7.15625,
      "learning_rate": 2.100024826037933e-06,
      "loss": 0.8667,
      "num_input_tokens_seen": 19255400,
      "step": 34745
    },
    {
      "epoch": 609.6548672566372,
      "grad_norm": 6.75,
      "learning_rate": 2.0960880112084027e-06,
      "loss": 0.9015,
      "num_input_tokens_seen": 19258040,
      "step": 34750
    },
    {
      "epoch": 609.7433628318585,
      "grad_norm": 8.5,
      "learning_rate": 2.092154728449927e-06,
      "loss": 1.1133,
      "num_input_tokens_seen": 19260552,
      "step": 34755
    },
    {
      "epoch": 609.8318584070796,
      "grad_norm": 8.0625,
      "learning_rate": 2.0882249783690687e-06,
      "loss": 1.1752,
      "num_input_tokens_seen": 19263336,
      "step": 34760
    },
    {
      "epoch": 609.9203539823009,
      "grad_norm": 10.625,
      "learning_rate": 2.084298761571851e-06,
      "loss": 1.1933,
      "num_input_tokens_seen": 19265992,
      "step": 34765
    },
    {
      "epoch": 610.0,
      "grad_norm": 11.3125,
      "learning_rate": 2.080376078663737e-06,
      "loss": 0.9712,
      "num_input_tokens_seen": 19268296,
      "step": 34770
    },
    {
      "epoch": 610.0884955752213,
      "grad_norm": 8.4375,
      "learning_rate": 2.0764569302496593e-06,
      "loss": 0.9445,
      "num_input_tokens_seen": 19271656,
      "step": 34775
    },
    {
      "epoch": 610.1769911504425,
      "grad_norm": 9.125,
      "learning_rate": 2.0725413169339957e-06,
      "loss": 1.1357,
      "num_input_tokens_seen": 19274472,
      "step": 34780
    },
    {
      "epoch": 610.2654867256637,
      "grad_norm": 9.0625,
      "learning_rate": 2.068629239320588e-06,
      "loss": 1.1327,
      "num_input_tokens_seen": 19277096,
      "step": 34785
    },
    {
      "epoch": 610.3539823008849,
      "grad_norm": 9.5625,
      "learning_rate": 2.064720698012726e-06,
      "loss": 1.238,
      "num_input_tokens_seen": 19279896,
      "step": 34790
    },
    {
      "epoch": 610.4424778761062,
      "grad_norm": 8.6875,
      "learning_rate": 2.0608156936131522e-06,
      "loss": 1.0938,
      "num_input_tokens_seen": 19282792,
      "step": 34795
    },
    {
      "epoch": 610.5309734513274,
      "grad_norm": 8.9375,
      "learning_rate": 2.056914226724074e-06,
      "loss": 1.007,
      "num_input_tokens_seen": 19285352,
      "step": 34800
    },
    {
      "epoch": 610.5309734513274,
      "eval_loss": 1.0310944318771362,
      "eval_runtime": 0.9394,
      "eval_samples_per_second": 26.613,
      "eval_steps_per_second": 13.839,
      "num_input_tokens_seen": 19285352,
      "step": 34800
    },
    {
      "epoch": 610.6194690265487,
      "grad_norm": 9.0,
      "learning_rate": 2.0530162979471385e-06,
      "loss": 1.0645,
      "num_input_tokens_seen": 19287944,
      "step": 34805
    },
    {
      "epoch": 610.70796460177,
      "grad_norm": 6.96875,
      "learning_rate": 2.0491219078834667e-06,
      "loss": 0.9741,
      "num_input_tokens_seen": 19290680,
      "step": 34810
    },
    {
      "epoch": 610.7964601769911,
      "grad_norm": 7.09375,
      "learning_rate": 2.045231057133612e-06,
      "loss": 1.005,
      "num_input_tokens_seen": 19293432,
      "step": 34815
    },
    {
      "epoch": 610.8849557522124,
      "grad_norm": 7.8125,
      "learning_rate": 2.0413437462975944e-06,
      "loss": 1.1257,
      "num_input_tokens_seen": 19296216,
      "step": 34820
    },
    {
      "epoch": 610.9734513274336,
      "grad_norm": 6.34375,
      "learning_rate": 2.0374599759748843e-06,
      "loss": 0.9832,
      "num_input_tokens_seen": 19298808,
      "step": 34825
    },
    {
      "epoch": 611.0530973451328,
      "grad_norm": 12.125,
      "learning_rate": 2.033579746764419e-06,
      "loss": 0.8164,
      "num_input_tokens_seen": 19301368,
      "step": 34830
    },
    {
      "epoch": 611.141592920354,
      "grad_norm": 8.375,
      "learning_rate": 2.029703059264565e-06,
      "loss": 1.0545,
      "num_input_tokens_seen": 19304504,
      "step": 34835
    },
    {
      "epoch": 611.2300884955753,
      "grad_norm": 10.125,
      "learning_rate": 2.02582991407316e-06,
      "loss": 0.9782,
      "num_input_tokens_seen": 19307496,
      "step": 34840
    },
    {
      "epoch": 611.3185840707964,
      "grad_norm": 7.625,
      "learning_rate": 2.0219603117874992e-06,
      "loss": 1.1648,
      "num_input_tokens_seen": 19310152,
      "step": 34845
    },
    {
      "epoch": 611.4070796460177,
      "grad_norm": 7.15625,
      "learning_rate": 2.0180942530043156e-06,
      "loss": 0.9455,
      "num_input_tokens_seen": 19312696,
      "step": 34850
    },
    {
      "epoch": 611.4955752212389,
      "grad_norm": 7.4375,
      "learning_rate": 2.0142317383198107e-06,
      "loss": 1.1487,
      "num_input_tokens_seen": 19315016,
      "step": 34855
    },
    {
      "epoch": 611.5840707964602,
      "grad_norm": 6.84375,
      "learning_rate": 2.0103727683296243e-06,
      "loss": 1.0507,
      "num_input_tokens_seen": 19318120,
      "step": 34860
    },
    {
      "epoch": 611.6725663716815,
      "grad_norm": 8.1875,
      "learning_rate": 2.0065173436288636e-06,
      "loss": 1.0956,
      "num_input_tokens_seen": 19321192,
      "step": 34865
    },
    {
      "epoch": 611.7610619469026,
      "grad_norm": 8.8125,
      "learning_rate": 2.002665464812087e-06,
      "loss": 0.8409,
      "num_input_tokens_seen": 19323880,
      "step": 34870
    },
    {
      "epoch": 611.8495575221239,
      "grad_norm": 8.0625,
      "learning_rate": 1.998817132473291e-06,
      "loss": 0.9707,
      "num_input_tokens_seen": 19326920,
      "step": 34875
    },
    {
      "epoch": 611.9380530973451,
      "grad_norm": 12.75,
      "learning_rate": 1.9949723472059507e-06,
      "loss": 1.0586,
      "num_input_tokens_seen": 19330024,
      "step": 34880
    },
    {
      "epoch": 612.0176991150443,
      "grad_norm": 8.875,
      "learning_rate": 1.9911311096029726e-06,
      "loss": 1.203,
      "num_input_tokens_seen": 19332016,
      "step": 34885
    },
    {
      "epoch": 612.1061946902655,
      "grad_norm": 8.375,
      "learning_rate": 1.9872934202567224e-06,
      "loss": 1.0551,
      "num_input_tokens_seen": 19334688,
      "step": 34890
    },
    {
      "epoch": 612.1946902654868,
      "grad_norm": 9.25,
      "learning_rate": 1.9834592797590257e-06,
      "loss": 1.0918,
      "num_input_tokens_seen": 19337600,
      "step": 34895
    },
    {
      "epoch": 612.2831858407079,
      "grad_norm": 6.0,
      "learning_rate": 1.979628688701149e-06,
      "loss": 1.1435,
      "num_input_tokens_seen": 19340144,
      "step": 34900
    },
    {
      "epoch": 612.3716814159292,
      "grad_norm": 8.875,
      "learning_rate": 1.9758016476738193e-06,
      "loss": 1.0182,
      "num_input_tokens_seen": 19342656,
      "step": 34905
    },
    {
      "epoch": 612.4601769911504,
      "grad_norm": 8.0625,
      "learning_rate": 1.971978157267221e-06,
      "loss": 0.9875,
      "num_input_tokens_seen": 19345344,
      "step": 34910
    },
    {
      "epoch": 612.5486725663717,
      "grad_norm": 10.375,
      "learning_rate": 1.968158218070973e-06,
      "loss": 1.1366,
      "num_input_tokens_seen": 19348944,
      "step": 34915
    },
    {
      "epoch": 612.637168141593,
      "grad_norm": 5.9375,
      "learning_rate": 1.9643418306741682e-06,
      "loss": 0.8122,
      "num_input_tokens_seen": 19352400,
      "step": 34920
    },
    {
      "epoch": 612.7256637168142,
      "grad_norm": 8.5625,
      "learning_rate": 1.9605289956653337e-06,
      "loss": 1.0162,
      "num_input_tokens_seen": 19355056,
      "step": 34925
    },
    {
      "epoch": 612.8141592920354,
      "grad_norm": 8.625,
      "learning_rate": 1.9567197136324626e-06,
      "loss": 1.1361,
      "num_input_tokens_seen": 19357424,
      "step": 34930
    },
    {
      "epoch": 612.9026548672566,
      "grad_norm": 8.875,
      "learning_rate": 1.9529139851629935e-06,
      "loss": 1.0274,
      "num_input_tokens_seen": 19360032,
      "step": 34935
    },
    {
      "epoch": 612.9911504424779,
      "grad_norm": 8.75,
      "learning_rate": 1.949111810843812e-06,
      "loss": 1.0984,
      "num_input_tokens_seen": 19362784,
      "step": 34940
    },
    {
      "epoch": 613.070796460177,
      "grad_norm": 8.8125,
      "learning_rate": 1.9453131912612694e-06,
      "loss": 1.0854,
      "num_input_tokens_seen": 19365048,
      "step": 34945
    },
    {
      "epoch": 613.1592920353983,
      "grad_norm": 5.125,
      "learning_rate": 1.941518127001149e-06,
      "loss": 0.9245,
      "num_input_tokens_seen": 19367720,
      "step": 34950
    },
    {
      "epoch": 613.2477876106195,
      "grad_norm": 9.1875,
      "learning_rate": 1.9377266186487107e-06,
      "loss": 1.1281,
      "num_input_tokens_seen": 19370360,
      "step": 34955
    },
    {
      "epoch": 613.3362831858407,
      "grad_norm": 8.5625,
      "learning_rate": 1.9339386667886483e-06,
      "loss": 1.0654,
      "num_input_tokens_seen": 19373144,
      "step": 34960
    },
    {
      "epoch": 613.4247787610619,
      "grad_norm": 8.5,
      "learning_rate": 1.9301542720051024e-06,
      "loss": 1.2028,
      "num_input_tokens_seen": 19376216,
      "step": 34965
    },
    {
      "epoch": 613.5132743362832,
      "grad_norm": 7.875,
      "learning_rate": 1.926373434881684e-06,
      "loss": 0.8613,
      "num_input_tokens_seen": 19379304,
      "step": 34970
    },
    {
      "epoch": 613.6017699115044,
      "grad_norm": 8.0625,
      "learning_rate": 1.9225961560014468e-06,
      "loss": 1.0091,
      "num_input_tokens_seen": 19381928,
      "step": 34975
    },
    {
      "epoch": 613.6902654867257,
      "grad_norm": 13.0,
      "learning_rate": 1.918822435946885e-06,
      "loss": 0.9885,
      "num_input_tokens_seen": 19384792,
      "step": 34980
    },
    {
      "epoch": 613.7787610619469,
      "grad_norm": 9.5,
      "learning_rate": 1.915052275299961e-06,
      "loss": 1.0331,
      "num_input_tokens_seen": 19387176,
      "step": 34985
    },
    {
      "epoch": 613.8672566371681,
      "grad_norm": 7.75,
      "learning_rate": 1.9112856746420854e-06,
      "loss": 1.2692,
      "num_input_tokens_seen": 19390360,
      "step": 34990
    },
    {
      "epoch": 613.9557522123894,
      "grad_norm": 7.5,
      "learning_rate": 1.907522634554104e-06,
      "loss": 0.9377,
      "num_input_tokens_seen": 19393608,
      "step": 34995
    },
    {
      "epoch": 614.0353982300885,
      "grad_norm": 6.46875,
      "learning_rate": 1.9037631556163337e-06,
      "loss": 1.0601,
      "num_input_tokens_seen": 19395536,
      "step": 35000
    },
    {
      "epoch": 614.0353982300885,
      "eval_loss": 1.0310944318771362,
      "eval_runtime": 0.9368,
      "eval_samples_per_second": 26.687,
      "eval_steps_per_second": 13.877,
      "num_input_tokens_seen": 19395536,
      "step": 35000
    },
    {
      "epoch": 614.1238938053098,
      "grad_norm": 7.40625,
      "learning_rate": 1.9000072384085272e-06,
      "loss": 0.9831,
      "num_input_tokens_seen": 19398608,
      "step": 35005
    },
    {
      "epoch": 614.212389380531,
      "grad_norm": 6.9375,
      "learning_rate": 1.8962548835098987e-06,
      "loss": 1.0273,
      "num_input_tokens_seen": 19401280,
      "step": 35010
    },
    {
      "epoch": 614.3008849557522,
      "grad_norm": 9.5,
      "learning_rate": 1.8925060914991077e-06,
      "loss": 1.1435,
      "num_input_tokens_seen": 19404672,
      "step": 35015
    },
    {
      "epoch": 614.3893805309734,
      "grad_norm": 7.21875,
      "learning_rate": 1.888760862954264e-06,
      "loss": 1.1524,
      "num_input_tokens_seen": 19407552,
      "step": 35020
    },
    {
      "epoch": 614.4778761061947,
      "grad_norm": 6.5,
      "learning_rate": 1.8850191984529309e-06,
      "loss": 1.0735,
      "num_input_tokens_seen": 19410128,
      "step": 35025
    },
    {
      "epoch": 614.566371681416,
      "grad_norm": 7.59375,
      "learning_rate": 1.8812810985721186e-06,
      "loss": 1.0514,
      "num_input_tokens_seen": 19412848,
      "step": 35030
    },
    {
      "epoch": 614.6548672566372,
      "grad_norm": 8.8125,
      "learning_rate": 1.8775465638882856e-06,
      "loss": 0.9999,
      "num_input_tokens_seen": 19415136,
      "step": 35035
    },
    {
      "epoch": 614.7433628318585,
      "grad_norm": 8.1875,
      "learning_rate": 1.8738155949773517e-06,
      "loss": 1.1191,
      "num_input_tokens_seen": 19418176,
      "step": 35040
    },
    {
      "epoch": 614.8318584070796,
      "grad_norm": 8.6875,
      "learning_rate": 1.8700881924146707e-06,
      "loss": 0.9365,
      "num_input_tokens_seen": 19420960,
      "step": 35045
    },
    {
      "epoch": 614.9203539823009,
      "grad_norm": 8.375,
      "learning_rate": 1.8663643567750577e-06,
      "loss": 1.1491,
      "num_input_tokens_seen": 19423648,
      "step": 35050
    },
    {
      "epoch": 615.0,
      "grad_norm": 8.8125,
      "learning_rate": 1.8626440886327813e-06,
      "loss": 1.0115,
      "num_input_tokens_seen": 19426592,
      "step": 35055
    },
    {
      "epoch": 615.0884955752213,
      "grad_norm": 8.0,
      "learning_rate": 1.8589273885615432e-06,
      "loss": 1.0953,
      "num_input_tokens_seen": 19429440,
      "step": 35060
    },
    {
      "epoch": 615.1769911504425,
      "grad_norm": 8.3125,
      "learning_rate": 1.8552142571345133e-06,
      "loss": 0.9189,
      "num_input_tokens_seen": 19432272,
      "step": 35065
    },
    {
      "epoch": 615.2654867256637,
      "grad_norm": 8.375,
      "learning_rate": 1.8515046949243025e-06,
      "loss": 1.0287,
      "num_input_tokens_seen": 19434704,
      "step": 35070
    },
    {
      "epoch": 615.3539823008849,
      "grad_norm": 8.8125,
      "learning_rate": 1.8477987025029674e-06,
      "loss": 1.0286,
      "num_input_tokens_seen": 19437312,
      "step": 35075
    },
    {
      "epoch": 615.4424778761062,
      "grad_norm": 6.875,
      "learning_rate": 1.8440962804420232e-06,
      "loss": 0.7585,
      "num_input_tokens_seen": 19440240,
      "step": 35080
    },
    {
      "epoch": 615.5309734513274,
      "grad_norm": 14.1875,
      "learning_rate": 1.8403974293124265e-06,
      "loss": 1.1806,
      "num_input_tokens_seen": 19443152,
      "step": 35085
    },
    {
      "epoch": 615.6194690265487,
      "grad_norm": 7.9375,
      "learning_rate": 1.8367021496845854e-06,
      "loss": 1.141,
      "num_input_tokens_seen": 19446288,
      "step": 35090
    },
    {
      "epoch": 615.70796460177,
      "grad_norm": 11.0625,
      "learning_rate": 1.8330104421283662e-06,
      "loss": 1.2286,
      "num_input_tokens_seen": 19448960,
      "step": 35095
    },
    {
      "epoch": 615.7964601769911,
      "grad_norm": 8.0625,
      "learning_rate": 1.8293223072130717e-06,
      "loss": 1.1206,
      "num_input_tokens_seen": 19451440,
      "step": 35100
    },
    {
      "epoch": 615.8849557522124,
      "grad_norm": 8.125,
      "learning_rate": 1.8256377455074525e-06,
      "loss": 1.0944,
      "num_input_tokens_seen": 19454192,
      "step": 35105
    },
    {
      "epoch": 615.9734513274336,
      "grad_norm": 9.375,
      "learning_rate": 1.8219567575797263e-06,
      "loss": 0.9762,
      "num_input_tokens_seen": 19457296,
      "step": 35110
    },
    {
      "epoch": 616.0530973451328,
      "grad_norm": 7.34375,
      "learning_rate": 1.8182793439975365e-06,
      "loss": 1.0237,
      "num_input_tokens_seen": 19459744,
      "step": 35115
    },
    {
      "epoch": 616.141592920354,
      "grad_norm": 8.0,
      "learning_rate": 1.8146055053279958e-06,
      "loss": 1.0691,
      "num_input_tokens_seen": 19463200,
      "step": 35120
    },
    {
      "epoch": 616.2300884955753,
      "grad_norm": 9.1875,
      "learning_rate": 1.8109352421376486e-06,
      "loss": 1.1477,
      "num_input_tokens_seen": 19465744,
      "step": 35125
    },
    {
      "epoch": 616.3185840707964,
      "grad_norm": 7.15625,
      "learning_rate": 1.8072685549924972e-06,
      "loss": 0.9321,
      "num_input_tokens_seen": 19468304,
      "step": 35130
    },
    {
      "epoch": 616.4070796460177,
      "grad_norm": 8.8125,
      "learning_rate": 1.8036054444579982e-06,
      "loss": 1.1656,
      "num_input_tokens_seen": 19471360,
      "step": 35135
    },
    {
      "epoch": 616.4955752212389,
      "grad_norm": 8.25,
      "learning_rate": 1.7999459110990407e-06,
      "loss": 1.0722,
      "num_input_tokens_seen": 19473712,
      "step": 35140
    },
    {
      "epoch": 616.5840707964602,
      "grad_norm": 5.28125,
      "learning_rate": 1.7962899554799712e-06,
      "loss": 0.8479,
      "num_input_tokens_seen": 19476464,
      "step": 35145
    },
    {
      "epoch": 616.6725663716815,
      "grad_norm": 8.5,
      "learning_rate": 1.7926375781645937e-06,
      "loss": 0.8356,
      "num_input_tokens_seen": 19479584,
      "step": 35150
    },
    {
      "epoch": 616.7610619469026,
      "grad_norm": 9.9375,
      "learning_rate": 1.7889887797161359e-06,
      "loss": 1.301,
      "num_input_tokens_seen": 19482064,
      "step": 35155
    },
    {
      "epoch": 616.8495575221239,
      "grad_norm": 6.375,
      "learning_rate": 1.7853435606973028e-06,
      "loss": 1.0424,
      "num_input_tokens_seen": 19484912,
      "step": 35160
    },
    {
      "epoch": 616.9380530973451,
      "grad_norm": 8.5,
      "learning_rate": 1.781701921670223e-06,
      "loss": 1.0574,
      "num_input_tokens_seen": 19488016,
      "step": 35165
    },
    {
      "epoch": 617.0176991150443,
      "grad_norm": 7.9375,
      "learning_rate": 1.7780638631964886e-06,
      "loss": 1.1076,
      "num_input_tokens_seen": 19490176,
      "step": 35170
    },
    {
      "epoch": 617.1061946902655,
      "grad_norm": 6.78125,
      "learning_rate": 1.7744293858371314e-06,
      "loss": 1.0276,
      "num_input_tokens_seen": 19492704,
      "step": 35175
    },
    {
      "epoch": 617.1946902654868,
      "grad_norm": 8.75,
      "learning_rate": 1.770798490152631e-06,
      "loss": 1.1014,
      "num_input_tokens_seen": 19495392,
      "step": 35180
    },
    {
      "epoch": 617.2831858407079,
      "grad_norm": 8.6875,
      "learning_rate": 1.767171176702917e-06,
      "loss": 1.1569,
      "num_input_tokens_seen": 19498176,
      "step": 35185
    },
    {
      "epoch": 617.3716814159292,
      "grad_norm": 7.53125,
      "learning_rate": 1.7635474460473755e-06,
      "loss": 0.9032,
      "num_input_tokens_seen": 19500752,
      "step": 35190
    },
    {
      "epoch": 617.4601769911504,
      "grad_norm": 9.25,
      "learning_rate": 1.7599272987448206e-06,
      "loss": 1.0774,
      "num_input_tokens_seen": 19503392,
      "step": 35195
    },
    {
      "epoch": 617.5486725663717,
      "grad_norm": 7.5,
      "learning_rate": 1.7563107353535362e-06,
      "loss": 0.8663,
      "num_input_tokens_seen": 19506864,
      "step": 35200
    },
    {
      "epoch": 617.5486725663717,
      "eval_loss": 1.0310944318771362,
      "eval_runtime": 0.9365,
      "eval_samples_per_second": 26.696,
      "eval_steps_per_second": 13.882,
      "num_input_tokens_seen": 19506864,
      "step": 35200
    },
    {
      "epoch": 617.637168141593,
      "grad_norm": 8.9375,
      "learning_rate": 1.7526977564312263e-06,
      "loss": 1.0231,
      "num_input_tokens_seen": 19510128,
      "step": 35205
    },
    {
      "epoch": 617.7256637168142,
      "grad_norm": 10.1875,
      "learning_rate": 1.7490883625350701e-06,
      "loss": 1.0026,
      "num_input_tokens_seen": 19513056,
      "step": 35210
    },
    {
      "epoch": 617.8141592920354,
      "grad_norm": 10.0,
      "learning_rate": 1.7454825542216807e-06,
      "loss": 1.1169,
      "num_input_tokens_seen": 19515952,
      "step": 35215
    },
    {
      "epoch": 617.9026548672566,
      "grad_norm": 8.9375,
      "learning_rate": 1.7418803320471105e-06,
      "loss": 1.2227,
      "num_input_tokens_seen": 19518704,
      "step": 35220
    },
    {
      "epoch": 617.9911504424779,
      "grad_norm": 11.625,
      "learning_rate": 1.7382816965668737e-06,
      "loss": 1.0743,
      "num_input_tokens_seen": 19521328,
      "step": 35225
    },
    {
      "epoch": 618.070796460177,
      "grad_norm": 8.75,
      "learning_rate": 1.7346866483359285e-06,
      "loss": 0.8734,
      "num_input_tokens_seen": 19524216,
      "step": 35230
    },
    {
      "epoch": 618.1592920353983,
      "grad_norm": 8.25,
      "learning_rate": 1.7310951879086657e-06,
      "loss": 1.0176,
      "num_input_tokens_seen": 19527272,
      "step": 35235
    },
    {
      "epoch": 618.2477876106195,
      "grad_norm": 6.53125,
      "learning_rate": 1.7275073158389471e-06,
      "loss": 1.0918,
      "num_input_tokens_seen": 19529896,
      "step": 35240
    },
    {
      "epoch": 618.3362831858407,
      "grad_norm": 7.875,
      "learning_rate": 1.723923032680061e-06,
      "loss": 1.0064,
      "num_input_tokens_seen": 19532824,
      "step": 35245
    },
    {
      "epoch": 618.4247787610619,
      "grad_norm": 9.125,
      "learning_rate": 1.7203423389847428e-06,
      "loss": 0.9913,
      "num_input_tokens_seen": 19535480,
      "step": 35250
    },
    {
      "epoch": 618.5132743362832,
      "grad_norm": 8.3125,
      "learning_rate": 1.7167652353051928e-06,
      "loss": 1.0604,
      "num_input_tokens_seen": 19537960,
      "step": 35255
    },
    {
      "epoch": 618.6017699115044,
      "grad_norm": 7.5625,
      "learning_rate": 1.7131917221930333e-06,
      "loss": 1.1054,
      "num_input_tokens_seen": 19540312,
      "step": 35260
    },
    {
      "epoch": 618.6902654867257,
      "grad_norm": 8.4375,
      "learning_rate": 1.7096218001993513e-06,
      "loss": 1.0868,
      "num_input_tokens_seen": 19543144,
      "step": 35265
    },
    {
      "epoch": 618.7787610619469,
      "grad_norm": 6.9375,
      "learning_rate": 1.706055469874676e-06,
      "loss": 0.8897,
      "num_input_tokens_seen": 19545816,
      "step": 35270
    },
    {
      "epoch": 618.8672566371681,
      "grad_norm": 10.25,
      "learning_rate": 1.702492731768976e-06,
      "loss": 1.0466,
      "num_input_tokens_seen": 19548776,
      "step": 35275
    },
    {
      "epoch": 618.9557522123894,
      "grad_norm": 9.0,
      "learning_rate": 1.6989335864316724e-06,
      "loss": 1.3099,
      "num_input_tokens_seen": 19551688,
      "step": 35280
    },
    {
      "epoch": 619.0353982300885,
      "grad_norm": 7.6875,
      "learning_rate": 1.6953780344116265e-06,
      "loss": 1.0231,
      "num_input_tokens_seen": 19553696,
      "step": 35285
    },
    {
      "epoch": 619.1238938053098,
      "grad_norm": 7.375,
      "learning_rate": 1.6918260762571497e-06,
      "loss": 1.1313,
      "num_input_tokens_seen": 19556560,
      "step": 35290
    },
    {
      "epoch": 619.212389380531,
      "grad_norm": 7.34375,
      "learning_rate": 1.6882777125160093e-06,
      "loss": 1.0964,
      "num_input_tokens_seen": 19559440,
      "step": 35295
    },
    {
      "epoch": 619.3008849557522,
      "grad_norm": 13.125,
      "learning_rate": 1.6847329437353899e-06,
      "loss": 1.0025,
      "num_input_tokens_seen": 19562256,
      "step": 35300
    },
    {
      "epoch": 619.3893805309734,
      "grad_norm": 10.625,
      "learning_rate": 1.6811917704619511e-06,
      "loss": 0.9558,
      "num_input_tokens_seen": 19564816,
      "step": 35305
    },
    {
      "epoch": 619.4778761061947,
      "grad_norm": 8.1875,
      "learning_rate": 1.67765419324179e-06,
      "loss": 0.8694,
      "num_input_tokens_seen": 19567696,
      "step": 35310
    },
    {
      "epoch": 619.566371681416,
      "grad_norm": 9.875,
      "learning_rate": 1.6741202126204364e-06,
      "loss": 1.0734,
      "num_input_tokens_seen": 19570672,
      "step": 35315
    },
    {
      "epoch": 619.6548672566372,
      "grad_norm": 6.71875,
      "learning_rate": 1.6705898291428767e-06,
      "loss": 1.0262,
      "num_input_tokens_seen": 19574048,
      "step": 35320
    },
    {
      "epoch": 619.7433628318585,
      "grad_norm": 8.5,
      "learning_rate": 1.6670630433535395e-06,
      "loss": 1.0434,
      "num_input_tokens_seen": 19576576,
      "step": 35325
    },
    {
      "epoch": 619.8318584070796,
      "grad_norm": 9.8125,
      "learning_rate": 1.6635398557962979e-06,
      "loss": 0.9617,
      "num_input_tokens_seen": 19579408,
      "step": 35330
    },
    {
      "epoch": 619.9203539823009,
      "grad_norm": 10.3125,
      "learning_rate": 1.660020267014481e-06,
      "loss": 1.1386,
      "num_input_tokens_seen": 19581872,
      "step": 35335
    },
    {
      "epoch": 620.0,
      "grad_norm": 8.0,
      "learning_rate": 1.6565042775508438e-06,
      "loss": 1.2129,
      "num_input_tokens_seen": 19584488,
      "step": 35340
    },
    {
      "epoch": 620.0884955752213,
      "grad_norm": 8.6875,
      "learning_rate": 1.6529918879475997e-06,
      "loss": 1.1049,
      "num_input_tokens_seen": 19587976,
      "step": 35345
    },
    {
      "epoch": 620.1769911504425,
      "grad_norm": 6.25,
      "learning_rate": 1.6494830987464043e-06,
      "loss": 1.0372,
      "num_input_tokens_seen": 19590792,
      "step": 35350
    },
    {
      "epoch": 620.2654867256637,
      "grad_norm": 7.9375,
      "learning_rate": 1.6459779104883555e-06,
      "loss": 0.9967,
      "num_input_tokens_seen": 19593464,
      "step": 35355
    },
    {
      "epoch": 620.3539823008849,
      "grad_norm": 7.96875,
      "learning_rate": 1.6424763237140013e-06,
      "loss": 1.0651,
      "num_input_tokens_seen": 19596024,
      "step": 35360
    },
    {
      "epoch": 620.4424778761062,
      "grad_norm": 8.125,
      "learning_rate": 1.6389783389633207e-06,
      "loss": 1.1045,
      "num_input_tokens_seen": 19598488,
      "step": 35365
    },
    {
      "epoch": 620.5309734513274,
      "grad_norm": 9.6875,
      "learning_rate": 1.6354839567757546e-06,
      "loss": 1.076,
      "num_input_tokens_seen": 19601080,
      "step": 35370
    },
    {
      "epoch": 620.6194690265487,
      "grad_norm": 7.96875,
      "learning_rate": 1.6319931776901831e-06,
      "loss": 0.9769,
      "num_input_tokens_seen": 19603992,
      "step": 35375
    },
    {
      "epoch": 620.70796460177,
      "grad_norm": 8.4375,
      "learning_rate": 1.6285060022449229e-06,
      "loss": 1.0056,
      "num_input_tokens_seen": 19606856,
      "step": 35380
    },
    {
      "epoch": 620.7964601769911,
      "grad_norm": 8.5625,
      "learning_rate": 1.6250224309777434e-06,
      "loss": 1.0779,
      "num_input_tokens_seen": 19609832,
      "step": 35385
    },
    {
      "epoch": 620.8849557522124,
      "grad_norm": 8.5625,
      "learning_rate": 1.6215424644258515e-06,
      "loss": 1.194,
      "num_input_tokens_seen": 19612824,
      "step": 35390
    },
    {
      "epoch": 620.9734513274336,
      "grad_norm": 8.625,
      "learning_rate": 1.6180661031259036e-06,
      "loss": 1.1109,
      "num_input_tokens_seen": 19615272,
      "step": 35395
    },
    {
      "epoch": 621.0530973451328,
      "grad_norm": 9.25,
      "learning_rate": 1.614593347613999e-06,
      "loss": 0.9241,
      "num_input_tokens_seen": 19617648,
      "step": 35400
    },
    {
      "epoch": 621.0530973451328,
      "eval_loss": 1.0310944318771362,
      "eval_runtime": 0.9359,
      "eval_samples_per_second": 26.711,
      "eval_steps_per_second": 13.89,
      "num_input_tokens_seen": 19617648,
      "step": 35400
    },
    {
      "epoch": 621.141592920354,
      "grad_norm": 12.625,
      "learning_rate": 1.6111241984256758e-06,
      "loss": 0.9613,
      "num_input_tokens_seen": 19620128,
      "step": 35405
    },
    {
      "epoch": 621.2300884955753,
      "grad_norm": 7.0625,
      "learning_rate": 1.6076586560959257e-06,
      "loss": 1.099,
      "num_input_tokens_seen": 19623344,
      "step": 35410
    },
    {
      "epoch": 621.3185840707964,
      "grad_norm": 6.9375,
      "learning_rate": 1.604196721159182e-06,
      "loss": 0.9726,
      "num_input_tokens_seen": 19625728,
      "step": 35415
    },
    {
      "epoch": 621.4070796460177,
      "grad_norm": 9.25,
      "learning_rate": 1.6007383941493092e-06,
      "loss": 1.0416,
      "num_input_tokens_seen": 19628640,
      "step": 35420
    },
    {
      "epoch": 621.4955752212389,
      "grad_norm": 8.5,
      "learning_rate": 1.5972836755996285e-06,
      "loss": 0.9145,
      "num_input_tokens_seen": 19631744,
      "step": 35425
    },
    {
      "epoch": 621.5840707964602,
      "grad_norm": 8.0625,
      "learning_rate": 1.5938325660429076e-06,
      "loss": 0.9914,
      "num_input_tokens_seen": 19634288,
      "step": 35430
    },
    {
      "epoch": 621.6725663716815,
      "grad_norm": 7.875,
      "learning_rate": 1.5903850660113378e-06,
      "loss": 1.0856,
      "num_input_tokens_seen": 19637104,
      "step": 35435
    },
    {
      "epoch": 621.7610619469026,
      "grad_norm": 9.125,
      "learning_rate": 1.5869411760365826e-06,
      "loss": 0.9573,
      "num_input_tokens_seen": 19640048,
      "step": 35440
    },
    {
      "epoch": 621.8495575221239,
      "grad_norm": 8.0625,
      "learning_rate": 1.58350089664972e-06,
      "loss": 1.1254,
      "num_input_tokens_seen": 19642528,
      "step": 35445
    },
    {
      "epoch": 621.9380530973451,
      "grad_norm": 9.125,
      "learning_rate": 1.5800642283812865e-06,
      "loss": 1.1667,
      "num_input_tokens_seen": 19646016,
      "step": 35450
    },
    {
      "epoch": 622.0176991150443,
      "grad_norm": 12.125,
      "learning_rate": 1.5766311717612698e-06,
      "loss": 1.1488,
      "num_input_tokens_seen": 19648368,
      "step": 35455
    },
    {
      "epoch": 622.1061946902655,
      "grad_norm": 7.9375,
      "learning_rate": 1.5732017273190818e-06,
      "loss": 0.9311,
      "num_input_tokens_seen": 19651040,
      "step": 35460
    },
    {
      "epoch": 622.1946902654868,
      "grad_norm": 8.75,
      "learning_rate": 1.5697758955835806e-06,
      "loss": 1.0336,
      "num_input_tokens_seen": 19653936,
      "step": 35465
    },
    {
      "epoch": 622.2831858407079,
      "grad_norm": 8.625,
      "learning_rate": 1.566353677083085e-06,
      "loss": 0.8906,
      "num_input_tokens_seen": 19656640,
      "step": 35470
    },
    {
      "epoch": 622.3716814159292,
      "grad_norm": 16.25,
      "learning_rate": 1.562935072345334e-06,
      "loss": 1.286,
      "num_input_tokens_seen": 19659456,
      "step": 35475
    },
    {
      "epoch": 622.4601769911504,
      "grad_norm": 7.59375,
      "learning_rate": 1.5595200818975281e-06,
      "loss": 1.1676,
      "num_input_tokens_seen": 19661920,
      "step": 35480
    },
    {
      "epoch": 622.5486725663717,
      "grad_norm": 9.0,
      "learning_rate": 1.5561087062662905e-06,
      "loss": 1.1661,
      "num_input_tokens_seen": 19665040,
      "step": 35485
    },
    {
      "epoch": 622.637168141593,
      "grad_norm": 14.125,
      "learning_rate": 1.5527009459777087e-06,
      "loss": 1.0345,
      "num_input_tokens_seen": 19667664,
      "step": 35490
    },
    {
      "epoch": 622.7256637168142,
      "grad_norm": 8.375,
      "learning_rate": 1.5492968015572984e-06,
      "loss": 1.0976,
      "num_input_tokens_seen": 19670416,
      "step": 35495
    },
    {
      "epoch": 622.8141592920354,
      "grad_norm": 8.25,
      "learning_rate": 1.5458962735300203e-06,
      "loss": 1.0697,
      "num_input_tokens_seen": 19673168,
      "step": 35500
    },
    {
      "epoch": 622.9026548672566,
      "grad_norm": 8.125,
      "learning_rate": 1.54249936242028e-06,
      "loss": 0.9347,
      "num_input_tokens_seen": 19676352,
      "step": 35505
    },
    {
      "epoch": 622.9911504424779,
      "grad_norm": 6.53125,
      "learning_rate": 1.5391060687519222e-06,
      "loss": 1.0018,
      "num_input_tokens_seen": 19679264,
      "step": 35510
    },
    {
      "epoch": 623.070796460177,
      "grad_norm": 5.28125,
      "learning_rate": 1.5357163930482367e-06,
      "loss": 0.9006,
      "num_input_tokens_seen": 19681456,
      "step": 35515
    },
    {
      "epoch": 623.1592920353983,
      "grad_norm": 8.6875,
      "learning_rate": 1.532330335831955e-06,
      "loss": 1.1556,
      "num_input_tokens_seen": 19684128,
      "step": 35520
    },
    {
      "epoch": 623.2477876106195,
      "grad_norm": 8.5,
      "learning_rate": 1.5289478976252491e-06,
      "loss": 0.9583,
      "num_input_tokens_seen": 19686512,
      "step": 35525
    },
    {
      "epoch": 623.3362831858407,
      "grad_norm": 7.21875,
      "learning_rate": 1.5255690789497345e-06,
      "loss": 0.9029,
      "num_input_tokens_seen": 19689168,
      "step": 35530
    },
    {
      "epoch": 623.4247787610619,
      "grad_norm": 11.0,
      "learning_rate": 1.5221938803264641e-06,
      "loss": 1.0208,
      "num_input_tokens_seen": 19691664,
      "step": 35535
    },
    {
      "epoch": 623.5132743362832,
      "grad_norm": 8.3125,
      "learning_rate": 1.518822302275938e-06,
      "loss": 1.0638,
      "num_input_tokens_seen": 19694832,
      "step": 35540
    },
    {
      "epoch": 623.6017699115044,
      "grad_norm": 8.8125,
      "learning_rate": 1.5154543453180958e-06,
      "loss": 1.0704,
      "num_input_tokens_seen": 19697584,
      "step": 35545
    },
    {
      "epoch": 623.6902654867257,
      "grad_norm": 7.8125,
      "learning_rate": 1.5120900099723167e-06,
      "loss": 1.1286,
      "num_input_tokens_seen": 19700352,
      "step": 35550
    },
    {
      "epoch": 623.7787610619469,
      "grad_norm": 10.3125,
      "learning_rate": 1.5087292967574273e-06,
      "loss": 1.0296,
      "num_input_tokens_seen": 19703568,
      "step": 35555
    },
    {
      "epoch": 623.8672566371681,
      "grad_norm": 8.9375,
      "learning_rate": 1.5053722061916908e-06,
      "loss": 1.2239,
      "num_input_tokens_seen": 19706640,
      "step": 35560
    },
    {
      "epoch": 623.9557522123894,
      "grad_norm": 10.1875,
      "learning_rate": 1.5020187387928124e-06,
      "loss": 1.0643,
      "num_input_tokens_seen": 19709232,
      "step": 35565
    },
    {
      "epoch": 624.0353982300885,
      "grad_norm": 6.5,
      "learning_rate": 1.4986688950779343e-06,
      "loss": 1.0355,
      "num_input_tokens_seen": 19711872,
      "step": 35570
    },
    {
      "epoch": 624.1238938053098,
      "grad_norm": 9.3125,
      "learning_rate": 1.495322675563654e-06,
      "loss": 1.0954,
      "num_input_tokens_seen": 19714400,
      "step": 35575
    },
    {
      "epoch": 624.212389380531,
      "grad_norm": 9.1875,
      "learning_rate": 1.4919800807659922e-06,
      "loss": 1.1873,
      "num_input_tokens_seen": 19717440,
      "step": 35580
    },
    {
      "epoch": 624.3008849557522,
      "grad_norm": 7.71875,
      "learning_rate": 1.4886411112004255e-06,
      "loss": 1.2221,
      "num_input_tokens_seen": 19719984,
      "step": 35585
    },
    {
      "epoch": 624.3893805309734,
      "grad_norm": 8.5625,
      "learning_rate": 1.4853057673818588e-06,
      "loss": 1.0531,
      "num_input_tokens_seen": 19722848,
      "step": 35590
    },
    {
      "epoch": 624.4778761061947,
      "grad_norm": 6.8125,
      "learning_rate": 1.481974049824647e-06,
      "loss": 0.8721,
      "num_input_tokens_seen": 19725504,
      "step": 35595
    },
    {
      "epoch": 624.566371681416,
      "grad_norm": 9.5625,
      "learning_rate": 1.4786459590425849e-06,
      "loss": 0.9727,
      "num_input_tokens_seen": 19728144,
      "step": 35600
    },
    {
      "epoch": 624.566371681416,
      "eval_loss": 1.0310944318771362,
      "eval_runtime": 0.9338,
      "eval_samples_per_second": 26.771,
      "eval_steps_per_second": 13.921,
      "num_input_tokens_seen": 19728144,
      "step": 35600
    },
    {
      "epoch": 624.6548672566372,
      "grad_norm": 6.5,
      "learning_rate": 1.4753214955489036e-06,
      "loss": 0.9268,
      "num_input_tokens_seen": 19731472,
      "step": 35605
    },
    {
      "epoch": 624.7433628318585,
      "grad_norm": 10.625,
      "learning_rate": 1.4720006598562737e-06,
      "loss": 1.0853,
      "num_input_tokens_seen": 19734064,
      "step": 35610
    },
    {
      "epoch": 624.8318584070796,
      "grad_norm": 8.4375,
      "learning_rate": 1.4686834524768185e-06,
      "loss": 0.9994,
      "num_input_tokens_seen": 19736912,
      "step": 35615
    },
    {
      "epoch": 624.9203539823009,
      "grad_norm": 7.5625,
      "learning_rate": 1.4653698739220844e-06,
      "loss": 1.0402,
      "num_input_tokens_seen": 19739664,
      "step": 35620
    },
    {
      "epoch": 625.0,
      "grad_norm": 12.5625,
      "learning_rate": 1.4620599247030715e-06,
      "loss": 1.1112,
      "num_input_tokens_seen": 19742088,
      "step": 35625
    },
    {
      "epoch": 625.0884955752213,
      "grad_norm": 7.96875,
      "learning_rate": 1.4587536053302125e-06,
      "loss": 1.193,
      "num_input_tokens_seen": 19744328,
      "step": 35630
    },
    {
      "epoch": 625.1769911504425,
      "grad_norm": 9.8125,
      "learning_rate": 1.4554509163133862e-06,
      "loss": 1.1065,
      "num_input_tokens_seen": 19747032,
      "step": 35635
    },
    {
      "epoch": 625.2654867256637,
      "grad_norm": 9.5625,
      "learning_rate": 1.4521518581619098e-06,
      "loss": 1.0127,
      "num_input_tokens_seen": 19749880,
      "step": 35640
    },
    {
      "epoch": 625.3539823008849,
      "grad_norm": 7.4375,
      "learning_rate": 1.4488564313845348e-06,
      "loss": 1.053,
      "num_input_tokens_seen": 19752776,
      "step": 35645
    },
    {
      "epoch": 625.4424778761062,
      "grad_norm": 6.96875,
      "learning_rate": 1.4455646364894603e-06,
      "loss": 1.1096,
      "num_input_tokens_seen": 19755512,
      "step": 35650
    },
    {
      "epoch": 625.5309734513274,
      "grad_norm": 8.125,
      "learning_rate": 1.4422764739843247e-06,
      "loss": 1.2383,
      "num_input_tokens_seen": 19758360,
      "step": 35655
    },
    {
      "epoch": 625.6194690265487,
      "grad_norm": 9.75,
      "learning_rate": 1.4389919443762e-06,
      "loss": 0.7419,
      "num_input_tokens_seen": 19761208,
      "step": 35660
    },
    {
      "epoch": 625.70796460177,
      "grad_norm": 9.125,
      "learning_rate": 1.4357110481716063e-06,
      "loss": 0.9195,
      "num_input_tokens_seen": 19763976,
      "step": 35665
    },
    {
      "epoch": 625.7964601769911,
      "grad_norm": 7.875,
      "learning_rate": 1.4324337858764941e-06,
      "loss": 0.8866,
      "num_input_tokens_seen": 19766408,
      "step": 35670
    },
    {
      "epoch": 625.8849557522124,
      "grad_norm": 11.5,
      "learning_rate": 1.4291601579962622e-06,
      "loss": 1.0754,
      "num_input_tokens_seen": 19769688,
      "step": 35675
    },
    {
      "epoch": 625.9734513274336,
      "grad_norm": 8.9375,
      "learning_rate": 1.42589016503574e-06,
      "loss": 1.1602,
      "num_input_tokens_seen": 19772600,
      "step": 35680
    },
    {
      "epoch": 626.0530973451328,
      "grad_norm": 8.875,
      "learning_rate": 1.4226238074992099e-06,
      "loss": 0.9428,
      "num_input_tokens_seen": 19775552,
      "step": 35685
    },
    {
      "epoch": 626.141592920354,
      "grad_norm": 7.875,
      "learning_rate": 1.4193610858903778e-06,
      "loss": 1.0807,
      "num_input_tokens_seen": 19778512,
      "step": 35690
    },
    {
      "epoch": 626.2300884955753,
      "grad_norm": 13.375,
      "learning_rate": 1.416102000712402e-06,
      "loss": 0.8607,
      "num_input_tokens_seen": 19781280,
      "step": 35695
    },
    {
      "epoch": 626.3185840707964,
      "grad_norm": 9.125,
      "learning_rate": 1.4128465524678668e-06,
      "loss": 1.0666,
      "num_input_tokens_seen": 19784048,
      "step": 35700
    },
    {
      "epoch": 626.4070796460177,
      "grad_norm": 8.4375,
      "learning_rate": 1.4095947416588124e-06,
      "loss": 1.139,
      "num_input_tokens_seen": 19786720,
      "step": 35705
    },
    {
      "epoch": 626.4955752212389,
      "grad_norm": 8.9375,
      "learning_rate": 1.4063465687866983e-06,
      "loss": 0.9357,
      "num_input_tokens_seen": 19789440,
      "step": 35710
    },
    {
      "epoch": 626.5840707964602,
      "grad_norm": 8.0,
      "learning_rate": 1.4031020343524438e-06,
      "loss": 1.0425,
      "num_input_tokens_seen": 19791920,
      "step": 35715
    },
    {
      "epoch": 626.6725663716815,
      "grad_norm": 9.0,
      "learning_rate": 1.3998611388563926e-06,
      "loss": 1.1649,
      "num_input_tokens_seen": 19794560,
      "step": 35720
    },
    {
      "epoch": 626.7610619469026,
      "grad_norm": 8.0625,
      "learning_rate": 1.3966238827983314e-06,
      "loss": 1.1557,
      "num_input_tokens_seen": 19797504,
      "step": 35725
    },
    {
      "epoch": 626.8495575221239,
      "grad_norm": 7.78125,
      "learning_rate": 1.393390266677483e-06,
      "loss": 1.0017,
      "num_input_tokens_seen": 19800048,
      "step": 35730
    },
    {
      "epoch": 626.9380530973451,
      "grad_norm": 7.34375,
      "learning_rate": 1.3901602909925204e-06,
      "loss": 1.0729,
      "num_input_tokens_seen": 19802832,
      "step": 35735
    },
    {
      "epoch": 627.0176991150443,
      "grad_norm": 7.1875,
      "learning_rate": 1.3869339562415373e-06,
      "loss": 1.103,
      "num_input_tokens_seen": 19805640,
      "step": 35740
    },
    {
      "epoch": 627.1061946902655,
      "grad_norm": 7.0,
      "learning_rate": 1.38371126292208e-06,
      "loss": 1.0574,
      "num_input_tokens_seen": 19808568,
      "step": 35745
    },
    {
      "epoch": 627.1946902654868,
      "grad_norm": 8.0625,
      "learning_rate": 1.3804922115311286e-06,
      "loss": 1.0084,
      "num_input_tokens_seen": 19811704,
      "step": 35750
    },
    {
      "epoch": 627.2831858407079,
      "grad_norm": 9.4375,
      "learning_rate": 1.3772768025650945e-06,
      "loss": 1.008,
      "num_input_tokens_seen": 19814376,
      "step": 35755
    },
    {
      "epoch": 627.3716814159292,
      "grad_norm": 8.25,
      "learning_rate": 1.3740650365198448e-06,
      "loss": 1.0697,
      "num_input_tokens_seen": 19817224,
      "step": 35760
    },
    {
      "epoch": 627.4601769911504,
      "grad_norm": 8.3125,
      "learning_rate": 1.3708569138906612e-06,
      "loss": 0.9297,
      "num_input_tokens_seen": 19819576,
      "step": 35765
    },
    {
      "epoch": 627.5486725663717,
      "grad_norm": 7.8125,
      "learning_rate": 1.367652435172287e-06,
      "loss": 1.1957,
      "num_input_tokens_seen": 19822360,
      "step": 35770
    },
    {
      "epoch": 627.637168141593,
      "grad_norm": 8.125,
      "learning_rate": 1.364451600858893e-06,
      "loss": 1.054,
      "num_input_tokens_seen": 19825752,
      "step": 35775
    },
    {
      "epoch": 627.7256637168142,
      "grad_norm": 7.28125,
      "learning_rate": 1.3612544114440823e-06,
      "loss": 0.9902,
      "num_input_tokens_seen": 19828584,
      "step": 35780
    },
    {
      "epoch": 627.8141592920354,
      "grad_norm": 7.875,
      "learning_rate": 1.3580608674209072e-06,
      "loss": 1.0397,
      "num_input_tokens_seen": 19830920,
      "step": 35785
    },
    {
      "epoch": 627.9026548672566,
      "grad_norm": 7.6875,
      "learning_rate": 1.3548709692818434e-06,
      "loss": 1.0599,
      "num_input_tokens_seen": 19833512,
      "step": 35790
    },
    {
      "epoch": 627.9911504424779,
      "grad_norm": 8.125,
      "learning_rate": 1.3516847175188223e-06,
      "loss": 1.1208,
      "num_input_tokens_seen": 19836168,
      "step": 35795
    },
    {
      "epoch": 628.070796460177,
      "grad_norm": 8.625,
      "learning_rate": 1.348502112623204e-06,
      "loss": 1.1257,
      "num_input_tokens_seen": 19838296,
      "step": 35800
    },
    {
      "epoch": 628.070796460177,
      "eval_loss": 1.0310944318771362,
      "eval_runtime": 0.9362,
      "eval_samples_per_second": 26.703,
      "eval_steps_per_second": 13.886,
      "num_input_tokens_seen": 19838296,
      "step": 35800
    },
    {
      "epoch": 628.1592920353983,
      "grad_norm": 8.625,
      "learning_rate": 1.3453231550857787e-06,
      "loss": 1.1302,
      "num_input_tokens_seen": 19841432,
      "step": 35805
    },
    {
      "epoch": 628.2477876106195,
      "grad_norm": 11.5625,
      "learning_rate": 1.3421478453967878e-06,
      "loss": 1.0829,
      "num_input_tokens_seen": 19844088,
      "step": 35810
    },
    {
      "epoch": 628.3362831858407,
      "grad_norm": 7.75,
      "learning_rate": 1.3389761840459065e-06,
      "loss": 0.9316,
      "num_input_tokens_seen": 19846584,
      "step": 35815
    },
    {
      "epoch": 628.4247787610619,
      "grad_norm": 9.3125,
      "learning_rate": 1.3358081715222376e-06,
      "loss": 1.1328,
      "num_input_tokens_seen": 19849096,
      "step": 35820
    },
    {
      "epoch": 628.5132743362832,
      "grad_norm": 8.3125,
      "learning_rate": 1.3326438083143295e-06,
      "loss": 0.8875,
      "num_input_tokens_seen": 19851736,
      "step": 35825
    },
    {
      "epoch": 628.6017699115044,
      "grad_norm": 9.875,
      "learning_rate": 1.3294830949101723e-06,
      "loss": 1.1371,
      "num_input_tokens_seen": 19854792,
      "step": 35830
    },
    {
      "epoch": 628.6902654867257,
      "grad_norm": 8.625,
      "learning_rate": 1.3263260317971815e-06,
      "loss": 0.911,
      "num_input_tokens_seen": 19858008,
      "step": 35835
    },
    {
      "epoch": 628.7787610619469,
      "grad_norm": 7.25,
      "learning_rate": 1.3231726194622208e-06,
      "loss": 1.1367,
      "num_input_tokens_seen": 19860520,
      "step": 35840
    },
    {
      "epoch": 628.8672566371681,
      "grad_norm": 8.625,
      "learning_rate": 1.3200228583915814e-06,
      "loss": 1.1662,
      "num_input_tokens_seen": 19863096,
      "step": 35845
    },
    {
      "epoch": 628.9557522123894,
      "grad_norm": 9.8125,
      "learning_rate": 1.3168767490709971e-06,
      "loss": 0.9636,
      "num_input_tokens_seen": 19866200,
      "step": 35850
    },
    {
      "epoch": 629.0353982300885,
      "grad_norm": 8.4375,
      "learning_rate": 1.3137342919856437e-06,
      "loss": 1.0201,
      "num_input_tokens_seen": 19868416,
      "step": 35855
    },
    {
      "epoch": 629.1238938053098,
      "grad_norm": 11.25,
      "learning_rate": 1.310595487620117e-06,
      "loss": 1.0226,
      "num_input_tokens_seen": 19871264,
      "step": 35860
    },
    {
      "epoch": 629.212389380531,
      "grad_norm": 9.0625,
      "learning_rate": 1.3074603364584715e-06,
      "loss": 1.0569,
      "num_input_tokens_seen": 19873728,
      "step": 35865
    },
    {
      "epoch": 629.3008849557522,
      "grad_norm": 12.1875,
      "learning_rate": 1.3043288389841758e-06,
      "loss": 1.0187,
      "num_input_tokens_seen": 19876960,
      "step": 35870
    },
    {
      "epoch": 629.3893805309734,
      "grad_norm": 8.625,
      "learning_rate": 1.3012009956801546e-06,
      "loss": 1.1437,
      "num_input_tokens_seen": 19879952,
      "step": 35875
    },
    {
      "epoch": 629.4778761061947,
      "grad_norm": 7.34375,
      "learning_rate": 1.2980768070287586e-06,
      "loss": 0.7905,
      "num_input_tokens_seen": 19882944,
      "step": 35880
    },
    {
      "epoch": 629.566371681416,
      "grad_norm": 8.25,
      "learning_rate": 1.2949562735117716e-06,
      "loss": 1.1022,
      "num_input_tokens_seen": 19885472,
      "step": 35885
    },
    {
      "epoch": 629.6548672566372,
      "grad_norm": 9.9375,
      "learning_rate": 1.291839395610428e-06,
      "loss": 1.2817,
      "num_input_tokens_seen": 19888416,
      "step": 35890
    },
    {
      "epoch": 629.7433628318585,
      "grad_norm": 7.96875,
      "learning_rate": 1.2887261738053852e-06,
      "loss": 0.9064,
      "num_input_tokens_seen": 19891440,
      "step": 35895
    },
    {
      "epoch": 629.8318584070796,
      "grad_norm": 9.3125,
      "learning_rate": 1.2856166085767396e-06,
      "loss": 1.1234,
      "num_input_tokens_seen": 19893840,
      "step": 35900
    },
    {
      "epoch": 629.9203539823009,
      "grad_norm": 8.75,
      "learning_rate": 1.2825107004040272e-06,
      "loss": 0.9484,
      "num_input_tokens_seen": 19896864,
      "step": 35905
    },
    {
      "epoch": 630.0,
      "grad_norm": 27.5,
      "learning_rate": 1.2794084497662146e-06,
      "loss": 1.5946,
      "num_input_tokens_seen": 19898800,
      "step": 35910
    },
    {
      "epoch": 630.0884955752213,
      "grad_norm": 7.34375,
      "learning_rate": 1.276309857141711e-06,
      "loss": 1.2154,
      "num_input_tokens_seen": 19901424,
      "step": 35915
    },
    {
      "epoch": 630.1769911504425,
      "grad_norm": 7.65625,
      "learning_rate": 1.273214923008359e-06,
      "loss": 0.947,
      "num_input_tokens_seen": 19904032,
      "step": 35920
    },
    {
      "epoch": 630.2654867256637,
      "grad_norm": 9.0625,
      "learning_rate": 1.2701236478434352e-06,
      "loss": 0.9858,
      "num_input_tokens_seen": 19906656,
      "step": 35925
    },
    {
      "epoch": 630.3539823008849,
      "grad_norm": 8.5625,
      "learning_rate": 1.2670360321236502e-06,
      "loss": 1.1498,
      "num_input_tokens_seen": 19909184,
      "step": 35930
    },
    {
      "epoch": 630.4424778761062,
      "grad_norm": 7.4375,
      "learning_rate": 1.2639520763251617e-06,
      "loss": 0.8814,
      "num_input_tokens_seen": 19912160,
      "step": 35935
    },
    {
      "epoch": 630.5309734513274,
      "grad_norm": 9.0625,
      "learning_rate": 1.2608717809235448e-06,
      "loss": 1.0371,
      "num_input_tokens_seen": 19915232,
      "step": 35940
    },
    {
      "epoch": 630.6194690265487,
      "grad_norm": 7.6875,
      "learning_rate": 1.2577951463938282e-06,
      "loss": 0.8765,
      "num_input_tokens_seen": 19918368,
      "step": 35945
    },
    {
      "epoch": 630.70796460177,
      "grad_norm": 9.25,
      "learning_rate": 1.2547221732104569e-06,
      "loss": 1.0173,
      "num_input_tokens_seen": 19920800,
      "step": 35950
    },
    {
      "epoch": 630.7964601769911,
      "grad_norm": 8.5,
      "learning_rate": 1.25165286184733e-06,
      "loss": 1.0461,
      "num_input_tokens_seen": 19923952,
      "step": 35955
    },
    {
      "epoch": 630.8849557522124,
      "grad_norm": 6.8125,
      "learning_rate": 1.248587212777777e-06,
      "loss": 1.0848,
      "num_input_tokens_seen": 19926608,
      "step": 35960
    },
    {
      "epoch": 630.9734513274336,
      "grad_norm": 8.25,
      "learning_rate": 1.2455252264745532e-06,
      "loss": 1.1664,
      "num_input_tokens_seen": 19929424,
      "step": 35965
    },
    {
      "epoch": 631.0530973451328,
      "grad_norm": 6.125,
      "learning_rate": 1.2424669034098528e-06,
      "loss": 1.6141,
      "num_input_tokens_seen": 19931896,
      "step": 35970
    },
    {
      "epoch": 631.141592920354,
      "grad_norm": 9.25,
      "learning_rate": 1.2394122440553185e-06,
      "loss": 1.0309,
      "num_input_tokens_seen": 19935064,
      "step": 35975
    },
    {
      "epoch": 631.2300884955753,
      "grad_norm": 8.1875,
      "learning_rate": 1.2363612488820037e-06,
      "loss": 0.9278,
      "num_input_tokens_seen": 19937640,
      "step": 35980
    },
    {
      "epoch": 631.3185840707964,
      "grad_norm": 10.1875,
      "learning_rate": 1.2333139183604208e-06,
      "loss": 1.0152,
      "num_input_tokens_seen": 19940392,
      "step": 35985
    },
    {
      "epoch": 631.4070796460177,
      "grad_norm": 7.84375,
      "learning_rate": 1.2302702529604998e-06,
      "loss": 1.061,
      "num_input_tokens_seen": 19943336,
      "step": 35990
    },
    {
      "epoch": 631.4955752212389,
      "grad_norm": 9.375,
      "learning_rate": 1.227230253151615e-06,
      "loss": 1.0275,
      "num_input_tokens_seen": 19945720,
      "step": 35995
    },
    {
      "epoch": 631.5840707964602,
      "grad_norm": 8.0625,
      "learning_rate": 1.2241939194025748e-06,
      "loss": 0.9849,
      "num_input_tokens_seen": 19948392,
      "step": 36000
    },
    {
      "epoch": 631.5840707964602,
      "eval_loss": 1.0310944318771362,
      "eval_runtime": 0.9374,
      "eval_samples_per_second": 26.671,
      "eval_steps_per_second": 13.869,
      "num_input_tokens_seen": 19948392,
      "step": 36000
    },
    {
      "epoch": 631.6725663716815,
      "grad_norm": 8.3125,
      "learning_rate": 1.2211612521816156e-06,
      "loss": 1.0275,
      "num_input_tokens_seen": 19951000,
      "step": 36005
    },
    {
      "epoch": 631.7610619469026,
      "grad_norm": 9.0,
      "learning_rate": 1.2181322519564137e-06,
      "loss": 1.215,
      "num_input_tokens_seen": 19953656,
      "step": 36010
    },
    {
      "epoch": 631.8495575221239,
      "grad_norm": 9.875,
      "learning_rate": 1.2151069191940839e-06,
      "loss": 1.0501,
      "num_input_tokens_seen": 19956552,
      "step": 36015
    },
    {
      "epoch": 631.9380530973451,
      "grad_norm": 7.25,
      "learning_rate": 1.2120852543611644e-06,
      "loss": 1.1281,
      "num_input_tokens_seen": 19959704,
      "step": 36020
    },
    {
      "epoch": 632.0176991150443,
      "grad_norm": 9.625,
      "learning_rate": 1.2090672579236379e-06,
      "loss": 1.0361,
      "num_input_tokens_seen": 19962000,
      "step": 36025
    },
    {
      "epoch": 632.1061946902655,
      "grad_norm": 10.375,
      "learning_rate": 1.2060529303469126e-06,
      "loss": 1.2164,
      "num_input_tokens_seen": 19964368,
      "step": 36030
    },
    {
      "epoch": 632.1946902654868,
      "grad_norm": 9.375,
      "learning_rate": 1.2030422720958445e-06,
      "loss": 1.198,
      "num_input_tokens_seen": 19966960,
      "step": 36035
    },
    {
      "epoch": 632.2831858407079,
      "grad_norm": 7.5625,
      "learning_rate": 1.200035283634704e-06,
      "loss": 1.0007,
      "num_input_tokens_seen": 19969968,
      "step": 36040
    },
    {
      "epoch": 632.3716814159292,
      "grad_norm": 6.8125,
      "learning_rate": 1.1970319654272144e-06,
      "loss": 1.0735,
      "num_input_tokens_seen": 19972336,
      "step": 36045
    },
    {
      "epoch": 632.4601769911504,
      "grad_norm": 8.25,
      "learning_rate": 1.1940323179365192e-06,
      "loss": 1.1098,
      "num_input_tokens_seen": 19975568,
      "step": 36050
    },
    {
      "epoch": 632.5486725663717,
      "grad_norm": 10.6875,
      "learning_rate": 1.1910363416252095e-06,
      "loss": 1.0703,
      "num_input_tokens_seen": 19978480,
      "step": 36055
    },
    {
      "epoch": 632.637168141593,
      "grad_norm": 11.375,
      "learning_rate": 1.1880440369552964e-06,
      "loss": 1.0229,
      "num_input_tokens_seen": 19981280,
      "step": 36060
    },
    {
      "epoch": 632.7256637168142,
      "grad_norm": 8.4375,
      "learning_rate": 1.1850554043882328e-06,
      "loss": 0.9035,
      "num_input_tokens_seen": 19984608,
      "step": 36065
    },
    {
      "epoch": 632.8141592920354,
      "grad_norm": 7.21875,
      "learning_rate": 1.1820704443849028e-06,
      "loss": 0.9728,
      "num_input_tokens_seen": 19987296,
      "step": 36070
    },
    {
      "epoch": 632.9026548672566,
      "grad_norm": 9.0,
      "learning_rate": 1.1790891574056219e-06,
      "loss": 0.9765,
      "num_input_tokens_seen": 19990064,
      "step": 36075
    },
    {
      "epoch": 632.9911504424779,
      "grad_norm": 9.9375,
      "learning_rate": 1.1761115439101523e-06,
      "loss": 0.9741,
      "num_input_tokens_seen": 19992896,
      "step": 36080
    },
    {
      "epoch": 633.070796460177,
      "grad_norm": 8.25,
      "learning_rate": 1.1731376043576659e-06,
      "loss": 1.0926,
      "num_input_tokens_seen": 19995776,
      "step": 36085
    },
    {
      "epoch": 633.1592920353983,
      "grad_norm": 7.125,
      "learning_rate": 1.1701673392067875e-06,
      "loss": 0.9709,
      "num_input_tokens_seen": 19998496,
      "step": 36090
    },
    {
      "epoch": 633.2477876106195,
      "grad_norm": 8.1875,
      "learning_rate": 1.1672007489155757e-06,
      "loss": 1.1482,
      "num_input_tokens_seen": 20000992,
      "step": 36095
    },
    {
      "epoch": 633.3362831858407,
      "grad_norm": 9.8125,
      "learning_rate": 1.164237833941506e-06,
      "loss": 0.9792,
      "num_input_tokens_seen": 20003632,
      "step": 36100
    },
    {
      "epoch": 633.4247787610619,
      "grad_norm": 8.1875,
      "learning_rate": 1.1612785947415022e-06,
      "loss": 1.1114,
      "num_input_tokens_seen": 20006288,
      "step": 36105
    },
    {
      "epoch": 633.5132743362832,
      "grad_norm": 7.75,
      "learning_rate": 1.1583230317719185e-06,
      "loss": 0.9988,
      "num_input_tokens_seen": 20009792,
      "step": 36110
    },
    {
      "epoch": 633.6017699115044,
      "grad_norm": 7.96875,
      "learning_rate": 1.1553711454885318e-06,
      "loss": 0.9409,
      "num_input_tokens_seen": 20012752,
      "step": 36115
    },
    {
      "epoch": 633.6902654867257,
      "grad_norm": 10.6875,
      "learning_rate": 1.152422936346567e-06,
      "loss": 1.1778,
      "num_input_tokens_seen": 20015520,
      "step": 36120
    },
    {
      "epoch": 633.7787610619469,
      "grad_norm": 7.8125,
      "learning_rate": 1.1494784048006718e-06,
      "loss": 0.9598,
      "num_input_tokens_seen": 20018064,
      "step": 36125
    },
    {
      "epoch": 633.8672566371681,
      "grad_norm": 7.4375,
      "learning_rate": 1.1465375513049326e-06,
      "loss": 0.968,
      "num_input_tokens_seen": 20020608,
      "step": 36130
    },
    {
      "epoch": 633.9557522123894,
      "grad_norm": 7.46875,
      "learning_rate": 1.1436003763128616e-06,
      "loss": 1.0665,
      "num_input_tokens_seen": 20023296,
      "step": 36135
    },
    {
      "epoch": 634.0353982300885,
      "grad_norm": 8.4375,
      "learning_rate": 1.1406668802774106e-06,
      "loss": 1.2115,
      "num_input_tokens_seen": 20025824,
      "step": 36140
    },
    {
      "epoch": 634.1238938053098,
      "grad_norm": 12.125,
      "learning_rate": 1.137737063650965e-06,
      "loss": 1.0498,
      "num_input_tokens_seen": 20028160,
      "step": 36145
    },
    {
      "epoch": 634.212389380531,
      "grad_norm": 7.28125,
      "learning_rate": 1.1348109268853323e-06,
      "loss": 1.0131,
      "num_input_tokens_seen": 20030896,
      "step": 36150
    },
    {
      "epoch": 634.3008849557522,
      "grad_norm": 7.8125,
      "learning_rate": 1.1318884704317634e-06,
      "loss": 1.148,
      "num_input_tokens_seen": 20033472,
      "step": 36155
    },
    {
      "epoch": 634.3893805309734,
      "grad_norm": 7.75,
      "learning_rate": 1.1289696947409417e-06,
      "loss": 1.0084,
      "num_input_tokens_seen": 20036384,
      "step": 36160
    },
    {
      "epoch": 634.4778761061947,
      "grad_norm": 7.34375,
      "learning_rate": 1.126054600262974e-06,
      "loss": 1.0096,
      "num_input_tokens_seen": 20039328,
      "step": 36165
    },
    {
      "epoch": 634.566371681416,
      "grad_norm": 8.3125,
      "learning_rate": 1.1231431874474064e-06,
      "loss": 1.1097,
      "num_input_tokens_seen": 20041920,
      "step": 36170
    },
    {
      "epoch": 634.6548672566372,
      "grad_norm": 8.4375,
      "learning_rate": 1.12023545674321e-06,
      "loss": 1.1526,
      "num_input_tokens_seen": 20044880,
      "step": 36175
    },
    {
      "epoch": 634.7433628318585,
      "grad_norm": 9.0,
      "learning_rate": 1.117331408598804e-06,
      "loss": 1.1837,
      "num_input_tokens_seen": 20047552,
      "step": 36180
    },
    {
      "epoch": 634.8318584070796,
      "grad_norm": 7.71875,
      "learning_rate": 1.1144310434620191e-06,
      "loss": 0.8716,
      "num_input_tokens_seen": 20050240,
      "step": 36185
    },
    {
      "epoch": 634.9203539823009,
      "grad_norm": 6.03125,
      "learning_rate": 1.1115343617801365e-06,
      "loss": 0.8327,
      "num_input_tokens_seen": 20053632,
      "step": 36190
    },
    {
      "epoch": 635.0,
      "grad_norm": 24.75,
      "learning_rate": 1.1086413639998515e-06,
      "loss": 1.217,
      "num_input_tokens_seen": 20056128,
      "step": 36195
    },
    {
      "epoch": 635.0884955752213,
      "grad_norm": 9.3125,
      "learning_rate": 1.1057520505673103e-06,
      "loss": 0.921,
      "num_input_tokens_seen": 20059232,
      "step": 36200
    },
    {
      "epoch": 635.0884955752213,
      "eval_loss": 1.0310944318771362,
      "eval_runtime": 0.9381,
      "eval_samples_per_second": 26.649,
      "eval_steps_per_second": 13.857,
      "num_input_tokens_seen": 20059232,
      "step": 36200
    },
    {
      "epoch": 635.1769911504425,
      "grad_norm": 8.3125,
      "learning_rate": 1.1028664219280727e-06,
      "loss": 1.0699,
      "num_input_tokens_seen": 20062208,
      "step": 36205
    },
    {
      "epoch": 635.2654867256637,
      "grad_norm": 8.5625,
      "learning_rate": 1.0999844785271468e-06,
      "loss": 1.1582,
      "num_input_tokens_seen": 20064880,
      "step": 36210
    },
    {
      "epoch": 635.3539823008849,
      "grad_norm": 8.0,
      "learning_rate": 1.097106220808955e-06,
      "loss": 1.1023,
      "num_input_tokens_seen": 20067680,
      "step": 36215
    },
    {
      "epoch": 635.4424778761062,
      "grad_norm": 7.46875,
      "learning_rate": 1.0942316492173698e-06,
      "loss": 0.9689,
      "num_input_tokens_seen": 20070432,
      "step": 36220
    },
    {
      "epoch": 635.5309734513274,
      "grad_norm": 9.4375,
      "learning_rate": 1.0913607641956841e-06,
      "loss": 0.9906,
      "num_input_tokens_seen": 20072912,
      "step": 36225
    },
    {
      "epoch": 635.6194690265487,
      "grad_norm": 7.84375,
      "learning_rate": 1.0884935661866213e-06,
      "loss": 0.8741,
      "num_input_tokens_seen": 20075536,
      "step": 36230
    },
    {
      "epoch": 635.70796460177,
      "grad_norm": 8.625,
      "learning_rate": 1.0856300556323418e-06,
      "loss": 0.8579,
      "num_input_tokens_seen": 20078240,
      "step": 36235
    },
    {
      "epoch": 635.7964601769911,
      "grad_norm": 8.1875,
      "learning_rate": 1.0827702329744365e-06,
      "loss": 1.1139,
      "num_input_tokens_seen": 20080912,
      "step": 36240
    },
    {
      "epoch": 635.8849557522124,
      "grad_norm": 10.5,
      "learning_rate": 1.0799140986539197e-06,
      "loss": 1.187,
      "num_input_tokens_seen": 20083696,
      "step": 36245
    },
    {
      "epoch": 635.9734513274336,
      "grad_norm": 10.8125,
      "learning_rate": 1.0770616531112526e-06,
      "loss": 1.3394,
      "num_input_tokens_seen": 20086640,
      "step": 36250
    },
    {
      "epoch": 636.0530973451328,
      "grad_norm": 7.53125,
      "learning_rate": 1.0742128967863085e-06,
      "loss": 0.9709,
      "num_input_tokens_seen": 20089080,
      "step": 36255
    },
    {
      "epoch": 636.141592920354,
      "grad_norm": 6.84375,
      "learning_rate": 1.071367830118411e-06,
      "loss": 1.0384,
      "num_input_tokens_seen": 20091416,
      "step": 36260
    },
    {
      "epoch": 636.2300884955753,
      "grad_norm": 9.375,
      "learning_rate": 1.068526453546298e-06,
      "loss": 0.9963,
      "num_input_tokens_seen": 20094824,
      "step": 36265
    },
    {
      "epoch": 636.3185840707964,
      "grad_norm": 7.84375,
      "learning_rate": 1.0656887675081467e-06,
      "loss": 1.1078,
      "num_input_tokens_seen": 20097480,
      "step": 36270
    },
    {
      "epoch": 636.4070796460177,
      "grad_norm": 12.625,
      "learning_rate": 1.0628547724415628e-06,
      "loss": 1.4635,
      "num_input_tokens_seen": 20100456,
      "step": 36275
    },
    {
      "epoch": 636.4955752212389,
      "grad_norm": 8.75,
      "learning_rate": 1.0600244687835881e-06,
      "loss": 0.8198,
      "num_input_tokens_seen": 20103576,
      "step": 36280
    },
    {
      "epoch": 636.5840707964602,
      "grad_norm": 7.53125,
      "learning_rate": 1.0571978569706876e-06,
      "loss": 0.92,
      "num_input_tokens_seen": 20106392,
      "step": 36285
    },
    {
      "epoch": 636.6725663716815,
      "grad_norm": 7.96875,
      "learning_rate": 1.0543749374387652e-06,
      "loss": 1.1693,
      "num_input_tokens_seen": 20109048,
      "step": 36290
    },
    {
      "epoch": 636.7610619469026,
      "grad_norm": 8.3125,
      "learning_rate": 1.051555710623142e-06,
      "loss": 0.9804,
      "num_input_tokens_seen": 20111688,
      "step": 36295
    },
    {
      "epoch": 636.8495575221239,
      "grad_norm": 10.75,
      "learning_rate": 1.0487401769585847e-06,
      "loss": 1.171,
      "num_input_tokens_seen": 20114776,
      "step": 36300
    },
    {
      "epoch": 636.9380530973451,
      "grad_norm": 7.25,
      "learning_rate": 1.0459283368792845e-06,
      "loss": 0.8438,
      "num_input_tokens_seen": 20117272,
      "step": 36305
    },
    {
      "epoch": 637.0176991150443,
      "grad_norm": 6.96875,
      "learning_rate": 1.043120190818858e-06,
      "loss": 1.077,
      "num_input_tokens_seen": 20119800,
      "step": 36310
    },
    {
      "epoch": 637.1061946902655,
      "grad_norm": 9.875,
      "learning_rate": 1.0403157392103596e-06,
      "loss": 1.1678,
      "num_input_tokens_seen": 20122088,
      "step": 36315
    },
    {
      "epoch": 637.1946902654868,
      "grad_norm": 9.25,
      "learning_rate": 1.0375149824862735e-06,
      "loss": 0.9725,
      "num_input_tokens_seen": 20125432,
      "step": 36320
    },
    {
      "epoch": 637.2831858407079,
      "grad_norm": 8.25,
      "learning_rate": 1.034717921078507e-06,
      "loss": 1.0497,
      "num_input_tokens_seen": 20128184,
      "step": 36325
    },
    {
      "epoch": 637.3716814159292,
      "grad_norm": 9.0,
      "learning_rate": 1.0319245554184009e-06,
      "loss": 0.9324,
      "num_input_tokens_seen": 20131432,
      "step": 36330
    },
    {
      "epoch": 637.4601769911504,
      "grad_norm": 11.125,
      "learning_rate": 1.0291348859367361e-06,
      "loss": 1.3271,
      "num_input_tokens_seen": 20134152,
      "step": 36335
    },
    {
      "epoch": 637.5486725663717,
      "grad_norm": 6.28125,
      "learning_rate": 1.0263489130637016e-06,
      "loss": 0.958,
      "num_input_tokens_seen": 20137208,
      "step": 36340
    },
    {
      "epoch": 637.637168141593,
      "grad_norm": 7.9375,
      "learning_rate": 1.0235666372289427e-06,
      "loss": 0.8785,
      "num_input_tokens_seen": 20140040,
      "step": 36345
    },
    {
      "epoch": 637.7256637168142,
      "grad_norm": 10.5625,
      "learning_rate": 1.0207880588615076e-06,
      "loss": 1.109,
      "num_input_tokens_seen": 20142808,
      "step": 36350
    },
    {
      "epoch": 637.8141592920354,
      "grad_norm": 7.375,
      "learning_rate": 1.0180131783898984e-06,
      "loss": 1.155,
      "num_input_tokens_seen": 20145368,
      "step": 36355
    },
    {
      "epoch": 637.9026548672566,
      "grad_norm": 12.625,
      "learning_rate": 1.0152419962420362e-06,
      "loss": 1.1806,
      "num_input_tokens_seen": 20148216,
      "step": 36360
    },
    {
      "epoch": 637.9911504424779,
      "grad_norm": 6.03125,
      "learning_rate": 1.0124745128452685e-06,
      "loss": 0.8626,
      "num_input_tokens_seen": 20151064,
      "step": 36365
    },
    {
      "epoch": 638.070796460177,
      "grad_norm": 8.8125,
      "learning_rate": 1.0097107286263758e-06,
      "loss": 1.1495,
      "num_input_tokens_seen": 20153648,
      "step": 36370
    },
    {
      "epoch": 638.1592920353983,
      "grad_norm": 6.5625,
      "learning_rate": 1.00695064401157e-06,
      "loss": 1.0728,
      "num_input_tokens_seen": 20156320,
      "step": 36375
    },
    {
      "epoch": 638.2477876106195,
      "grad_norm": 8.125,
      "learning_rate": 1.0041942594264886e-06,
      "loss": 1.1635,
      "num_input_tokens_seen": 20158640,
      "step": 36380
    },
    {
      "epoch": 638.3362831858407,
      "grad_norm": 8.8125,
      "learning_rate": 1.001441575296208e-06,
      "loss": 1.0841,
      "num_input_tokens_seen": 20161200,
      "step": 36385
    },
    {
      "epoch": 638.4247787610619,
      "grad_norm": 11.1875,
      "learning_rate": 9.986925920452139e-07,
      "loss": 1.0835,
      "num_input_tokens_seen": 20164240,
      "step": 36390
    },
    {
      "epoch": 638.5132743362832,
      "grad_norm": 9.8125,
      "learning_rate": 9.959473100974475e-07,
      "loss": 1.0123,
      "num_input_tokens_seen": 20167056,
      "step": 36395
    },
    {
      "epoch": 638.6017699115044,
      "grad_norm": 9.625,
      "learning_rate": 9.932057298762564e-07,
      "loss": 0.9354,
      "num_input_tokens_seen": 20170032,
      "step": 36400
    },
    {
      "epoch": 638.6017699115044,
      "eval_loss": 1.0310944318771362,
      "eval_runtime": 0.9363,
      "eval_samples_per_second": 26.7,
      "eval_steps_per_second": 13.884,
      "num_input_tokens_seen": 20170032,
      "step": 36400
    },
    {
      "epoch": 638.6902654867257,
      "grad_norm": 9.4375,
      "learning_rate": 9.90467851804433e-07,
      "loss": 1.2335,
      "num_input_tokens_seen": 20173040,
      "step": 36405
    },
    {
      "epoch": 638.7787610619469,
      "grad_norm": 10.9375,
      "learning_rate": 9.877336763041895e-07,
      "loss": 1.0073,
      "num_input_tokens_seen": 20176000,
      "step": 36410
    },
    {
      "epoch": 638.8672566371681,
      "grad_norm": 11.1875,
      "learning_rate": 9.850032037971662e-07,
      "loss": 0.9467,
      "num_input_tokens_seen": 20178704,
      "step": 36415
    },
    {
      "epoch": 638.9557522123894,
      "grad_norm": 8.0,
      "learning_rate": 9.822764347044406e-07,
      "loss": 1.0768,
      "num_input_tokens_seen": 20181040,
      "step": 36420
    },
    {
      "epoch": 639.0353982300885,
      "grad_norm": 9.25,
      "learning_rate": 9.795533694465175e-07,
      "loss": 0.8887,
      "num_input_tokens_seen": 20183320,
      "step": 36425
    },
    {
      "epoch": 639.1238938053098,
      "grad_norm": 8.5625,
      "learning_rate": 9.768340084433197e-07,
      "loss": 1.1941,
      "num_input_tokens_seen": 20185608,
      "step": 36430
    },
    {
      "epoch": 639.212389380531,
      "grad_norm": 8.875,
      "learning_rate": 9.741183521142143e-07,
      "loss": 1.1887,
      "num_input_tokens_seen": 20188072,
      "step": 36435
    },
    {
      "epoch": 639.3008849557522,
      "grad_norm": 6.84375,
      "learning_rate": 9.714064008779889e-07,
      "loss": 0.904,
      "num_input_tokens_seen": 20190664,
      "step": 36440
    },
    {
      "epoch": 639.3893805309734,
      "grad_norm": 7.4375,
      "learning_rate": 9.686981551528584e-07,
      "loss": 0.9544,
      "num_input_tokens_seen": 20193976,
      "step": 36445
    },
    {
      "epoch": 639.4778761061947,
      "grad_norm": 10.875,
      "learning_rate": 9.65993615356467e-07,
      "loss": 1.0302,
      "num_input_tokens_seen": 20196808,
      "step": 36450
    },
    {
      "epoch": 639.566371681416,
      "grad_norm": 7.625,
      "learning_rate": 9.632927819058917e-07,
      "loss": 0.898,
      "num_input_tokens_seen": 20199992,
      "step": 36455
    },
    {
      "epoch": 639.6548672566372,
      "grad_norm": 8.25,
      "learning_rate": 9.605956552176305e-07,
      "loss": 1.038,
      "num_input_tokens_seen": 20202536,
      "step": 36460
    },
    {
      "epoch": 639.7433628318585,
      "grad_norm": 8.6875,
      "learning_rate": 9.579022357076223e-07,
      "loss": 1.2203,
      "num_input_tokens_seen": 20205432,
      "step": 36465
    },
    {
      "epoch": 639.8318584070796,
      "grad_norm": 7.9375,
      "learning_rate": 9.552125237912158e-07,
      "loss": 0.9116,
      "num_input_tokens_seen": 20208296,
      "step": 36470
    },
    {
      "epoch": 639.9203539823009,
      "grad_norm": 7.8125,
      "learning_rate": 9.525265198832096e-07,
      "loss": 0.9827,
      "num_input_tokens_seen": 20211672,
      "step": 36475
    },
    {
      "epoch": 640.0,
      "grad_norm": 24.875,
      "learning_rate": 9.498442243978112e-07,
      "loss": 1.4163,
      "num_input_tokens_seen": 20213824,
      "step": 36480
    },
    {
      "epoch": 640.0884955752213,
      "grad_norm": 9.25,
      "learning_rate": 9.471656377486649e-07,
      "loss": 1.1092,
      "num_input_tokens_seen": 20216352,
      "step": 36485
    },
    {
      "epoch": 640.1769911504425,
      "grad_norm": 7.59375,
      "learning_rate": 9.444907603488456e-07,
      "loss": 1.0385,
      "num_input_tokens_seen": 20219344,
      "step": 36490
    },
    {
      "epoch": 640.2654867256637,
      "grad_norm": 11.125,
      "learning_rate": 9.418195926108514e-07,
      "loss": 1.2227,
      "num_input_tokens_seen": 20222272,
      "step": 36495
    },
    {
      "epoch": 640.3539823008849,
      "grad_norm": 8.6875,
      "learning_rate": 9.391521349466053e-07,
      "loss": 0.9807,
      "num_input_tokens_seen": 20225024,
      "step": 36500
    },
    {
      "epoch": 640.4424778761062,
      "grad_norm": 8.3125,
      "learning_rate": 9.364883877674758e-07,
      "loss": 1.1797,
      "num_input_tokens_seen": 20227504,
      "step": 36505
    },
    {
      "epoch": 640.5309734513274,
      "grad_norm": 9.375,
      "learning_rate": 9.33828351484231e-07,
      "loss": 1.1205,
      "num_input_tokens_seen": 20230144,
      "step": 36510
    },
    {
      "epoch": 640.6194690265487,
      "grad_norm": 7.9375,
      "learning_rate": 9.311720265070906e-07,
      "loss": 1.012,
      "num_input_tokens_seen": 20233024,
      "step": 36515
    },
    {
      "epoch": 640.70796460177,
      "grad_norm": 8.0625,
      "learning_rate": 9.285194132456931e-07,
      "loss": 1.0557,
      "num_input_tokens_seen": 20236032,
      "step": 36520
    },
    {
      "epoch": 640.7964601769911,
      "grad_norm": 6.875,
      "learning_rate": 9.258705121091032e-07,
      "loss": 1.0064,
      "num_input_tokens_seen": 20239504,
      "step": 36525
    },
    {
      "epoch": 640.8849557522124,
      "grad_norm": 8.875,
      "learning_rate": 9.232253235058136e-07,
      "loss": 0.883,
      "num_input_tokens_seen": 20242272,
      "step": 36530
    },
    {
      "epoch": 640.9734513274336,
      "grad_norm": 9.25,
      "learning_rate": 9.205838478437478e-07,
      "loss": 1.0006,
      "num_input_tokens_seen": 20244720,
      "step": 36535
    },
    {
      "epoch": 641.0530973451328,
      "grad_norm": 10.375,
      "learning_rate": 9.179460855302524e-07,
      "loss": 1.0653,
      "num_input_tokens_seen": 20246928,
      "step": 36540
    },
    {
      "epoch": 641.141592920354,
      "grad_norm": 8.1875,
      "learning_rate": 9.153120369721046e-07,
      "loss": 1.0109,
      "num_input_tokens_seen": 20249536,
      "step": 36545
    },
    {
      "epoch": 641.2300884955753,
      "grad_norm": 8.875,
      "learning_rate": 9.126817025755103e-07,
      "loss": 1.0848,
      "num_input_tokens_seen": 20251968,
      "step": 36550
    },
    {
      "epoch": 641.3185840707964,
      "grad_norm": 7.90625,
      "learning_rate": 9.100550827460947e-07,
      "loss": 1.1351,
      "num_input_tokens_seen": 20255312,
      "step": 36555
    },
    {
      "epoch": 641.4070796460177,
      "grad_norm": 5.84375,
      "learning_rate": 9.0743217788892e-07,
      "loss": 1.2027,
      "num_input_tokens_seen": 20257680,
      "step": 36560
    },
    {
      "epoch": 641.4955752212389,
      "grad_norm": 7.9375,
      "learning_rate": 9.048129884084683e-07,
      "loss": 1.1227,
      "num_input_tokens_seen": 20260384,
      "step": 36565
    },
    {
      "epoch": 641.5840707964602,
      "grad_norm": 6.71875,
      "learning_rate": 9.021975147086553e-07,
      "loss": 0.9248,
      "num_input_tokens_seen": 20262928,
      "step": 36570
    },
    {
      "epoch": 641.6725663716815,
      "grad_norm": 6.78125,
      "learning_rate": 8.995857571928141e-07,
      "loss": 0.8675,
      "num_input_tokens_seen": 20265136,
      "step": 36575
    },
    {
      "epoch": 641.7610619469026,
      "grad_norm": 11.125,
      "learning_rate": 8.969777162637139e-07,
      "loss": 0.9893,
      "num_input_tokens_seen": 20268816,
      "step": 36580
    },
    {
      "epoch": 641.8495575221239,
      "grad_norm": 9.5,
      "learning_rate": 8.943733923235525e-07,
      "loss": 0.9048,
      "num_input_tokens_seen": 20271968,
      "step": 36585
    },
    {
      "epoch": 641.9380530973451,
      "grad_norm": 8.0625,
      "learning_rate": 8.917727857739394e-07,
      "loss": 1.1054,
      "num_input_tokens_seen": 20274656,
      "step": 36590
    },
    {
      "epoch": 642.0176991150443,
      "grad_norm": 8.375,
      "learning_rate": 8.891758970159258e-07,
      "loss": 1.2415,
      "num_input_tokens_seen": 20277208,
      "step": 36595
    },
    {
      "epoch": 642.1061946902655,
      "grad_norm": 8.0625,
      "learning_rate": 8.86582726449986e-07,
      "loss": 1.1421,
      "num_input_tokens_seen": 20279560,
      "step": 36600
    },
    {
      "epoch": 642.1061946902655,
      "eval_loss": 1.0310944318771362,
      "eval_runtime": 0.9481,
      "eval_samples_per_second": 26.37,
      "eval_steps_per_second": 13.712,
      "num_input_tokens_seen": 20279560,
      "step": 36600
    },
    {
      "epoch": 642.1946902654868,
      "grad_norm": 8.25,
      "learning_rate": 8.839932744760165e-07,
      "loss": 1.0265,
      "num_input_tokens_seen": 20282360,
      "step": 36605
    },
    {
      "epoch": 642.2831858407079,
      "grad_norm": 7.34375,
      "learning_rate": 8.814075414933482e-07,
      "loss": 0.9491,
      "num_input_tokens_seen": 20285576,
      "step": 36610
    },
    {
      "epoch": 642.3716814159292,
      "grad_norm": 7.34375,
      "learning_rate": 8.788255279007257e-07,
      "loss": 1.0532,
      "num_input_tokens_seen": 20288504,
      "step": 36615
    },
    {
      "epoch": 642.4601769911504,
      "grad_norm": 9.9375,
      "learning_rate": 8.762472340963362e-07,
      "loss": 1.1371,
      "num_input_tokens_seen": 20291112,
      "step": 36620
    },
    {
      "epoch": 642.5486725663717,
      "grad_norm": 8.5,
      "learning_rate": 8.736726604777811e-07,
      "loss": 1.0146,
      "num_input_tokens_seen": 20294280,
      "step": 36625
    },
    {
      "epoch": 642.637168141593,
      "grad_norm": 7.65625,
      "learning_rate": 8.711018074420901e-07,
      "loss": 1.0293,
      "num_input_tokens_seen": 20297352,
      "step": 36630
    },
    {
      "epoch": 642.7256637168142,
      "grad_norm": 7.8125,
      "learning_rate": 8.685346753857209e-07,
      "loss": 1.0163,
      "num_input_tokens_seen": 20299880,
      "step": 36635
    },
    {
      "epoch": 642.8141592920354,
      "grad_norm": 6.90625,
      "learning_rate": 8.659712647045654e-07,
      "loss": 1.1543,
      "num_input_tokens_seen": 20302520,
      "step": 36640
    },
    {
      "epoch": 642.9026548672566,
      "grad_norm": 9.8125,
      "learning_rate": 8.634115757939209e-07,
      "loss": 0.8964,
      "num_input_tokens_seen": 20305064,
      "step": 36645
    },
    {
      "epoch": 642.9911504424779,
      "grad_norm": 8.25,
      "learning_rate": 8.608556090485387e-07,
      "loss": 1.1506,
      "num_input_tokens_seen": 20307512,
      "step": 36650
    },
    {
      "epoch": 643.070796460177,
      "grad_norm": 7.59375,
      "learning_rate": 8.583033648625671e-07,
      "loss": 0.8421,
      "num_input_tokens_seen": 20309920,
      "step": 36655
    },
    {
      "epoch": 643.1592920353983,
      "grad_norm": 7.6875,
      "learning_rate": 8.557548436295998e-07,
      "loss": 1.0076,
      "num_input_tokens_seen": 20312784,
      "step": 36660
    },
    {
      "epoch": 643.2477876106195,
      "grad_norm": 9.3125,
      "learning_rate": 8.532100457426556e-07,
      "loss": 1.0905,
      "num_input_tokens_seen": 20315472,
      "step": 36665
    },
    {
      "epoch": 643.3362831858407,
      "grad_norm": 8.25,
      "learning_rate": 8.506689715941679e-07,
      "loss": 0.9887,
      "num_input_tokens_seen": 20318400,
      "step": 36670
    },
    {
      "epoch": 643.4247787610619,
      "grad_norm": 8.0625,
      "learning_rate": 8.481316215760011e-07,
      "loss": 1.1506,
      "num_input_tokens_seen": 20321056,
      "step": 36675
    },
    {
      "epoch": 643.5132743362832,
      "grad_norm": 11.4375,
      "learning_rate": 8.455979960794558e-07,
      "loss": 0.981,
      "num_input_tokens_seen": 20323952,
      "step": 36680
    },
    {
      "epoch": 643.6017699115044,
      "grad_norm": 7.78125,
      "learning_rate": 8.430680954952364e-07,
      "loss": 1.0436,
      "num_input_tokens_seen": 20327056,
      "step": 36685
    },
    {
      "epoch": 643.6902654867257,
      "grad_norm": 7.46875,
      "learning_rate": 8.405419202134974e-07,
      "loss": 1.032,
      "num_input_tokens_seen": 20329936,
      "step": 36690
    },
    {
      "epoch": 643.7787610619469,
      "grad_norm": 8.0,
      "learning_rate": 8.380194706237993e-07,
      "loss": 1.1033,
      "num_input_tokens_seen": 20332592,
      "step": 36695
    },
    {
      "epoch": 643.8672566371681,
      "grad_norm": 6.0625,
      "learning_rate": 8.355007471151366e-07,
      "loss": 1.1241,
      "num_input_tokens_seen": 20335184,
      "step": 36700
    },
    {
      "epoch": 643.9557522123894,
      "grad_norm": 8.6875,
      "learning_rate": 8.329857500759292e-07,
      "loss": 1.0509,
      "num_input_tokens_seen": 20338320,
      "step": 36705
    },
    {
      "epoch": 644.0353982300885,
      "grad_norm": 7.21875,
      "learning_rate": 8.304744798940194e-07,
      "loss": 1.1621,
      "num_input_tokens_seen": 20340312,
      "step": 36710
    },
    {
      "epoch": 644.1238938053098,
      "grad_norm": 6.96875,
      "learning_rate": 8.279669369566756e-07,
      "loss": 0.9297,
      "num_input_tokens_seen": 20342920,
      "step": 36715
    },
    {
      "epoch": 644.212389380531,
      "grad_norm": 8.5625,
      "learning_rate": 8.254631216505993e-07,
      "loss": 1.1457,
      "num_input_tokens_seen": 20345672,
      "step": 36720
    },
    {
      "epoch": 644.3008849557522,
      "grad_norm": 6.9375,
      "learning_rate": 8.229630343619038e-07,
      "loss": 1.0404,
      "num_input_tokens_seen": 20348808,
      "step": 36725
    },
    {
      "epoch": 644.3893805309734,
      "grad_norm": 8.8125,
      "learning_rate": 8.204666754761392e-07,
      "loss": 1.1476,
      "num_input_tokens_seen": 20351752,
      "step": 36730
    },
    {
      "epoch": 644.4778761061947,
      "grad_norm": 10.375,
      "learning_rate": 8.179740453782669e-07,
      "loss": 1.2628,
      "num_input_tokens_seen": 20354872,
      "step": 36735
    },
    {
      "epoch": 644.566371681416,
      "grad_norm": 7.375,
      "learning_rate": 8.154851444526907e-07,
      "loss": 1.1374,
      "num_input_tokens_seen": 20357880,
      "step": 36740
    },
    {
      "epoch": 644.6548672566372,
      "grad_norm": 10.625,
      "learning_rate": 8.129999730832283e-07,
      "loss": 0.9243,
      "num_input_tokens_seen": 20360312,
      "step": 36745
    },
    {
      "epoch": 644.7433628318585,
      "grad_norm": 7.625,
      "learning_rate": 8.105185316531178e-07,
      "loss": 1.0288,
      "num_input_tokens_seen": 20362552,
      "step": 36750
    },
    {
      "epoch": 644.8318584070796,
      "grad_norm": 8.75,
      "learning_rate": 8.08040820545039e-07,
      "loss": 0.8703,
      "num_input_tokens_seen": 20365496,
      "step": 36755
    },
    {
      "epoch": 644.9203539823009,
      "grad_norm": 8.0,
      "learning_rate": 8.055668401410782e-07,
      "loss": 1.0701,
      "num_input_tokens_seen": 20368472,
      "step": 36760
    },
    {
      "epoch": 645.0,
      "grad_norm": 8.6875,
      "learning_rate": 8.030965908227578e-07,
      "loss": 0.9212,
      "num_input_tokens_seen": 20370688,
      "step": 36765
    },
    {
      "epoch": 645.0884955752213,
      "grad_norm": 9.875,
      "learning_rate": 8.006300729710203e-07,
      "loss": 1.0469,
      "num_input_tokens_seen": 20373440,
      "step": 36770
    },
    {
      "epoch": 645.1769911504425,
      "grad_norm": 7.65625,
      "learning_rate": 7.981672869662337e-07,
      "loss": 0.9918,
      "num_input_tokens_seen": 20376448,
      "step": 36775
    },
    {
      "epoch": 645.2654867256637,
      "grad_norm": 8.0,
      "learning_rate": 7.957082331881888e-07,
      "loss": 0.9192,
      "num_input_tokens_seen": 20378768,
      "step": 36780
    },
    {
      "epoch": 645.3539823008849,
      "grad_norm": 9.625,
      "learning_rate": 7.932529120161069e-07,
      "loss": 1.0614,
      "num_input_tokens_seen": 20381664,
      "step": 36785
    },
    {
      "epoch": 645.4424778761062,
      "grad_norm": 8.9375,
      "learning_rate": 7.908013238286243e-07,
      "loss": 1.1346,
      "num_input_tokens_seen": 20384432,
      "step": 36790
    },
    {
      "epoch": 645.5309734513274,
      "grad_norm": 12.0625,
      "learning_rate": 7.883534690038136e-07,
      "loss": 0.9899,
      "num_input_tokens_seen": 20387200,
      "step": 36795
    },
    {
      "epoch": 645.6194690265487,
      "grad_norm": 7.5,
      "learning_rate": 7.859093479191559e-07,
      "loss": 0.9622,
      "num_input_tokens_seen": 20389936,
      "step": 36800
    },
    {
      "epoch": 645.6194690265487,
      "eval_loss": 1.0310944318771362,
      "eval_runtime": 0.9466,
      "eval_samples_per_second": 26.41,
      "eval_steps_per_second": 13.733,
      "num_input_tokens_seen": 20389936,
      "step": 36800
    },
    {
      "epoch": 645.70796460177,
      "grad_norm": 7.5625,
      "learning_rate": 7.834689609515722e-07,
      "loss": 1.1753,
      "num_input_tokens_seen": 20392656,
      "step": 36805
    },
    {
      "epoch": 645.7964601769911,
      "grad_norm": 6.59375,
      "learning_rate": 7.810323084774002e-07,
      "loss": 1.1417,
      "num_input_tokens_seen": 20395504,
      "step": 36810
    },
    {
      "epoch": 645.8849557522124,
      "grad_norm": 9.125,
      "learning_rate": 7.785993908723976e-07,
      "loss": 1.1167,
      "num_input_tokens_seen": 20398256,
      "step": 36815
    },
    {
      "epoch": 645.9734513274336,
      "grad_norm": 7.9375,
      "learning_rate": 7.761702085117534e-07,
      "loss": 0.9427,
      "num_input_tokens_seen": 20401568,
      "step": 36820
    },
    {
      "epoch": 646.0530973451328,
      "grad_norm": 9.8125,
      "learning_rate": 7.737447617700844e-07,
      "loss": 1.1186,
      "num_input_tokens_seen": 20404240,
      "step": 36825
    },
    {
      "epoch": 646.141592920354,
      "grad_norm": 11.3125,
      "learning_rate": 7.713230510214136e-07,
      "loss": 1.0125,
      "num_input_tokens_seen": 20407072,
      "step": 36830
    },
    {
      "epoch": 646.2300884955753,
      "grad_norm": 12.5,
      "learning_rate": 7.689050766392092e-07,
      "loss": 1.1016,
      "num_input_tokens_seen": 20409984,
      "step": 36835
    },
    {
      "epoch": 646.3185840707964,
      "grad_norm": 6.8125,
      "learning_rate": 7.664908389963477e-07,
      "loss": 1.0695,
      "num_input_tokens_seen": 20413040,
      "step": 36840
    },
    {
      "epoch": 646.4070796460177,
      "grad_norm": 11.6875,
      "learning_rate": 7.64080338465134e-07,
      "loss": 1.1242,
      "num_input_tokens_seen": 20415456,
      "step": 36845
    },
    {
      "epoch": 646.4955752212389,
      "grad_norm": 8.0,
      "learning_rate": 7.616735754173043e-07,
      "loss": 1.1715,
      "num_input_tokens_seen": 20418048,
      "step": 36850
    },
    {
      "epoch": 646.5840707964602,
      "grad_norm": 5.4375,
      "learning_rate": 7.592705502240005e-07,
      "loss": 0.8442,
      "num_input_tokens_seen": 20420752,
      "step": 36855
    },
    {
      "epoch": 646.6725663716815,
      "grad_norm": 6.59375,
      "learning_rate": 7.568712632558095e-07,
      "loss": 1.0454,
      "num_input_tokens_seen": 20423680,
      "step": 36860
    },
    {
      "epoch": 646.7610619469026,
      "grad_norm": 9.0625,
      "learning_rate": 7.544757148827297e-07,
      "loss": 1.0048,
      "num_input_tokens_seen": 20426576,
      "step": 36865
    },
    {
      "epoch": 646.8495575221239,
      "grad_norm": 7.59375,
      "learning_rate": 7.520839054741797e-07,
      "loss": 0.9337,
      "num_input_tokens_seen": 20429408,
      "step": 36870
    },
    {
      "epoch": 646.9380530973451,
      "grad_norm": 8.4375,
      "learning_rate": 7.496958353990113e-07,
      "loss": 1.134,
      "num_input_tokens_seen": 20432352,
      "step": 36875
    },
    {
      "epoch": 647.0176991150443,
      "grad_norm": 8.0,
      "learning_rate": 7.473115050254941e-07,
      "loss": 0.9722,
      "num_input_tokens_seen": 20434360,
      "step": 36880
    },
    {
      "epoch": 647.1061946902655,
      "grad_norm": 7.90625,
      "learning_rate": 7.449309147213173e-07,
      "loss": 0.9464,
      "num_input_tokens_seen": 20437096,
      "step": 36885
    },
    {
      "epoch": 647.1946902654868,
      "grad_norm": 8.5,
      "learning_rate": 7.425540648536067e-07,
      "loss": 1.0117,
      "num_input_tokens_seen": 20439704,
      "step": 36890
    },
    {
      "epoch": 647.2831858407079,
      "grad_norm": 8.3125,
      "learning_rate": 7.40180955788894e-07,
      "loss": 1.1082,
      "num_input_tokens_seen": 20442120,
      "step": 36895
    },
    {
      "epoch": 647.3716814159292,
      "grad_norm": 7.53125,
      "learning_rate": 7.378115878931474e-07,
      "loss": 1.0422,
      "num_input_tokens_seen": 20444664,
      "step": 36900
    },
    {
      "epoch": 647.4601769911504,
      "grad_norm": 7.9375,
      "learning_rate": 7.354459615317527e-07,
      "loss": 1.1753,
      "num_input_tokens_seen": 20447528,
      "step": 36905
    },
    {
      "epoch": 647.5486725663717,
      "grad_norm": 8.375,
      "learning_rate": 7.33084077069518e-07,
      "loss": 1.0853,
      "num_input_tokens_seen": 20450824,
      "step": 36910
    },
    {
      "epoch": 647.637168141593,
      "grad_norm": 8.5625,
      "learning_rate": 7.307259348706768e-07,
      "loss": 1.1383,
      "num_input_tokens_seen": 20453672,
      "step": 36915
    },
    {
      "epoch": 647.7256637168142,
      "grad_norm": 9.9375,
      "learning_rate": 7.283715352988801e-07,
      "loss": 1.1404,
      "num_input_tokens_seen": 20456376,
      "step": 36920
    },
    {
      "epoch": 647.8141592920354,
      "grad_norm": 9.0625,
      "learning_rate": 7.260208787172068e-07,
      "loss": 1.0396,
      "num_input_tokens_seen": 20459464,
      "step": 36925
    },
    {
      "epoch": 647.9026548672566,
      "grad_norm": 6.46875,
      "learning_rate": 7.23673965488167e-07,
      "loss": 0.9252,
      "num_input_tokens_seen": 20462120,
      "step": 36930
    },
    {
      "epoch": 647.9911504424779,
      "grad_norm": 9.625,
      "learning_rate": 7.213307959736709e-07,
      "loss": 0.9604,
      "num_input_tokens_seen": 20465096,
      "step": 36935
    },
    {
      "epoch": 648.070796460177,
      "grad_norm": 6.96875,
      "learning_rate": 7.189913705350715e-07,
      "loss": 1.0086,
      "num_input_tokens_seen": 20467656,
      "step": 36940
    },
    {
      "epoch": 648.1592920353983,
      "grad_norm": 6.5625,
      "learning_rate": 7.166556895331411e-07,
      "loss": 0.9215,
      "num_input_tokens_seen": 20470136,
      "step": 36945
    },
    {
      "epoch": 648.2477876106195,
      "grad_norm": 8.625,
      "learning_rate": 7.143237533280639e-07,
      "loss": 1.1113,
      "num_input_tokens_seen": 20472840,
      "step": 36950
    },
    {
      "epoch": 648.3362831858407,
      "grad_norm": 8.125,
      "learning_rate": 7.119955622794578e-07,
      "loss": 1.3372,
      "num_input_tokens_seen": 20475496,
      "step": 36955
    },
    {
      "epoch": 648.4247787610619,
      "grad_norm": 6.28125,
      "learning_rate": 7.096711167463577e-07,
      "loss": 0.8418,
      "num_input_tokens_seen": 20478072,
      "step": 36960
    },
    {
      "epoch": 648.5132743362832,
      "grad_norm": 10.8125,
      "learning_rate": 7.073504170872213e-07,
      "loss": 1.1373,
      "num_input_tokens_seen": 20480888,
      "step": 36965
    },
    {
      "epoch": 648.6017699115044,
      "grad_norm": 6.0,
      "learning_rate": 7.05033463659932e-07,
      "loss": 0.8965,
      "num_input_tokens_seen": 20483800,
      "step": 36970
    },
    {
      "epoch": 648.6902654867257,
      "grad_norm": 9.0,
      "learning_rate": 7.027202568217928e-07,
      "loss": 1.0943,
      "num_input_tokens_seen": 20486712,
      "step": 36975
    },
    {
      "epoch": 648.7787610619469,
      "grad_norm": 14.8125,
      "learning_rate": 7.004107969295293e-07,
      "loss": 1.2279,
      "num_input_tokens_seen": 20489592,
      "step": 36980
    },
    {
      "epoch": 648.8672566371681,
      "grad_norm": 9.9375,
      "learning_rate": 6.9810508433929e-07,
      "loss": 0.9473,
      "num_input_tokens_seen": 20492920,
      "step": 36985
    },
    {
      "epoch": 648.9557522123894,
      "grad_norm": 8.3125,
      "learning_rate": 6.958031194066406e-07,
      "loss": 1.0198,
      "num_input_tokens_seen": 20495464,
      "step": 36990
    },
    {
      "epoch": 649.0353982300885,
      "grad_norm": 9.125,
      "learning_rate": 6.935049024865776e-07,
      "loss": 1.3284,
      "num_input_tokens_seen": 20497584,
      "step": 36995
    },
    {
      "epoch": 649.1238938053098,
      "grad_norm": 8.3125,
      "learning_rate": 6.912104339335118e-07,
      "loss": 1.0777,
      "num_input_tokens_seen": 20499984,
      "step": 37000
    },
    {
      "epoch": 649.1238938053098,
      "eval_loss": 1.0310944318771362,
      "eval_runtime": 0.9467,
      "eval_samples_per_second": 26.406,
      "eval_steps_per_second": 13.731,
      "num_input_tokens_seen": 20499984,
      "step": 37000
    },
    {
      "epoch": 649.212389380531,
      "grad_norm": 9.875,
      "learning_rate": 6.889197141012799e-07,
      "loss": 1.1808,
      "num_input_tokens_seen": 20502688,
      "step": 37005
    },
    {
      "epoch": 649.3008849557522,
      "grad_norm": 8.5625,
      "learning_rate": 6.866327433431435e-07,
      "loss": 1.2527,
      "num_input_tokens_seen": 20505392,
      "step": 37010
    },
    {
      "epoch": 649.3893805309734,
      "grad_norm": 8.1875,
      "learning_rate": 6.843495220117735e-07,
      "loss": 0.9279,
      "num_input_tokens_seen": 20508240,
      "step": 37015
    },
    {
      "epoch": 649.4778761061947,
      "grad_norm": 7.1875,
      "learning_rate": 6.820700504592798e-07,
      "loss": 0.9839,
      "num_input_tokens_seen": 20511600,
      "step": 37020
    },
    {
      "epoch": 649.566371681416,
      "grad_norm": 5.90625,
      "learning_rate": 6.797943290371839e-07,
      "loss": 0.9246,
      "num_input_tokens_seen": 20514720,
      "step": 37025
    },
    {
      "epoch": 649.6548672566372,
      "grad_norm": 9.0625,
      "learning_rate": 6.775223580964274e-07,
      "loss": 1.0521,
      "num_input_tokens_seen": 20516976,
      "step": 37030
    },
    {
      "epoch": 649.7433628318585,
      "grad_norm": 10.6875,
      "learning_rate": 6.7525413798738e-07,
      "loss": 1.156,
      "num_input_tokens_seen": 20519936,
      "step": 37035
    },
    {
      "epoch": 649.8318584070796,
      "grad_norm": 9.8125,
      "learning_rate": 6.729896690598259e-07,
      "loss": 0.9598,
      "num_input_tokens_seen": 20522688,
      "step": 37040
    },
    {
      "epoch": 649.9203539823009,
      "grad_norm": 7.59375,
      "learning_rate": 6.707289516629772e-07,
      "loss": 1.0869,
      "num_input_tokens_seen": 20525440,
      "step": 37045
    },
    {
      "epoch": 650.0,
      "grad_norm": 8.4375,
      "learning_rate": 6.684719861454692e-07,
      "loss": 0.9266,
      "num_input_tokens_seen": 20528032,
      "step": 37050
    },
    {
      "epoch": 650.0884955752213,
      "grad_norm": 8.5625,
      "learning_rate": 6.662187728553481e-07,
      "loss": 0.8914,
      "num_input_tokens_seen": 20531056,
      "step": 37055
    },
    {
      "epoch": 650.1769911504425,
      "grad_norm": 7.5,
      "learning_rate": 6.639693121400892e-07,
      "loss": 1.1003,
      "num_input_tokens_seen": 20533616,
      "step": 37060
    },
    {
      "epoch": 650.2654867256637,
      "grad_norm": 8.375,
      "learning_rate": 6.617236043465868e-07,
      "loss": 1.1757,
      "num_input_tokens_seen": 20536528,
      "step": 37065
    },
    {
      "epoch": 650.3539823008849,
      "grad_norm": 7.5625,
      "learning_rate": 6.594816498211587e-07,
      "loss": 1.025,
      "num_input_tokens_seen": 20538928,
      "step": 37070
    },
    {
      "epoch": 650.4424778761062,
      "grad_norm": 8.0625,
      "learning_rate": 6.572434489095447e-07,
      "loss": 0.8778,
      "num_input_tokens_seen": 20542064,
      "step": 37075
    },
    {
      "epoch": 650.5309734513274,
      "grad_norm": 7.21875,
      "learning_rate": 6.550090019568994e-07,
      "loss": 1.1118,
      "num_input_tokens_seen": 20544768,
      "step": 37080
    },
    {
      "epoch": 650.6194690265487,
      "grad_norm": 7.59375,
      "learning_rate": 6.527783093078027e-07,
      "loss": 1.1666,
      "num_input_tokens_seen": 20548240,
      "step": 37085
    },
    {
      "epoch": 650.70796460177,
      "grad_norm": 7.09375,
      "learning_rate": 6.5055137130626e-07,
      "loss": 0.9362,
      "num_input_tokens_seen": 20551120,
      "step": 37090
    },
    {
      "epoch": 650.7964601769911,
      "grad_norm": 6.28125,
      "learning_rate": 6.483281882956854e-07,
      "loss": 0.9073,
      "num_input_tokens_seen": 20553616,
      "step": 37095
    },
    {
      "epoch": 650.8849557522124,
      "grad_norm": 6.6875,
      "learning_rate": 6.461087606189298e-07,
      "loss": 1.1288,
      "num_input_tokens_seen": 20556208,
      "step": 37100
    },
    {
      "epoch": 650.9734513274336,
      "grad_norm": 7.09375,
      "learning_rate": 6.438930886182554e-07,
      "loss": 1.1288,
      "num_input_tokens_seen": 20558912,
      "step": 37105
    },
    {
      "epoch": 651.0530973451328,
      "grad_norm": 9.4375,
      "learning_rate": 6.416811726353417e-07,
      "loss": 1.3391,
      "num_input_tokens_seen": 20561208,
      "step": 37110
    },
    {
      "epoch": 651.141592920354,
      "grad_norm": 8.9375,
      "learning_rate": 6.394730130112991e-07,
      "loss": 1.0115,
      "num_input_tokens_seen": 20564152,
      "step": 37115
    },
    {
      "epoch": 651.2300884955753,
      "grad_norm": 8.25,
      "learning_rate": 6.372686100866471e-07,
      "loss": 1.0365,
      "num_input_tokens_seen": 20566952,
      "step": 37120
    },
    {
      "epoch": 651.3185840707964,
      "grad_norm": 7.96875,
      "learning_rate": 6.350679642013413e-07,
      "loss": 0.9744,
      "num_input_tokens_seen": 20570168,
      "step": 37125
    },
    {
      "epoch": 651.4070796460177,
      "grad_norm": 7.8125,
      "learning_rate": 6.328710756947437e-07,
      "loss": 0.9318,
      "num_input_tokens_seen": 20572984,
      "step": 37130
    },
    {
      "epoch": 651.4955752212389,
      "grad_norm": 9.0,
      "learning_rate": 6.306779449056416e-07,
      "loss": 1.26,
      "num_input_tokens_seen": 20575208,
      "step": 37135
    },
    {
      "epoch": 651.5840707964602,
      "grad_norm": 8.4375,
      "learning_rate": 6.284885721722422e-07,
      "loss": 0.8988,
      "num_input_tokens_seen": 20578488,
      "step": 37140
    },
    {
      "epoch": 651.6725663716815,
      "grad_norm": 7.34375,
      "learning_rate": 6.26302957832181e-07,
      "loss": 1.0988,
      "num_input_tokens_seen": 20581192,
      "step": 37145
    },
    {
      "epoch": 651.7610619469026,
      "grad_norm": 8.375,
      "learning_rate": 6.241211022224997e-07,
      "loss": 1.1629,
      "num_input_tokens_seen": 20584008,
      "step": 37150
    },
    {
      "epoch": 651.8495575221239,
      "grad_norm": 7.09375,
      "learning_rate": 6.219430056796732e-07,
      "loss": 0.9139,
      "num_input_tokens_seen": 20587016,
      "step": 37155
    },
    {
      "epoch": 651.9380530973451,
      "grad_norm": 8.9375,
      "learning_rate": 6.19768668539586e-07,
      "loss": 0.9276,
      "num_input_tokens_seen": 20589944,
      "step": 37160
    },
    {
      "epoch": 652.0176991150443,
      "grad_norm": 6.8125,
      "learning_rate": 6.175980911375528e-07,
      "loss": 1.0361,
      "num_input_tokens_seen": 20592208,
      "step": 37165
    },
    {
      "epoch": 652.1061946902655,
      "grad_norm": 9.0,
      "learning_rate": 6.154312738083034e-07,
      "loss": 1.0263,
      "num_input_tokens_seen": 20595088,
      "step": 37170
    },
    {
      "epoch": 652.1946902654868,
      "grad_norm": 7.125,
      "learning_rate": 6.132682168859843e-07,
      "loss": 0.9792,
      "num_input_tokens_seen": 20598064,
      "step": 37175
    },
    {
      "epoch": 652.2831858407079,
      "grad_norm": 9.125,
      "learning_rate": 6.111089207041704e-07,
      "loss": 1.1166,
      "num_input_tokens_seen": 20600928,
      "step": 37180
    },
    {
      "epoch": 652.3716814159292,
      "grad_norm": 8.6875,
      "learning_rate": 6.089533855958507e-07,
      "loss": 1.1511,
      "num_input_tokens_seen": 20603920,
      "step": 37185
    },
    {
      "epoch": 652.4601769911504,
      "grad_norm": 9.1875,
      "learning_rate": 6.068016118934372e-07,
      "loss": 1.178,
      "num_input_tokens_seen": 20606560,
      "step": 37190
    },
    {
      "epoch": 652.5486725663717,
      "grad_norm": 7.03125,
      "learning_rate": 6.04653599928759e-07,
      "loss": 0.9837,
      "num_input_tokens_seen": 20609488,
      "step": 37195
    },
    {
      "epoch": 652.637168141593,
      "grad_norm": 6.15625,
      "learning_rate": 6.025093500330675e-07,
      "loss": 0.9475,
      "num_input_tokens_seen": 20612176,
      "step": 37200
    },
    {
      "epoch": 652.637168141593,
      "eval_loss": 1.0310944318771362,
      "eval_runtime": 0.936,
      "eval_samples_per_second": 26.71,
      "eval_steps_per_second": 13.889,
      "num_input_tokens_seen": 20612176,
      "step": 37200
    },
    {
      "epoch": 652.7256637168142,
      "grad_norm": 8.625,
      "learning_rate": 6.003688625370291e-07,
      "loss": 1.0629,
      "num_input_tokens_seen": 20615472,
      "step": 37205
    },
    {
      "epoch": 652.8141592920354,
      "grad_norm": 8.4375,
      "learning_rate": 5.982321377707406e-07,
      "loss": 1.1041,
      "num_input_tokens_seen": 20617760,
      "step": 37210
    },
    {
      "epoch": 652.9026548672566,
      "grad_norm": 8.625,
      "learning_rate": 5.96099176063708e-07,
      "loss": 0.991,
      "num_input_tokens_seen": 20620288,
      "step": 37215
    },
    {
      "epoch": 652.9911504424779,
      "grad_norm": 10.1875,
      "learning_rate": 5.93969977744857e-07,
      "loss": 1.0587,
      "num_input_tokens_seen": 20623104,
      "step": 37220
    },
    {
      "epoch": 653.070796460177,
      "grad_norm": 9.6875,
      "learning_rate": 5.918445431425445e-07,
      "loss": 1.1241,
      "num_input_tokens_seen": 20625432,
      "step": 37225
    },
    {
      "epoch": 653.1592920353983,
      "grad_norm": 12.3125,
      "learning_rate": 5.897228725845333e-07,
      "loss": 1.1798,
      "num_input_tokens_seen": 20627720,
      "step": 37230
    },
    {
      "epoch": 653.2477876106195,
      "grad_norm": 6.09375,
      "learning_rate": 5.876049663980171e-07,
      "loss": 0.8273,
      "num_input_tokens_seen": 20630456,
      "step": 37235
    },
    {
      "epoch": 653.3362831858407,
      "grad_norm": 8.5625,
      "learning_rate": 5.854908249095959e-07,
      "loss": 1.0682,
      "num_input_tokens_seen": 20633464,
      "step": 37240
    },
    {
      "epoch": 653.4247787610619,
      "grad_norm": 8.1875,
      "learning_rate": 5.833804484453031e-07,
      "loss": 0.9748,
      "num_input_tokens_seen": 20636440,
      "step": 37245
    },
    {
      "epoch": 653.5132743362832,
      "grad_norm": 7.75,
      "learning_rate": 5.81273837330587e-07,
      "loss": 1.0483,
      "num_input_tokens_seen": 20638984,
      "step": 37250
    },
    {
      "epoch": 653.6017699115044,
      "grad_norm": 8.75,
      "learning_rate": 5.791709918903071e-07,
      "loss": 0.9493,
      "num_input_tokens_seen": 20641640,
      "step": 37255
    },
    {
      "epoch": 653.6902654867257,
      "grad_norm": 7.71875,
      "learning_rate": 5.770719124487483e-07,
      "loss": 1.1674,
      "num_input_tokens_seen": 20644856,
      "step": 37260
    },
    {
      "epoch": 653.7787610619469,
      "grad_norm": 9.5625,
      "learning_rate": 5.749765993296241e-07,
      "loss": 1.1755,
      "num_input_tokens_seen": 20647432,
      "step": 37265
    },
    {
      "epoch": 653.8672566371681,
      "grad_norm": 7.9375,
      "learning_rate": 5.728850528560509e-07,
      "loss": 0.9648,
      "num_input_tokens_seen": 20650696,
      "step": 37270
    },
    {
      "epoch": 653.9557522123894,
      "grad_norm": 8.0625,
      "learning_rate": 5.707972733505707e-07,
      "loss": 1.128,
      "num_input_tokens_seen": 20653560,
      "step": 37275
    },
    {
      "epoch": 654.0353982300885,
      "grad_norm": 8.5625,
      "learning_rate": 5.687132611351509e-07,
      "loss": 1.4098,
      "num_input_tokens_seen": 20655992,
      "step": 37280
    },
    {
      "epoch": 654.1238938053098,
      "grad_norm": 7.5625,
      "learning_rate": 5.666330165311651e-07,
      "loss": 1.1481,
      "num_input_tokens_seen": 20658616,
      "step": 37285
    },
    {
      "epoch": 654.212389380531,
      "grad_norm": 6.4375,
      "learning_rate": 5.645565398594204e-07,
      "loss": 0.9766,
      "num_input_tokens_seen": 20661944,
      "step": 37290
    },
    {
      "epoch": 654.3008849557522,
      "grad_norm": 8.3125,
      "learning_rate": 5.624838314401304e-07,
      "loss": 1.0232,
      "num_input_tokens_seen": 20665176,
      "step": 37295
    },
    {
      "epoch": 654.3893805309734,
      "grad_norm": 8.125,
      "learning_rate": 5.604148915929336e-07,
      "loss": 0.9331,
      "num_input_tokens_seen": 20668216,
      "step": 37300
    },
    {
      "epoch": 654.4778761061947,
      "grad_norm": 7.90625,
      "learning_rate": 5.583497206368887e-07,
      "loss": 1.2093,
      "num_input_tokens_seen": 20670712,
      "step": 37305
    },
    {
      "epoch": 654.566371681416,
      "grad_norm": 7.5625,
      "learning_rate": 5.562883188904688e-07,
      "loss": 0.9372,
      "num_input_tokens_seen": 20673208,
      "step": 37310
    },
    {
      "epoch": 654.6548672566372,
      "grad_norm": 8.125,
      "learning_rate": 5.542306866715724e-07,
      "loss": 1.1163,
      "num_input_tokens_seen": 20675896,
      "step": 37315
    },
    {
      "epoch": 654.7433628318585,
      "grad_norm": 7.15625,
      "learning_rate": 5.52176824297504e-07,
      "loss": 0.974,
      "num_input_tokens_seen": 20678840,
      "step": 37320
    },
    {
      "epoch": 654.8318584070796,
      "grad_norm": 7.3125,
      "learning_rate": 5.501267320850018e-07,
      "loss": 1.1698,
      "num_input_tokens_seen": 20681384,
      "step": 37325
    },
    {
      "epoch": 654.9203539823009,
      "grad_norm": 10.5625,
      "learning_rate": 5.480804103502157e-07,
      "loss": 0.9728,
      "num_input_tokens_seen": 20684424,
      "step": 37330
    },
    {
      "epoch": 655.0,
      "grad_norm": 17.375,
      "learning_rate": 5.460378594087101e-07,
      "loss": 1.1254,
      "num_input_tokens_seen": 20686464,
      "step": 37335
    },
    {
      "epoch": 655.0884955752213,
      "grad_norm": 5.59375,
      "learning_rate": 5.439990795754773e-07,
      "loss": 0.9731,
      "num_input_tokens_seen": 20689728,
      "step": 37340
    },
    {
      "epoch": 655.1769911504425,
      "grad_norm": 8.5,
      "learning_rate": 5.419640711649188e-07,
      "loss": 1.0545,
      "num_input_tokens_seen": 20692480,
      "step": 37345
    },
    {
      "epoch": 655.2654867256637,
      "grad_norm": 10.4375,
      "learning_rate": 5.399328344908583e-07,
      "loss": 1.0461,
      "num_input_tokens_seen": 20695072,
      "step": 37350
    },
    {
      "epoch": 655.3539823008849,
      "grad_norm": 8.625,
      "learning_rate": 5.379053698665399e-07,
      "loss": 0.9965,
      "num_input_tokens_seen": 20697936,
      "step": 37355
    },
    {
      "epoch": 655.4424778761062,
      "grad_norm": 7.78125,
      "learning_rate": 5.358816776046216e-07,
      "loss": 0.9482,
      "num_input_tokens_seen": 20700576,
      "step": 37360
    },
    {
      "epoch": 655.5309734513274,
      "grad_norm": 8.5,
      "learning_rate": 5.338617580171817e-07,
      "loss": 1.1435,
      "num_input_tokens_seen": 20703024,
      "step": 37365
    },
    {
      "epoch": 655.6194690265487,
      "grad_norm": 6.09375,
      "learning_rate": 5.318456114157239e-07,
      "loss": 1.0159,
      "num_input_tokens_seen": 20706064,
      "step": 37370
    },
    {
      "epoch": 655.70796460177,
      "grad_norm": 9.625,
      "learning_rate": 5.298332381111576e-07,
      "loss": 1.2072,
      "num_input_tokens_seen": 20708544,
      "step": 37375
    },
    {
      "epoch": 655.7964601769911,
      "grad_norm": 11.8125,
      "learning_rate": 5.27824638413818e-07,
      "loss": 1.0529,
      "num_input_tokens_seen": 20711536,
      "step": 37380
    },
    {
      "epoch": 655.8849557522124,
      "grad_norm": 10.5,
      "learning_rate": 5.258198126334546e-07,
      "loss": 0.912,
      "num_input_tokens_seen": 20714560,
      "step": 37385
    },
    {
      "epoch": 655.9734513274336,
      "grad_norm": 9.25,
      "learning_rate": 5.238187610792367e-07,
      "loss": 1.0977,
      "num_input_tokens_seen": 20717520,
      "step": 37390
    },
    {
      "epoch": 656.0530973451328,
      "grad_norm": 7.96875,
      "learning_rate": 5.218214840597563e-07,
      "loss": 1.1713,
      "num_input_tokens_seen": 20719960,
      "step": 37395
    },
    {
      "epoch": 656.141592920354,
      "grad_norm": 8.0625,
      "learning_rate": 5.198279818830115e-07,
      "loss": 1.3035,
      "num_input_tokens_seen": 20722344,
      "step": 37400
    },
    {
      "epoch": 656.141592920354,
      "eval_loss": 1.0310944318771362,
      "eval_runtime": 0.934,
      "eval_samples_per_second": 26.765,
      "eval_steps_per_second": 13.918,
      "num_input_tokens_seen": 20722344,
      "step": 37400
    },
    {
      "epoch": 656.2300884955753,
      "grad_norm": 7.78125,
      "learning_rate": 5.178382548564287e-07,
      "loss": 1.1476,
      "num_input_tokens_seen": 20724984,
      "step": 37405
    },
    {
      "epoch": 656.3185840707964,
      "grad_norm": 8.0,
      "learning_rate": 5.15852303286854e-07,
      "loss": 0.9792,
      "num_input_tokens_seen": 20727896,
      "step": 37410
    },
    {
      "epoch": 656.4070796460177,
      "grad_norm": 8.0625,
      "learning_rate": 5.138701274805396e-07,
      "loss": 0.8433,
      "num_input_tokens_seen": 20730744,
      "step": 37415
    },
    {
      "epoch": 656.4955752212389,
      "grad_norm": 9.125,
      "learning_rate": 5.118917277431606e-07,
      "loss": 1.2299,
      "num_input_tokens_seen": 20733432,
      "step": 37420
    },
    {
      "epoch": 656.5840707964602,
      "grad_norm": 9.1875,
      "learning_rate": 5.099171043798145e-07,
      "loss": 1.0411,
      "num_input_tokens_seen": 20736776,
      "step": 37425
    },
    {
      "epoch": 656.6725663716815,
      "grad_norm": 13.625,
      "learning_rate": 5.079462576950133e-07,
      "loss": 1.0096,
      "num_input_tokens_seen": 20739448,
      "step": 37430
    },
    {
      "epoch": 656.7610619469026,
      "grad_norm": 7.1875,
      "learning_rate": 5.059791879926862e-07,
      "loss": 1.0887,
      "num_input_tokens_seen": 20742072,
      "step": 37435
    },
    {
      "epoch": 656.8495575221239,
      "grad_norm": 10.0625,
      "learning_rate": 5.040158955761793e-07,
      "loss": 1.0621,
      "num_input_tokens_seen": 20744904,
      "step": 37440
    },
    {
      "epoch": 656.9380530973451,
      "grad_norm": 6.65625,
      "learning_rate": 5.020563807482559e-07,
      "loss": 0.9192,
      "num_input_tokens_seen": 20747928,
      "step": 37445
    },
    {
      "epoch": 657.0176991150443,
      "grad_norm": 7.9375,
      "learning_rate": 5.001006438110995e-07,
      "loss": 0.8142,
      "num_input_tokens_seen": 20750136,
      "step": 37450
    },
    {
      "epoch": 657.1061946902655,
      "grad_norm": 7.15625,
      "learning_rate": 4.981486850663075e-07,
      "loss": 1.0355,
      "num_input_tokens_seen": 20752536,
      "step": 37455
    },
    {
      "epoch": 657.1946902654868,
      "grad_norm": 8.3125,
      "learning_rate": 4.962005048149005e-07,
      "loss": 0.9909,
      "num_input_tokens_seen": 20755784,
      "step": 37460
    },
    {
      "epoch": 657.2831858407079,
      "grad_norm": 9.3125,
      "learning_rate": 4.942561033573073e-07,
      "loss": 1.1431,
      "num_input_tokens_seen": 20759000,
      "step": 37465
    },
    {
      "epoch": 657.3716814159292,
      "grad_norm": 9.1875,
      "learning_rate": 4.923154809933827e-07,
      "loss": 0.8553,
      "num_input_tokens_seen": 20762024,
      "step": 37470
    },
    {
      "epoch": 657.4601769911504,
      "grad_norm": 8.5,
      "learning_rate": 4.903786380223957e-07,
      "loss": 1.1585,
      "num_input_tokens_seen": 20764744,
      "step": 37475
    },
    {
      "epoch": 657.5486725663717,
      "grad_norm": 9.3125,
      "learning_rate": 4.884455747430266e-07,
      "loss": 1.0585,
      "num_input_tokens_seen": 20767352,
      "step": 37480
    },
    {
      "epoch": 657.637168141593,
      "grad_norm": 7.53125,
      "learning_rate": 4.865162914533816e-07,
      "loss": 0.9916,
      "num_input_tokens_seen": 20770216,
      "step": 37485
    },
    {
      "epoch": 657.7256637168142,
      "grad_norm": 9.5,
      "learning_rate": 4.845907884509809e-07,
      "loss": 1.0192,
      "num_input_tokens_seen": 20772648,
      "step": 37490
    },
    {
      "epoch": 657.8141592920354,
      "grad_norm": 8.75,
      "learning_rate": 4.82669066032762e-07,
      "loss": 0.9499,
      "num_input_tokens_seen": 20775912,
      "step": 37495
    },
    {
      "epoch": 657.9026548672566,
      "grad_norm": 9.5,
      "learning_rate": 4.807511244950768e-07,
      "loss": 1.2026,
      "num_input_tokens_seen": 20778552,
      "step": 37500
    },
    {
      "epoch": 657.9911504424779,
      "grad_norm": 9.375,
      "learning_rate": 4.788369641336943e-07,
      "loss": 1.1033,
      "num_input_tokens_seen": 20781128,
      "step": 37505
    },
    {
      "epoch": 658.070796460177,
      "grad_norm": 9.125,
      "learning_rate": 4.769265852438032e-07,
      "loss": 1.0769,
      "num_input_tokens_seen": 20783400,
      "step": 37510
    },
    {
      "epoch": 658.1592920353983,
      "grad_norm": 7.78125,
      "learning_rate": 4.750199881200124e-07,
      "loss": 1.0947,
      "num_input_tokens_seen": 20786664,
      "step": 37515
    },
    {
      "epoch": 658.2477876106195,
      "grad_norm": 10.4375,
      "learning_rate": 4.7311717305633664e-07,
      "loss": 0.9799,
      "num_input_tokens_seen": 20789544,
      "step": 37520
    },
    {
      "epoch": 658.3362831858407,
      "grad_norm": 9.6875,
      "learning_rate": 4.7121814034621623e-07,
      "loss": 1.0563,
      "num_input_tokens_seen": 20792456,
      "step": 37525
    },
    {
      "epoch": 658.4247787610619,
      "grad_norm": 7.78125,
      "learning_rate": 4.693228902825114e-07,
      "loss": 0.9762,
      "num_input_tokens_seen": 20795256,
      "step": 37530
    },
    {
      "epoch": 658.5132743362832,
      "grad_norm": 7.4375,
      "learning_rate": 4.6743142315748277e-07,
      "loss": 1.0264,
      "num_input_tokens_seen": 20797832,
      "step": 37535
    },
    {
      "epoch": 658.6017699115044,
      "grad_norm": 7.65625,
      "learning_rate": 4.655437392628276e-07,
      "loss": 1.0348,
      "num_input_tokens_seen": 20800504,
      "step": 37540
    },
    {
      "epoch": 658.6902654867257,
      "grad_norm": 8.5,
      "learning_rate": 4.636598388896463e-07,
      "loss": 1.0231,
      "num_input_tokens_seen": 20803400,
      "step": 37545
    },
    {
      "epoch": 658.7787610619469,
      "grad_norm": 9.375,
      "learning_rate": 4.6177972232845925e-07,
      "loss": 0.9932,
      "num_input_tokens_seen": 20806056,
      "step": 37550
    },
    {
      "epoch": 658.8672566371681,
      "grad_norm": 9.875,
      "learning_rate": 4.5990338986920953e-07,
      "loss": 1.0152,
      "num_input_tokens_seen": 20808728,
      "step": 37555
    },
    {
      "epoch": 658.9557522123894,
      "grad_norm": 8.8125,
      "learning_rate": 4.5803084180124633e-07,
      "loss": 1.236,
      "num_input_tokens_seen": 20811752,
      "step": 37560
    },
    {
      "epoch": 659.0353982300885,
      "grad_norm": 6.9375,
      "learning_rate": 4.561620784133386e-07,
      "loss": 0.9786,
      "num_input_tokens_seen": 20814056,
      "step": 37565
    },
    {
      "epoch": 659.1238938053098,
      "grad_norm": 6.78125,
      "learning_rate": 4.5429709999367796e-07,
      "loss": 1.1146,
      "num_input_tokens_seen": 20816744,
      "step": 37570
    },
    {
      "epoch": 659.212389380531,
      "grad_norm": 6.9375,
      "learning_rate": 4.5243590682986223e-07,
      "loss": 1.0089,
      "num_input_tokens_seen": 20819688,
      "step": 37575
    },
    {
      "epoch": 659.3008849557522,
      "grad_norm": 10.3125,
      "learning_rate": 4.5057849920891735e-07,
      "loss": 1.0868,
      "num_input_tokens_seen": 20822344,
      "step": 37580
    },
    {
      "epoch": 659.3893805309734,
      "grad_norm": 9.4375,
      "learning_rate": 4.487248774172698e-07,
      "loss": 1.0853,
      "num_input_tokens_seen": 20825256,
      "step": 37585
    },
    {
      "epoch": 659.4778761061947,
      "grad_norm": 9.6875,
      "learning_rate": 4.4687504174077965e-07,
      "loss": 1.2239,
      "num_input_tokens_seen": 20827832,
      "step": 37590
    },
    {
      "epoch": 659.566371681416,
      "grad_norm": 7.4375,
      "learning_rate": 4.450289924647133e-07,
      "loss": 0.9889,
      "num_input_tokens_seen": 20830648,
      "step": 37595
    },
    {
      "epoch": 659.6548672566372,
      "grad_norm": 8.375,
      "learning_rate": 4.431867298737513e-07,
      "loss": 0.9743,
      "num_input_tokens_seen": 20833640,
      "step": 37600
    },
    {
      "epoch": 659.6548672566372,
      "eval_loss": 1.0310944318771362,
      "eval_runtime": 0.9361,
      "eval_samples_per_second": 26.707,
      "eval_steps_per_second": 13.888,
      "num_input_tokens_seen": 20833640,
      "step": 37600
    },
    {
      "epoch": 659.7433628318585,
      "grad_norm": 8.9375,
      "learning_rate": 4.41348254251997e-07,
      "loss": 1.0729,
      "num_input_tokens_seen": 20836728,
      "step": 37605
    },
    {
      "epoch": 659.8318584070796,
      "grad_norm": 7.3125,
      "learning_rate": 4.395135658829652e-07,
      "loss": 1.0529,
      "num_input_tokens_seen": 20839384,
      "step": 37610
    },
    {
      "epoch": 659.9203539823009,
      "grad_norm": 6.59375,
      "learning_rate": 4.376826650495852e-07,
      "loss": 1.0164,
      "num_input_tokens_seen": 20842360,
      "step": 37615
    },
    {
      "epoch": 660.0,
      "grad_norm": 7.78125,
      "learning_rate": 4.358555520342117e-07,
      "loss": 0.9355,
      "num_input_tokens_seen": 20844680,
      "step": 37620
    },
    {
      "epoch": 660.0884955752213,
      "grad_norm": 7.625,
      "learning_rate": 4.3403222711860257e-07,
      "loss": 1.1003,
      "num_input_tokens_seen": 20848008,
      "step": 37625
    },
    {
      "epoch": 660.1769911504425,
      "grad_norm": 8.875,
      "learning_rate": 4.3221269058394133e-07,
      "loss": 1.0782,
      "num_input_tokens_seen": 20850680,
      "step": 37630
    },
    {
      "epoch": 660.2654867256637,
      "grad_norm": 7.625,
      "learning_rate": 4.303969427108173e-07,
      "loss": 1.1185,
      "num_input_tokens_seen": 20853144,
      "step": 37635
    },
    {
      "epoch": 660.3539823008849,
      "grad_norm": 8.9375,
      "learning_rate": 4.2858498377924825e-07,
      "loss": 0.9515,
      "num_input_tokens_seen": 20856088,
      "step": 37640
    },
    {
      "epoch": 660.4424778761062,
      "grad_norm": 8.375,
      "learning_rate": 4.267768140686579e-07,
      "loss": 1.0909,
      "num_input_tokens_seen": 20858744,
      "step": 37645
    },
    {
      "epoch": 660.5309734513274,
      "grad_norm": 11.0,
      "learning_rate": 4.2497243385788975e-07,
      "loss": 1.0054,
      "num_input_tokens_seen": 20860984,
      "step": 37650
    },
    {
      "epoch": 660.6194690265487,
      "grad_norm": 8.4375,
      "learning_rate": 4.231718434251991e-07,
      "loss": 1.0581,
      "num_input_tokens_seen": 20863784,
      "step": 37655
    },
    {
      "epoch": 660.70796460177,
      "grad_norm": 7.875,
      "learning_rate": 4.213750430482666e-07,
      "loss": 0.7878,
      "num_input_tokens_seen": 20866200,
      "step": 37660
    },
    {
      "epoch": 660.7964601769911,
      "grad_norm": 8.0,
      "learning_rate": 4.1958203300417054e-07,
      "loss": 1.1408,
      "num_input_tokens_seen": 20868952,
      "step": 37665
    },
    {
      "epoch": 660.8849557522124,
      "grad_norm": 7.0625,
      "learning_rate": 4.177928135694259e-07,
      "loss": 0.9335,
      "num_input_tokens_seen": 20872120,
      "step": 37670
    },
    {
      "epoch": 660.9734513274336,
      "grad_norm": 7.78125,
      "learning_rate": 4.1600738501994807e-07,
      "loss": 1.2579,
      "num_input_tokens_seen": 20875320,
      "step": 37675
    },
    {
      "epoch": 661.0530973451328,
      "grad_norm": 9.625,
      "learning_rate": 4.1422574763107237e-07,
      "loss": 1.5699,
      "num_input_tokens_seen": 20877640,
      "step": 37680
    },
    {
      "epoch": 661.141592920354,
      "grad_norm": 9.375,
      "learning_rate": 4.124479016775512e-07,
      "loss": 1.0971,
      "num_input_tokens_seen": 20880456,
      "step": 37685
    },
    {
      "epoch": 661.2300884955753,
      "grad_norm": 6.375,
      "learning_rate": 4.106738474335514e-07,
      "loss": 0.9086,
      "num_input_tokens_seen": 20882984,
      "step": 37690
    },
    {
      "epoch": 661.3185840707964,
      "grad_norm": 7.0625,
      "learning_rate": 4.089035851726486e-07,
      "loss": 1.0016,
      "num_input_tokens_seen": 20885752,
      "step": 37695
    },
    {
      "epoch": 661.4070796460177,
      "grad_norm": 6.65625,
      "learning_rate": 4.0713711516784937e-07,
      "loss": 1.2366,
      "num_input_tokens_seen": 20888104,
      "step": 37700
    },
    {
      "epoch": 661.4955752212389,
      "grad_norm": 8.625,
      "learning_rate": 4.05374437691558e-07,
      "loss": 0.8433,
      "num_input_tokens_seen": 20891224,
      "step": 37705
    },
    {
      "epoch": 661.5840707964602,
      "grad_norm": 8.25,
      "learning_rate": 4.036155530156044e-07,
      "loss": 1.0912,
      "num_input_tokens_seen": 20893976,
      "step": 37710
    },
    {
      "epoch": 661.6725663716815,
      "grad_norm": 12.3125,
      "learning_rate": 4.018604614112298e-07,
      "loss": 0.9553,
      "num_input_tokens_seen": 20897112,
      "step": 37715
    },
    {
      "epoch": 661.7610619469026,
      "grad_norm": 10.9375,
      "learning_rate": 4.0010916314908996e-07,
      "loss": 0.9951,
      "num_input_tokens_seen": 20899896,
      "step": 37720
    },
    {
      "epoch": 661.8495575221239,
      "grad_norm": 10.125,
      "learning_rate": 3.983616584992578e-07,
      "loss": 1.233,
      "num_input_tokens_seen": 20902520,
      "step": 37725
    },
    {
      "epoch": 661.9380530973451,
      "grad_norm": 6.75,
      "learning_rate": 3.9661794773122595e-07,
      "loss": 1.1352,
      "num_input_tokens_seen": 20905512,
      "step": 37730
    },
    {
      "epoch": 662.0176991150443,
      "grad_norm": 9.625,
      "learning_rate": 3.9487803111388777e-07,
      "loss": 1.0902,
      "num_input_tokens_seen": 20908032,
      "step": 37735
    },
    {
      "epoch": 662.1061946902655,
      "grad_norm": 5.53125,
      "learning_rate": 3.9314190891556747e-07,
      "loss": 0.9181,
      "num_input_tokens_seen": 20911088,
      "step": 37740
    },
    {
      "epoch": 662.1946902654868,
      "grad_norm": 6.6875,
      "learning_rate": 3.914095814039925e-07,
      "loss": 0.8201,
      "num_input_tokens_seen": 20913760,
      "step": 37745
    },
    {
      "epoch": 662.2831858407079,
      "grad_norm": 7.21875,
      "learning_rate": 3.896810488463104e-07,
      "loss": 1.2093,
      "num_input_tokens_seen": 20916416,
      "step": 37750
    },
    {
      "epoch": 662.3716814159292,
      "grad_norm": 9.9375,
      "learning_rate": 3.8795631150908565e-07,
      "loss": 1.0578,
      "num_input_tokens_seen": 20919056,
      "step": 37755
    },
    {
      "epoch": 662.4601769911504,
      "grad_norm": 10.4375,
      "learning_rate": 3.862353696582888e-07,
      "loss": 1.0963,
      "num_input_tokens_seen": 20922000,
      "step": 37760
    },
    {
      "epoch": 662.5486725663717,
      "grad_norm": 8.125,
      "learning_rate": 3.8451822355931313e-07,
      "loss": 1.049,
      "num_input_tokens_seen": 20924560,
      "step": 37765
    },
    {
      "epoch": 662.637168141593,
      "grad_norm": 12.875,
      "learning_rate": 3.82804873476969e-07,
      "loss": 1.1675,
      "num_input_tokens_seen": 20927776,
      "step": 37770
    },
    {
      "epoch": 662.7256637168142,
      "grad_norm": 9.0,
      "learning_rate": 3.810953196754702e-07,
      "loss": 1.2751,
      "num_input_tokens_seen": 20930928,
      "step": 37775
    },
    {
      "epoch": 662.8141592920354,
      "grad_norm": 7.28125,
      "learning_rate": 3.793895624184529e-07,
      "loss": 0.889,
      "num_input_tokens_seen": 20934000,
      "step": 37780
    },
    {
      "epoch": 662.9026548672566,
      "grad_norm": 8.0625,
      "learning_rate": 3.776876019689679e-07,
      "loss": 1.1102,
      "num_input_tokens_seen": 20936336,
      "step": 37785
    },
    {
      "epoch": 662.9911504424779,
      "grad_norm": 6.84375,
      "learning_rate": 3.7598943858947743e-07,
      "loss": 0.9761,
      "num_input_tokens_seen": 20939216,
      "step": 37790
    },
    {
      "epoch": 663.070796460177,
      "grad_norm": 7.59375,
      "learning_rate": 3.742950725418637e-07,
      "loss": 1.0138,
      "num_input_tokens_seen": 20941712,
      "step": 37795
    },
    {
      "epoch": 663.1592920353983,
      "grad_norm": 8.6875,
      "learning_rate": 3.726045040874093e-07,
      "loss": 1.2551,
      "num_input_tokens_seen": 20944256,
      "step": 37800
    },
    {
      "epoch": 663.1592920353983,
      "eval_loss": 1.0310944318771362,
      "eval_runtime": 0.9444,
      "eval_samples_per_second": 26.472,
      "eval_steps_per_second": 13.766,
      "num_input_tokens_seen": 20944256,
      "step": 37800
    },
    {
      "epoch": 663.2477876106195,
      "grad_norm": 9.0,
      "learning_rate": 3.709177334868308e-07,
      "loss": 1.0576,
      "num_input_tokens_seen": 20946784,
      "step": 37805
    },
    {
      "epoch": 663.3362831858407,
      "grad_norm": 8.3125,
      "learning_rate": 3.692347610002478e-07,
      "loss": 1.087,
      "num_input_tokens_seen": 20949744,
      "step": 37810
    },
    {
      "epoch": 663.4247787610619,
      "grad_norm": 7.75,
      "learning_rate": 3.675555868871916e-07,
      "loss": 1.0502,
      "num_input_tokens_seen": 20952624,
      "step": 37815
    },
    {
      "epoch": 663.5132743362832,
      "grad_norm": 8.0,
      "learning_rate": 3.658802114066162e-07,
      "loss": 0.8468,
      "num_input_tokens_seen": 20955824,
      "step": 37820
    },
    {
      "epoch": 663.6017699115044,
      "grad_norm": 8.5,
      "learning_rate": 3.6420863481688437e-07,
      "loss": 0.9611,
      "num_input_tokens_seen": 20958720,
      "step": 37825
    },
    {
      "epoch": 663.6902654867257,
      "grad_norm": 7.375,
      "learning_rate": 3.625408573757705e-07,
      "loss": 1.2353,
      "num_input_tokens_seen": 20961632,
      "step": 37830
    },
    {
      "epoch": 663.7787610619469,
      "grad_norm": 8.5,
      "learning_rate": 3.608768793404743e-07,
      "loss": 1.1392,
      "num_input_tokens_seen": 20964560,
      "step": 37835
    },
    {
      "epoch": 663.8672566371681,
      "grad_norm": 9.375,
      "learning_rate": 3.592167009675934e-07,
      "loss": 0.9299,
      "num_input_tokens_seen": 20967296,
      "step": 37840
    },
    {
      "epoch": 663.9557522123894,
      "grad_norm": 9.5,
      "learning_rate": 3.575603225131563e-07,
      "loss": 0.8603,
      "num_input_tokens_seen": 20970128,
      "step": 37845
    },
    {
      "epoch": 664.0353982300885,
      "grad_norm": 8.6875,
      "learning_rate": 3.55907744232592e-07,
      "loss": 0.9243,
      "num_input_tokens_seen": 20972472,
      "step": 37850
    },
    {
      "epoch": 664.1238938053098,
      "grad_norm": 10.8125,
      "learning_rate": 3.5425896638075217e-07,
      "loss": 1.0724,
      "num_input_tokens_seen": 20975160,
      "step": 37855
    },
    {
      "epoch": 664.212389380531,
      "grad_norm": 6.8125,
      "learning_rate": 3.5261398921189736e-07,
      "loss": 1.0135,
      "num_input_tokens_seen": 20978184,
      "step": 37860
    },
    {
      "epoch": 664.3008849557522,
      "grad_norm": 7.78125,
      "learning_rate": 3.509728129797024e-07,
      "loss": 1.1744,
      "num_input_tokens_seen": 20980792,
      "step": 37865
    },
    {
      "epoch": 664.3893805309734,
      "grad_norm": 8.8125,
      "learning_rate": 3.4933543793725656e-07,
      "loss": 1.0233,
      "num_input_tokens_seen": 20984104,
      "step": 37870
    },
    {
      "epoch": 664.4778761061947,
      "grad_norm": 7.09375,
      "learning_rate": 3.4770186433707163e-07,
      "loss": 0.9102,
      "num_input_tokens_seen": 20986920,
      "step": 37875
    },
    {
      "epoch": 664.566371681416,
      "grad_norm": 8.1875,
      "learning_rate": 3.4607209243105453e-07,
      "loss": 1.1176,
      "num_input_tokens_seen": 20989640,
      "step": 37880
    },
    {
      "epoch": 664.6548672566372,
      "grad_norm": 8.0,
      "learning_rate": 3.444461224705431e-07,
      "loss": 1.1561,
      "num_input_tokens_seen": 20992280,
      "step": 37885
    },
    {
      "epoch": 664.7433628318585,
      "grad_norm": 10.5,
      "learning_rate": 3.4282395470628116e-07,
      "loss": 1.1696,
      "num_input_tokens_seen": 20995096,
      "step": 37890
    },
    {
      "epoch": 664.8318584070796,
      "grad_norm": 12.75,
      "learning_rate": 3.4120558938842417e-07,
      "loss": 0.9182,
      "num_input_tokens_seen": 20997736,
      "step": 37895
    },
    {
      "epoch": 664.9203539823009,
      "grad_norm": 9.9375,
      "learning_rate": 3.395910267665503e-07,
      "loss": 1.0208,
      "num_input_tokens_seen": 21000792,
      "step": 37900
    },
    {
      "epoch": 665.0,
      "grad_norm": 11.375,
      "learning_rate": 3.3798026708964094e-07,
      "loss": 0.8837,
      "num_input_tokens_seen": 21003144,
      "step": 37905
    },
    {
      "epoch": 665.0884955752213,
      "grad_norm": 7.25,
      "learning_rate": 3.3637331060609456e-07,
      "loss": 1.0997,
      "num_input_tokens_seen": 21005352,
      "step": 37910
    },
    {
      "epoch": 665.1769911504425,
      "grad_norm": 8.5625,
      "learning_rate": 3.3477015756372966e-07,
      "loss": 1.156,
      "num_input_tokens_seen": 21008056,
      "step": 37915
    },
    {
      "epoch": 665.2654867256637,
      "grad_norm": 6.90625,
      "learning_rate": 3.3317080820976785e-07,
      "loss": 1.0284,
      "num_input_tokens_seen": 21010856,
      "step": 37920
    },
    {
      "epoch": 665.3539823008849,
      "grad_norm": 9.875,
      "learning_rate": 3.315752627908508e-07,
      "loss": 0.885,
      "num_input_tokens_seen": 21013912,
      "step": 37925
    },
    {
      "epoch": 665.4424778761062,
      "grad_norm": 9.0,
      "learning_rate": 3.299835215530317e-07,
      "loss": 1.075,
      "num_input_tokens_seen": 21016488,
      "step": 37930
    },
    {
      "epoch": 665.5309734513274,
      "grad_norm": 7.9375,
      "learning_rate": 3.2839558474177245e-07,
      "loss": 1.0927,
      "num_input_tokens_seen": 21019160,
      "step": 37935
    },
    {
      "epoch": 665.6194690265487,
      "grad_norm": 10.375,
      "learning_rate": 3.2681145260196056e-07,
      "loss": 1.1288,
      "num_input_tokens_seen": 21022440,
      "step": 37940
    },
    {
      "epoch": 665.70796460177,
      "grad_norm": 8.5,
      "learning_rate": 3.252311253778839e-07,
      "loss": 1.0246,
      "num_input_tokens_seen": 21025208,
      "step": 37945
    },
    {
      "epoch": 665.7964601769911,
      "grad_norm": 8.4375,
      "learning_rate": 3.2365460331325034e-07,
      "loss": 0.9807,
      "num_input_tokens_seen": 21028152,
      "step": 37950
    },
    {
      "epoch": 665.8849557522124,
      "grad_norm": 8.9375,
      "learning_rate": 3.2208188665117934e-07,
      "loss": 1.04,
      "num_input_tokens_seen": 21031448,
      "step": 37955
    },
    {
      "epoch": 665.9734513274336,
      "grad_norm": 8.4375,
      "learning_rate": 3.205129756342018e-07,
      "loss": 1.0425,
      "num_input_tokens_seen": 21033992,
      "step": 37960
    },
    {
      "epoch": 666.0530973451328,
      "grad_norm": 8.3125,
      "learning_rate": 3.189478705042659e-07,
      "loss": 1.1737,
      "num_input_tokens_seen": 21036280,
      "step": 37965
    },
    {
      "epoch": 666.141592920354,
      "grad_norm": 9.0625,
      "learning_rate": 3.173865715027341e-07,
      "loss": 1.0159,
      "num_input_tokens_seen": 21038984,
      "step": 37970
    },
    {
      "epoch": 666.2300884955753,
      "grad_norm": 5.46875,
      "learning_rate": 3.158290788703694e-07,
      "loss": 1.0351,
      "num_input_tokens_seen": 21041960,
      "step": 37975
    },
    {
      "epoch": 666.3185840707964,
      "grad_norm": 7.5625,
      "learning_rate": 3.1427539284736297e-07,
      "loss": 0.945,
      "num_input_tokens_seen": 21044824,
      "step": 37980
    },
    {
      "epoch": 666.4070796460177,
      "grad_norm": 9.25,
      "learning_rate": 3.127255136733093e-07,
      "loss": 1.217,
      "num_input_tokens_seen": 21047384,
      "step": 37985
    },
    {
      "epoch": 666.4955752212389,
      "grad_norm": 7.03125,
      "learning_rate": 3.1117944158722544e-07,
      "loss": 1.0937,
      "num_input_tokens_seen": 21050360,
      "step": 37990
    },
    {
      "epoch": 666.5840707964602,
      "grad_norm": 8.875,
      "learning_rate": 3.0963717682752635e-07,
      "loss": 1.0432,
      "num_input_tokens_seen": 21052968,
      "step": 37995
    },
    {
      "epoch": 666.6725663716815,
      "grad_norm": 5.625,
      "learning_rate": 3.080987196320578e-07,
      "loss": 0.7373,
      "num_input_tokens_seen": 21055624,
      "step": 38000
    },
    {
      "epoch": 666.6725663716815,
      "eval_loss": 1.0310944318771362,
      "eval_runtime": 0.9512,
      "eval_samples_per_second": 26.282,
      "eval_steps_per_second": 13.666,
      "num_input_tokens_seen": 21055624,
      "step": 38000
    },
    {
      "epoch": 666.7610619469026,
      "grad_norm": 8.1875,
      "learning_rate": 3.065640702380607e-07,
      "loss": 1.1273,
      "num_input_tokens_seen": 21058920,
      "step": 38005
    },
    {
      "epoch": 666.8495575221239,
      "grad_norm": 7.875,
      "learning_rate": 3.050332288822011e-07,
      "loss": 1.0845,
      "num_input_tokens_seen": 21061464,
      "step": 38010
    },
    {
      "epoch": 666.9380530973451,
      "grad_norm": 8.0625,
      "learning_rate": 3.035061958005542e-07,
      "loss": 1.1123,
      "num_input_tokens_seen": 21064168,
      "step": 38015
    },
    {
      "epoch": 667.0176991150443,
      "grad_norm": 8.125,
      "learning_rate": 3.019829712286093e-07,
      "loss": 0.9081,
      "num_input_tokens_seen": 21066264,
      "step": 38020
    },
    {
      "epoch": 667.1061946902655,
      "grad_norm": 9.625,
      "learning_rate": 3.004635554012647e-07,
      "loss": 0.9718,
      "num_input_tokens_seen": 21069592,
      "step": 38025
    },
    {
      "epoch": 667.1946902654868,
      "grad_norm": 9.5625,
      "learning_rate": 2.9894794855283017e-07,
      "loss": 1.1606,
      "num_input_tokens_seen": 21072184,
      "step": 38030
    },
    {
      "epoch": 667.2831858407079,
      "grad_norm": 8.1875,
      "learning_rate": 2.9743615091703816e-07,
      "loss": 1.0095,
      "num_input_tokens_seen": 21075032,
      "step": 38035
    },
    {
      "epoch": 667.3716814159292,
      "grad_norm": 12.0,
      "learning_rate": 2.959281627270216e-07,
      "loss": 1.1784,
      "num_input_tokens_seen": 21077736,
      "step": 38040
    },
    {
      "epoch": 667.4601769911504,
      "grad_norm": 8.8125,
      "learning_rate": 2.944239842153362e-07,
      "loss": 1.1022,
      "num_input_tokens_seen": 21080824,
      "step": 38045
    },
    {
      "epoch": 667.5486725663717,
      "grad_norm": 10.1875,
      "learning_rate": 2.929236156139381e-07,
      "loss": 1.1474,
      "num_input_tokens_seen": 21083384,
      "step": 38050
    },
    {
      "epoch": 667.637168141593,
      "grad_norm": 8.625,
      "learning_rate": 2.9142705715420883e-07,
      "loss": 0.9666,
      "num_input_tokens_seen": 21086648,
      "step": 38055
    },
    {
      "epoch": 667.7256637168142,
      "grad_norm": 7.75,
      "learning_rate": 2.8993430906693595e-07,
      "loss": 0.9803,
      "num_input_tokens_seen": 21088952,
      "step": 38060
    },
    {
      "epoch": 667.8141592920354,
      "grad_norm": 9.25,
      "learning_rate": 2.88445371582316e-07,
      "loss": 0.9624,
      "num_input_tokens_seen": 21091640,
      "step": 38065
    },
    {
      "epoch": 667.9026548672566,
      "grad_norm": 7.3125,
      "learning_rate": 2.8696024492996796e-07,
      "loss": 0.996,
      "num_input_tokens_seen": 21094712,
      "step": 38070
    },
    {
      "epoch": 667.9911504424779,
      "grad_norm": 7.0625,
      "learning_rate": 2.854789293389115e-07,
      "loss": 1.0422,
      "num_input_tokens_seen": 21097432,
      "step": 38075
    },
    {
      "epoch": 668.070796460177,
      "grad_norm": 7.75,
      "learning_rate": 2.8400142503758606e-07,
      "loss": 1.0656,
      "num_input_tokens_seen": 21099800,
      "step": 38080
    },
    {
      "epoch": 668.1592920353983,
      "grad_norm": 10.625,
      "learning_rate": 2.8252773225384276e-07,
      "loss": 1.0524,
      "num_input_tokens_seen": 21102440,
      "step": 38085
    },
    {
      "epoch": 668.2477876106195,
      "grad_norm": 6.625,
      "learning_rate": 2.8105785121494143e-07,
      "loss": 0.9015,
      "num_input_tokens_seen": 21105240,
      "step": 38090
    },
    {
      "epoch": 668.3362831858407,
      "grad_norm": 8.3125,
      "learning_rate": 2.795917821475563e-07,
      "loss": 1.1187,
      "num_input_tokens_seen": 21108040,
      "step": 38095
    },
    {
      "epoch": 668.4247787610619,
      "grad_norm": 9.75,
      "learning_rate": 2.78129525277776e-07,
      "loss": 1.1593,
      "num_input_tokens_seen": 21110904,
      "step": 38100
    },
    {
      "epoch": 668.5132743362832,
      "grad_norm": 8.0,
      "learning_rate": 2.766710808310952e-07,
      "loss": 0.8319,
      "num_input_tokens_seen": 21114120,
      "step": 38105
    },
    {
      "epoch": 668.6017699115044,
      "grad_norm": 11.375,
      "learning_rate": 2.7521644903242827e-07,
      "loss": 1.184,
      "num_input_tokens_seen": 21117224,
      "step": 38110
    },
    {
      "epoch": 668.6902654867257,
      "grad_norm": 10.3125,
      "learning_rate": 2.7376563010609593e-07,
      "loss": 1.0416,
      "num_input_tokens_seen": 21119688,
      "step": 38115
    },
    {
      "epoch": 668.7787610619469,
      "grad_norm": 8.1875,
      "learning_rate": 2.72318624275833e-07,
      "loss": 1.1108,
      "num_input_tokens_seen": 21122184,
      "step": 38120
    },
    {
      "epoch": 668.8672566371681,
      "grad_norm": 8.0625,
      "learning_rate": 2.7087543176478324e-07,
      "loss": 1.0236,
      "num_input_tokens_seen": 21125096,
      "step": 38125
    },
    {
      "epoch": 668.9557522123894,
      "grad_norm": 7.78125,
      "learning_rate": 2.694360527955103e-07,
      "loss": 1.0573,
      "num_input_tokens_seen": 21127736,
      "step": 38130
    },
    {
      "epoch": 669.0353982300885,
      "grad_norm": 8.0625,
      "learning_rate": 2.680004875899811e-07,
      "loss": 1.1872,
      "num_input_tokens_seen": 21129696,
      "step": 38135
    },
    {
      "epoch": 669.1238938053098,
      "grad_norm": 8.1875,
      "learning_rate": 2.665687363695768e-07,
      "loss": 1.0162,
      "num_input_tokens_seen": 21132384,
      "step": 38140
    },
    {
      "epoch": 669.212389380531,
      "grad_norm": 8.0625,
      "learning_rate": 2.6514079935509584e-07,
      "loss": 1.0758,
      "num_input_tokens_seen": 21134912,
      "step": 38145
    },
    {
      "epoch": 669.3008849557522,
      "grad_norm": 8.875,
      "learning_rate": 2.6371667676673983e-07,
      "loss": 1.1633,
      "num_input_tokens_seen": 21137472,
      "step": 38150
    },
    {
      "epoch": 669.3893805309734,
      "grad_norm": 9.875,
      "learning_rate": 2.6229636882412755e-07,
      "loss": 1.0916,
      "num_input_tokens_seen": 21140752,
      "step": 38155
    },
    {
      "epoch": 669.4778761061947,
      "grad_norm": 7.0,
      "learning_rate": 2.6087987574628935e-07,
      "loss": 1.0191,
      "num_input_tokens_seen": 21143520,
      "step": 38160
    },
    {
      "epoch": 669.566371681416,
      "grad_norm": 7.71875,
      "learning_rate": 2.5946719775166437e-07,
      "loss": 0.7388,
      "num_input_tokens_seen": 21146560,
      "step": 38165
    },
    {
      "epoch": 669.6548672566372,
      "grad_norm": 8.4375,
      "learning_rate": 2.5805833505810616e-07,
      "loss": 1.1273,
      "num_input_tokens_seen": 21149472,
      "step": 38170
    },
    {
      "epoch": 669.7433628318585,
      "grad_norm": 7.9375,
      "learning_rate": 2.566532878828798e-07,
      "loss": 1.0008,
      "num_input_tokens_seen": 21152464,
      "step": 38175
    },
    {
      "epoch": 669.8318584070796,
      "grad_norm": 10.5,
      "learning_rate": 2.552520564426619e-07,
      "loss": 1.2678,
      "num_input_tokens_seen": 21155168,
      "step": 38180
    },
    {
      "epoch": 669.9203539823009,
      "grad_norm": 10.9375,
      "learning_rate": 2.5385464095353803e-07,
      "loss": 0.875,
      "num_input_tokens_seen": 21158160,
      "step": 38185
    },
    {
      "epoch": 670.0,
      "grad_norm": 9.875,
      "learning_rate": 2.5246104163100804e-07,
      "loss": 1.2067,
      "num_input_tokens_seen": 21160240,
      "step": 38190
    },
    {
      "epoch": 670.0884955752213,
      "grad_norm": 9.4375,
      "learning_rate": 2.510712586899833e-07,
      "loss": 1.0719,
      "num_input_tokens_seen": 21162912,
      "step": 38195
    },
    {
      "epoch": 670.1769911504425,
      "grad_norm": 8.9375,
      "learning_rate": 2.4968529234478124e-07,
      "loss": 1.0018,
      "num_input_tokens_seen": 21165744,
      "step": 38200
    },
    {
      "epoch": 670.1769911504425,
      "eval_loss": 1.0310944318771362,
      "eval_runtime": 1.0124,
      "eval_samples_per_second": 24.695,
      "eval_steps_per_second": 12.841,
      "num_input_tokens_seen": 21165744,
      "step": 38200
    },
    {
      "epoch": 670.2654867256637,
      "grad_norm": 7.21875,
      "learning_rate": 2.483031428091448e-07,
      "loss": 1.0698,
      "num_input_tokens_seen": 21168624,
      "step": 38205
    },
    {
      "epoch": 670.3539823008849,
      "grad_norm": 6.65625,
      "learning_rate": 2.469248102962091e-07,
      "loss": 1.0595,
      "num_input_tokens_seen": 21171664,
      "step": 38210
    },
    {
      "epoch": 670.4424778761062,
      "grad_norm": 8.625,
      "learning_rate": 2.4555029501853455e-07,
      "loss": 0.9861,
      "num_input_tokens_seen": 21174592,
      "step": 38215
    },
    {
      "epoch": 670.5309734513274,
      "grad_norm": 8.25,
      "learning_rate": 2.441795971880906e-07,
      "loss": 1.0567,
      "num_input_tokens_seen": 21177456,
      "step": 38220
    },
    {
      "epoch": 670.6194690265487,
      "grad_norm": 9.1875,
      "learning_rate": 2.4281271701625255e-07,
      "loss": 1.1393,
      "num_input_tokens_seen": 21180368,
      "step": 38225
    },
    {
      "epoch": 670.70796460177,
      "grad_norm": 8.4375,
      "learning_rate": 2.4144965471381007e-07,
      "loss": 0.7925,
      "num_input_tokens_seen": 21183024,
      "step": 38230
    },
    {
      "epoch": 670.7964601769911,
      "grad_norm": 8.5625,
      "learning_rate": 2.400904104909674e-07,
      "loss": 1.1828,
      "num_input_tokens_seen": 21186176,
      "step": 38235
    },
    {
      "epoch": 670.8849557522124,
      "grad_norm": 9.1875,
      "learning_rate": 2.3873498455733725e-07,
      "loss": 1.1055,
      "num_input_tokens_seen": 21188944,
      "step": 38240
    },
    {
      "epoch": 670.9734513274336,
      "grad_norm": 10.5,
      "learning_rate": 2.3738337712194137e-07,
      "loss": 1.1337,
      "num_input_tokens_seen": 21191424,
      "step": 38245
    },
    {
      "epoch": 671.0530973451328,
      "grad_norm": 10.625,
      "learning_rate": 2.3603558839321305e-07,
      "loss": 0.685,
      "num_input_tokens_seen": 21194104,
      "step": 38250
    },
    {
      "epoch": 671.141592920354,
      "grad_norm": 8.5625,
      "learning_rate": 2.3469161857900267e-07,
      "loss": 0.9356,
      "num_input_tokens_seen": 21196872,
      "step": 38255
    },
    {
      "epoch": 671.2300884955753,
      "grad_norm": 7.03125,
      "learning_rate": 2.3335146788656393e-07,
      "loss": 1.0726,
      "num_input_tokens_seen": 21199624,
      "step": 38260
    },
    {
      "epoch": 671.3185840707964,
      "grad_norm": 9.875,
      "learning_rate": 2.3201513652256757e-07,
      "loss": 1.14,
      "num_input_tokens_seen": 21202232,
      "step": 38265
    },
    {
      "epoch": 671.4070796460177,
      "grad_norm": 9.125,
      "learning_rate": 2.3068262469308766e-07,
      "loss": 1.0135,
      "num_input_tokens_seen": 21205224,
      "step": 38270
    },
    {
      "epoch": 671.4955752212389,
      "grad_norm": 7.90625,
      "learning_rate": 2.2935393260362093e-07,
      "loss": 1.0813,
      "num_input_tokens_seen": 21207656,
      "step": 38275
    },
    {
      "epoch": 671.5840707964602,
      "grad_norm": 8.5625,
      "learning_rate": 2.2802906045906458e-07,
      "loss": 1.209,
      "num_input_tokens_seen": 21210280,
      "step": 38280
    },
    {
      "epoch": 671.6725663716815,
      "grad_norm": 5.21875,
      "learning_rate": 2.2670800846373018e-07,
      "loss": 0.9931,
      "num_input_tokens_seen": 21213240,
      "step": 38285
    },
    {
      "epoch": 671.7610619469026,
      "grad_norm": 8.875,
      "learning_rate": 2.2539077682134367e-07,
      "loss": 1.1048,
      "num_input_tokens_seen": 21215992,
      "step": 38290
    },
    {
      "epoch": 671.8495575221239,
      "grad_norm": 7.9375,
      "learning_rate": 2.2407736573503423e-07,
      "loss": 1.1071,
      "num_input_tokens_seen": 21218792,
      "step": 38295
    },
    {
      "epoch": 671.9380530973451,
      "grad_norm": 8.875,
      "learning_rate": 2.2276777540735093e-07,
      "loss": 1.1139,
      "num_input_tokens_seen": 21221320,
      "step": 38300
    },
    {
      "epoch": 672.0176991150443,
      "grad_norm": 8.1875,
      "learning_rate": 2.2146200604024613e-07,
      "loss": 0.7383,
      "num_input_tokens_seen": 21224056,
      "step": 38305
    },
    {
      "epoch": 672.1061946902655,
      "grad_norm": 7.125,
      "learning_rate": 2.2016005783508375e-07,
      "loss": 1.1037,
      "num_input_tokens_seen": 21226856,
      "step": 38310
    },
    {
      "epoch": 672.1946902654868,
      "grad_norm": 9.4375,
      "learning_rate": 2.1886193099264763e-07,
      "loss": 1.1473,
      "num_input_tokens_seen": 21229352,
      "step": 38315
    },
    {
      "epoch": 672.2831858407079,
      "grad_norm": 7.0625,
      "learning_rate": 2.175676257131165e-07,
      "loss": 1.1303,
      "num_input_tokens_seen": 21232328,
      "step": 38320
    },
    {
      "epoch": 672.3716814159292,
      "grad_norm": 8.6875,
      "learning_rate": 2.162771421960974e-07,
      "loss": 1.0979,
      "num_input_tokens_seen": 21234952,
      "step": 38325
    },
    {
      "epoch": 672.4601769911504,
      "grad_norm": 11.375,
      "learning_rate": 2.1499048064059224e-07,
      "loss": 0.9826,
      "num_input_tokens_seen": 21237560,
      "step": 38330
    },
    {
      "epoch": 672.5486725663717,
      "grad_norm": 7.375,
      "learning_rate": 2.1370764124502285e-07,
      "loss": 1.0132,
      "num_input_tokens_seen": 21240152,
      "step": 38335
    },
    {
      "epoch": 672.637168141593,
      "grad_norm": 8.75,
      "learning_rate": 2.1242862420721988e-07,
      "loss": 0.9753,
      "num_input_tokens_seen": 21242984,
      "step": 38340
    },
    {
      "epoch": 672.7256637168142,
      "grad_norm": 8.375,
      "learning_rate": 2.1115342972442276e-07,
      "loss": 0.9481,
      "num_input_tokens_seen": 21245720,
      "step": 38345
    },
    {
      "epoch": 672.8141592920354,
      "grad_norm": 8.625,
      "learning_rate": 2.0988205799328252e-07,
      "loss": 0.9343,
      "num_input_tokens_seen": 21249032,
      "step": 38350
    },
    {
      "epoch": 672.9026548672566,
      "grad_norm": 7.5625,
      "learning_rate": 2.0861450920986182e-07,
      "loss": 1.0523,
      "num_input_tokens_seen": 21252136,
      "step": 38355
    },
    {
      "epoch": 672.9911504424779,
      "grad_norm": 7.875,
      "learning_rate": 2.07350783569632e-07,
      "loss": 1.1536,
      "num_input_tokens_seen": 21254776,
      "step": 38360
    },
    {
      "epoch": 673.070796460177,
      "grad_norm": 10.1875,
      "learning_rate": 2.060908812674761e-07,
      "loss": 1.1527,
      "num_input_tokens_seen": 21257216,
      "step": 38365
    },
    {
      "epoch": 673.1592920353983,
      "grad_norm": 7.8125,
      "learning_rate": 2.0483480249768317e-07,
      "loss": 0.9894,
      "num_input_tokens_seen": 21259776,
      "step": 38370
    },
    {
      "epoch": 673.2477876106195,
      "grad_norm": 9.3125,
      "learning_rate": 2.035825474539621e-07,
      "loss": 1.027,
      "num_input_tokens_seen": 21262752,
      "step": 38375
    },
    {
      "epoch": 673.3362831858407,
      "grad_norm": 6.5625,
      "learning_rate": 2.0233411632942235e-07,
      "loss": 1.0381,
      "num_input_tokens_seen": 21265488,
      "step": 38380
    },
    {
      "epoch": 673.4247787610619,
      "grad_norm": 8.0625,
      "learning_rate": 2.0108950931658764e-07,
      "loss": 0.9759,
      "num_input_tokens_seen": 21268528,
      "step": 38385
    },
    {
      "epoch": 673.5132743362832,
      "grad_norm": 8.75,
      "learning_rate": 1.998487266073934e-07,
      "loss": 1.1238,
      "num_input_tokens_seen": 21271408,
      "step": 38390
    },
    {
      "epoch": 673.6017699115044,
      "grad_norm": 8.5625,
      "learning_rate": 1.986117683931865e-07,
      "loss": 1.1492,
      "num_input_tokens_seen": 21274336,
      "step": 38395
    },
    {
      "epoch": 673.6902654867257,
      "grad_norm": 8.125,
      "learning_rate": 1.9737863486471442e-07,
      "loss": 0.9958,
      "num_input_tokens_seen": 21277072,
      "step": 38400
    },
    {
      "epoch": 673.6902654867257,
      "eval_loss": 1.0310944318771362,
      "eval_runtime": 0.9942,
      "eval_samples_per_second": 25.145,
      "eval_steps_per_second": 13.075,
      "num_input_tokens_seen": 21277072,
      "step": 38400
    },
    {
      "epoch": 673.7787610619469,
      "grad_norm": 9.0625,
      "learning_rate": 1.9614932621215e-07,
      "loss": 1.0577,
      "num_input_tokens_seen": 21279888,
      "step": 38405
    },
    {
      "epoch": 673.8672566371681,
      "grad_norm": 9.75,
      "learning_rate": 1.9492384262506102e-07,
      "loss": 1.1962,
      "num_input_tokens_seen": 21282144,
      "step": 38410
    },
    {
      "epoch": 673.9557522123894,
      "grad_norm": 7.71875,
      "learning_rate": 1.9370218429243524e-07,
      "loss": 0.8943,
      "num_input_tokens_seen": 21285488,
      "step": 38415
    },
    {
      "epoch": 674.0353982300885,
      "grad_norm": 11.625,
      "learning_rate": 1.9248435140267197e-07,
      "loss": 0.9651,
      "num_input_tokens_seen": 21287864,
      "step": 38420
    },
    {
      "epoch": 674.1238938053098,
      "grad_norm": 10.0625,
      "learning_rate": 1.9127034414356814e-07,
      "loss": 1.0782,
      "num_input_tokens_seen": 21290456,
      "step": 38425
    },
    {
      "epoch": 674.212389380531,
      "grad_norm": 8.9375,
      "learning_rate": 1.9006016270234627e-07,
      "loss": 1.1239,
      "num_input_tokens_seen": 21293256,
      "step": 38430
    },
    {
      "epoch": 674.3008849557522,
      "grad_norm": 11.3125,
      "learning_rate": 1.888538072656293e-07,
      "loss": 1.172,
      "num_input_tokens_seen": 21296040,
      "step": 38435
    },
    {
      "epoch": 674.3893805309734,
      "grad_norm": 10.3125,
      "learning_rate": 1.8765127801944893e-07,
      "loss": 1.0602,
      "num_input_tokens_seen": 21299368,
      "step": 38440
    },
    {
      "epoch": 674.4778761061947,
      "grad_norm": 9.6875,
      "learning_rate": 1.8645257514925406e-07,
      "loss": 1.0151,
      "num_input_tokens_seen": 21302344,
      "step": 38445
    },
    {
      "epoch": 674.566371681416,
      "grad_norm": 6.34375,
      "learning_rate": 1.8525769883989685e-07,
      "loss": 1.2029,
      "num_input_tokens_seen": 21304648,
      "step": 38450
    },
    {
      "epoch": 674.6548672566372,
      "grad_norm": 8.25,
      "learning_rate": 1.8406664927564654e-07,
      "loss": 1.022,
      "num_input_tokens_seen": 21307688,
      "step": 38455
    },
    {
      "epoch": 674.7433628318585,
      "grad_norm": 7.28125,
      "learning_rate": 1.8287942664017566e-07,
      "loss": 1.0644,
      "num_input_tokens_seen": 21310584,
      "step": 38460
    },
    {
      "epoch": 674.8318584070796,
      "grad_norm": 7.65625,
      "learning_rate": 1.8169603111656552e-07,
      "loss": 1.1312,
      "num_input_tokens_seen": 21313000,
      "step": 38465
    },
    {
      "epoch": 674.9203539823009,
      "grad_norm": 7.8125,
      "learning_rate": 1.805164628873146e-07,
      "loss": 0.8793,
      "num_input_tokens_seen": 21315704,
      "step": 38470
    },
    {
      "epoch": 675.0,
      "grad_norm": 8.125,
      "learning_rate": 1.793407221343274e-07,
      "loss": 0.9537,
      "num_input_tokens_seen": 21318264,
      "step": 38475
    },
    {
      "epoch": 675.0884955752213,
      "grad_norm": 11.375,
      "learning_rate": 1.781688090389172e-07,
      "loss": 1.007,
      "num_input_tokens_seen": 21320760,
      "step": 38480
    },
    {
      "epoch": 675.1769911504425,
      "grad_norm": 7.84375,
      "learning_rate": 1.770007237818061e-07,
      "loss": 0.9795,
      "num_input_tokens_seen": 21323672,
      "step": 38485
    },
    {
      "epoch": 675.2654867256637,
      "grad_norm": 8.4375,
      "learning_rate": 1.7583646654313059e-07,
      "loss": 1.0941,
      "num_input_tokens_seen": 21326296,
      "step": 38490
    },
    {
      "epoch": 675.3539823008849,
      "grad_norm": 10.625,
      "learning_rate": 1.7467603750242757e-07,
      "loss": 0.9966,
      "num_input_tokens_seen": 21329464,
      "step": 38495
    },
    {
      "epoch": 675.4424778761062,
      "grad_norm": 7.40625,
      "learning_rate": 1.7351943683865944e-07,
      "loss": 1.012,
      "num_input_tokens_seen": 21331992,
      "step": 38500
    },
    {
      "epoch": 675.5309734513274,
      "grad_norm": 9.8125,
      "learning_rate": 1.723666647301808e-07,
      "loss": 0.9298,
      "num_input_tokens_seen": 21334936,
      "step": 38505
    },
    {
      "epoch": 675.6194690265487,
      "grad_norm": 7.65625,
      "learning_rate": 1.712177213547661e-07,
      "loss": 1.155,
      "num_input_tokens_seen": 21338024,
      "step": 38510
    },
    {
      "epoch": 675.70796460177,
      "grad_norm": 8.1875,
      "learning_rate": 1.7007260688959581e-07,
      "loss": 1.2194,
      "num_input_tokens_seen": 21340696,
      "step": 38515
    },
    {
      "epoch": 675.7964601769911,
      "grad_norm": 10.1875,
      "learning_rate": 1.68931321511262e-07,
      "loss": 1.226,
      "num_input_tokens_seen": 21343704,
      "step": 38520
    },
    {
      "epoch": 675.8849557522124,
      "grad_norm": 8.125,
      "learning_rate": 1.6779386539576835e-07,
      "loss": 0.846,
      "num_input_tokens_seen": 21346744,
      "step": 38525
    },
    {
      "epoch": 675.9734513274336,
      "grad_norm": 8.0,
      "learning_rate": 1.666602387185162e-07,
      "loss": 1.0018,
      "num_input_tokens_seen": 21349672,
      "step": 38530
    },
    {
      "epoch": 676.0530973451328,
      "grad_norm": 7.625,
      "learning_rate": 1.655304416543352e-07,
      "loss": 1.0556,
      "num_input_tokens_seen": 21351776,
      "step": 38535
    },
    {
      "epoch": 676.141592920354,
      "grad_norm": 8.5,
      "learning_rate": 1.6440447437744698e-07,
      "loss": 1.0486,
      "num_input_tokens_seen": 21354304,
      "step": 38540
    },
    {
      "epoch": 676.2300884955753,
      "grad_norm": 10.1875,
      "learning_rate": 1.6328233706149332e-07,
      "loss": 1.1924,
      "num_input_tokens_seen": 21357440,
      "step": 38545
    },
    {
      "epoch": 676.3185840707964,
      "grad_norm": 9.0,
      "learning_rate": 1.6216402987951906e-07,
      "loss": 1.2108,
      "num_input_tokens_seen": 21359744,
      "step": 38550
    },
    {
      "epoch": 676.4070796460177,
      "grad_norm": 8.3125,
      "learning_rate": 1.6104955300398627e-07,
      "loss": 1.0162,
      "num_input_tokens_seen": 21362528,
      "step": 38555
    },
    {
      "epoch": 676.4955752212389,
      "grad_norm": 14.3125,
      "learning_rate": 1.5993890660675748e-07,
      "loss": 1.422,
      "num_input_tokens_seen": 21365040,
      "step": 38560
    },
    {
      "epoch": 676.5840707964602,
      "grad_norm": 7.09375,
      "learning_rate": 1.5883209085910678e-07,
      "loss": 0.9425,
      "num_input_tokens_seen": 21368016,
      "step": 38565
    },
    {
      "epoch": 676.6725663716815,
      "grad_norm": 10.6875,
      "learning_rate": 1.5772910593172264e-07,
      "loss": 1.0468,
      "num_input_tokens_seen": 21370976,
      "step": 38570
    },
    {
      "epoch": 676.7610619469026,
      "grad_norm": 7.0625,
      "learning_rate": 1.5662995199469954e-07,
      "loss": 1.0743,
      "num_input_tokens_seen": 21374272,
      "step": 38575
    },
    {
      "epoch": 676.8495575221239,
      "grad_norm": 12.6875,
      "learning_rate": 1.5553462921753802e-07,
      "loss": 0.9048,
      "num_input_tokens_seen": 21377120,
      "step": 38580
    },
    {
      "epoch": 676.9380530973451,
      "grad_norm": 6.90625,
      "learning_rate": 1.544431377691502e-07,
      "loss": 0.7996,
      "num_input_tokens_seen": 21379904,
      "step": 38585
    },
    {
      "epoch": 677.0176991150443,
      "grad_norm": 9.0625,
      "learning_rate": 1.5335547781785975e-07,
      "loss": 0.8638,
      "num_input_tokens_seen": 21382496,
      "step": 38590
    },
    {
      "epoch": 677.1061946902655,
      "grad_norm": 8.125,
      "learning_rate": 1.5227164953139917e-07,
      "loss": 0.9582,
      "num_input_tokens_seen": 21385264,
      "step": 38595
    },
    {
      "epoch": 677.1946902654868,
      "grad_norm": 9.8125,
      "learning_rate": 1.511916530769042e-07,
      "loss": 1.1756,
      "num_input_tokens_seen": 21388128,
      "step": 38600
    },
    {
      "epoch": 677.1946902654868,
      "eval_loss": 1.0310944318771362,
      "eval_runtime": 1.0066,
      "eval_samples_per_second": 24.837,
      "eval_steps_per_second": 12.915,
      "num_input_tokens_seen": 21388128,
      "step": 38600
    },
    {
      "epoch": 677.2831858407079,
      "grad_norm": 5.625,
      "learning_rate": 1.5011548862092773e-07,
      "loss": 0.932,
      "num_input_tokens_seen": 21391040,
      "step": 38605
    },
    {
      "epoch": 677.3716814159292,
      "grad_norm": 6.6875,
      "learning_rate": 1.490431563294231e-07,
      "loss": 0.9792,
      "num_input_tokens_seen": 21393856,
      "step": 38610
    },
    {
      "epoch": 677.4601769911504,
      "grad_norm": 6.9375,
      "learning_rate": 1.4797465636776365e-07,
      "loss": 1.0202,
      "num_input_tokens_seen": 21396496,
      "step": 38615
    },
    {
      "epoch": 677.5486725663717,
      "grad_norm": 9.375,
      "learning_rate": 1.4690998890072027e-07,
      "loss": 1.0558,
      "num_input_tokens_seen": 21399360,
      "step": 38620
    },
    {
      "epoch": 677.637168141593,
      "grad_norm": 8.375,
      "learning_rate": 1.4584915409248112e-07,
      "loss": 0.9695,
      "num_input_tokens_seen": 21402032,
      "step": 38625
    },
    {
      "epoch": 677.7256637168142,
      "grad_norm": 7.0,
      "learning_rate": 1.4479215210663754e-07,
      "loss": 1.2219,
      "num_input_tokens_seen": 21404304,
      "step": 38630
    },
    {
      "epoch": 677.8141592920354,
      "grad_norm": 7.6875,
      "learning_rate": 1.4373898310619528e-07,
      "loss": 1.1199,
      "num_input_tokens_seen": 21407216,
      "step": 38635
    },
    {
      "epoch": 677.9026548672566,
      "grad_norm": 7.34375,
      "learning_rate": 1.4268964725356604e-07,
      "loss": 1.1885,
      "num_input_tokens_seen": 21410144,
      "step": 38640
    },
    {
      "epoch": 677.9911504424779,
      "grad_norm": 9.375,
      "learning_rate": 1.4164414471056764e-07,
      "loss": 0.9726,
      "num_input_tokens_seen": 21412912,
      "step": 38645
    },
    {
      "epoch": 678.070796460177,
      "grad_norm": 9.8125,
      "learning_rate": 1.4060247563843497e-07,
      "loss": 1.0354,
      "num_input_tokens_seen": 21415376,
      "step": 38650
    },
    {
      "epoch": 678.1592920353983,
      "grad_norm": 11.9375,
      "learning_rate": 1.3956464019780068e-07,
      "loss": 1.2392,
      "num_input_tokens_seen": 21418048,
      "step": 38655
    },
    {
      "epoch": 678.2477876106195,
      "grad_norm": 9.3125,
      "learning_rate": 1.385306385487145e-07,
      "loss": 0.9848,
      "num_input_tokens_seen": 21420656,
      "step": 38660
    },
    {
      "epoch": 678.3362831858407,
      "grad_norm": 7.4375,
      "learning_rate": 1.3750047085063222e-07,
      "loss": 0.9741,
      "num_input_tokens_seen": 21423776,
      "step": 38665
    },
    {
      "epoch": 678.4247787610619,
      "grad_norm": 8.5625,
      "learning_rate": 1.3647413726242119e-07,
      "loss": 1.08,
      "num_input_tokens_seen": 21426848,
      "step": 38670
    },
    {
      "epoch": 678.5132743362832,
      "grad_norm": 9.5,
      "learning_rate": 1.3545163794235205e-07,
      "loss": 1.1038,
      "num_input_tokens_seen": 21429472,
      "step": 38675
    },
    {
      "epoch": 678.6017699115044,
      "grad_norm": 6.65625,
      "learning_rate": 1.3443297304810698e-07,
      "loss": 0.9208,
      "num_input_tokens_seen": 21432528,
      "step": 38680
    },
    {
      "epoch": 678.6902654867257,
      "grad_norm": 7.90625,
      "learning_rate": 1.3341814273677977e-07,
      "loss": 0.9672,
      "num_input_tokens_seen": 21434944,
      "step": 38685
    },
    {
      "epoch": 678.7787610619469,
      "grad_norm": 9.5,
      "learning_rate": 1.324071471648647e-07,
      "loss": 1.057,
      "num_input_tokens_seen": 21437728,
      "step": 38690
    },
    {
      "epoch": 678.8672566371681,
      "grad_norm": 7.9375,
      "learning_rate": 1.3139998648827312e-07,
      "loss": 0.9396,
      "num_input_tokens_seen": 21441136,
      "step": 38695
    },
    {
      "epoch": 678.9557522123894,
      "grad_norm": 8.375,
      "learning_rate": 1.3039666086232526e-07,
      "loss": 1.054,
      "num_input_tokens_seen": 21444128,
      "step": 38700
    },
    {
      "epoch": 679.0353982300885,
      "grad_norm": 8.5625,
      "learning_rate": 1.2939717044174183e-07,
      "loss": 1.2888,
      "num_input_tokens_seen": 21446296,
      "step": 38705
    },
    {
      "epoch": 679.1238938053098,
      "grad_norm": 8.625,
      "learning_rate": 1.284015153806578e-07,
      "loss": 1.0037,
      "num_input_tokens_seen": 21449144,
      "step": 38710
    },
    {
      "epoch": 679.212389380531,
      "grad_norm": 10.0625,
      "learning_rate": 1.274096958326171e-07,
      "loss": 1.0383,
      "num_input_tokens_seen": 21452152,
      "step": 38715
    },
    {
      "epoch": 679.3008849557522,
      "grad_norm": 7.15625,
      "learning_rate": 1.2642171195056952e-07,
      "loss": 1.0538,
      "num_input_tokens_seen": 21454808,
      "step": 38720
    },
    {
      "epoch": 679.3893805309734,
      "grad_norm": 9.9375,
      "learning_rate": 1.2543756388687377e-07,
      "loss": 1.0718,
      "num_input_tokens_seen": 21457416,
      "step": 38725
    },
    {
      "epoch": 679.4778761061947,
      "grad_norm": 13.4375,
      "learning_rate": 1.2445725179330014e-07,
      "loss": 1.1045,
      "num_input_tokens_seen": 21460280,
      "step": 38730
    },
    {
      "epoch": 679.566371681416,
      "grad_norm": 10.0625,
      "learning_rate": 1.2348077582102212e-07,
      "loss": 1.1735,
      "num_input_tokens_seen": 21462696,
      "step": 38735
    },
    {
      "epoch": 679.6548672566372,
      "grad_norm": 8.1875,
      "learning_rate": 1.2250813612062762e-07,
      "loss": 0.8991,
      "num_input_tokens_seen": 21465528,
      "step": 38740
    },
    {
      "epoch": 679.7433628318585,
      "grad_norm": 10.125,
      "learning_rate": 1.215393328421105e-07,
      "loss": 1.0821,
      "num_input_tokens_seen": 21468312,
      "step": 38745
    },
    {
      "epoch": 679.8318584070796,
      "grad_norm": 8.5625,
      "learning_rate": 1.2057436613486796e-07,
      "loss": 1.1382,
      "num_input_tokens_seen": 21471096,
      "step": 38750
    },
    {
      "epoch": 679.9203539823009,
      "grad_norm": 6.09375,
      "learning_rate": 1.1961323614771424e-07,
      "loss": 1.0764,
      "num_input_tokens_seen": 21474392,
      "step": 38755
    },
    {
      "epoch": 680.0,
      "grad_norm": 9.875,
      "learning_rate": 1.1865594302886418e-07,
      "loss": 0.9231,
      "num_input_tokens_seen": 21477016,
      "step": 38760
    },
    {
      "epoch": 680.0884955752213,
      "grad_norm": 11.5625,
      "learning_rate": 1.1770248692594687e-07,
      "loss": 1.0636,
      "num_input_tokens_seen": 21479656,
      "step": 38765
    },
    {
      "epoch": 680.1769911504425,
      "grad_norm": 9.0,
      "learning_rate": 1.167528679859975e-07,
      "loss": 1.1738,
      "num_input_tokens_seen": 21482872,
      "step": 38770
    },
    {
      "epoch": 680.2654867256637,
      "grad_norm": 10.1875,
      "learning_rate": 1.1580708635545446e-07,
      "loss": 1.0615,
      "num_input_tokens_seen": 21485448,
      "step": 38775
    },
    {
      "epoch": 680.3539823008849,
      "grad_norm": 7.5625,
      "learning_rate": 1.1486514218017885e-07,
      "loss": 1.1121,
      "num_input_tokens_seen": 21488648,
      "step": 38780
    },
    {
      "epoch": 680.4424778761062,
      "grad_norm": 6.46875,
      "learning_rate": 1.1392703560542117e-07,
      "loss": 1.0698,
      "num_input_tokens_seen": 21491832,
      "step": 38785
    },
    {
      "epoch": 680.5309734513274,
      "grad_norm": 8.9375,
      "learning_rate": 1.129927667758518e-07,
      "loss": 1.1038,
      "num_input_tokens_seen": 21494120,
      "step": 38790
    },
    {
      "epoch": 680.6194690265487,
      "grad_norm": 7.46875,
      "learning_rate": 1.1206233583554992e-07,
      "loss": 1.2259,
      "num_input_tokens_seen": 21496808,
      "step": 38795
    },
    {
      "epoch": 680.70796460177,
      "grad_norm": 7.65625,
      "learning_rate": 1.1113574292799523e-07,
      "loss": 1.0102,
      "num_input_tokens_seen": 21499624,
      "step": 38800
    },
    {
      "epoch": 680.70796460177,
      "eval_loss": 1.0310944318771362,
      "eval_runtime": 0.9905,
      "eval_samples_per_second": 25.239,
      "eval_steps_per_second": 13.124,
      "num_input_tokens_seen": 21499624,
      "step": 38800
    },
    {
      "epoch": 680.7964601769911,
      "grad_norm": 8.5,
      "learning_rate": 1.1021298819608449e-07,
      "loss": 0.7588,
      "num_input_tokens_seen": 21502456,
      "step": 38805
    },
    {
      "epoch": 680.8849557522124,
      "grad_norm": 8.6875,
      "learning_rate": 1.0929407178211226e-07,
      "loss": 1.0605,
      "num_input_tokens_seen": 21505560,
      "step": 38810
    },
    {
      "epoch": 680.9734513274336,
      "grad_norm": 9.5625,
      "learning_rate": 1.0837899382779293e-07,
      "loss": 0.9657,
      "num_input_tokens_seen": 21508056,
      "step": 38815
    },
    {
      "epoch": 681.0530973451328,
      "grad_norm": 8.375,
      "learning_rate": 1.0746775447423862e-07,
      "loss": 0.981,
      "num_input_tokens_seen": 21510520,
      "step": 38820
    },
    {
      "epoch": 681.141592920354,
      "grad_norm": 8.6875,
      "learning_rate": 1.0656035386197583e-07,
      "loss": 1.1316,
      "num_input_tokens_seen": 21513912,
      "step": 38825
    },
    {
      "epoch": 681.2300884955753,
      "grad_norm": 7.90625,
      "learning_rate": 1.0565679213093982e-07,
      "loss": 0.9017,
      "num_input_tokens_seen": 21516840,
      "step": 38830
    },
    {
      "epoch": 681.3185840707964,
      "grad_norm": 6.78125,
      "learning_rate": 1.0475706942046638e-07,
      "loss": 0.9958,
      "num_input_tokens_seen": 21519352,
      "step": 38835
    },
    {
      "epoch": 681.4070796460177,
      "grad_norm": 7.75,
      "learning_rate": 1.0386118586930282e-07,
      "loss": 0.9126,
      "num_input_tokens_seen": 21522296,
      "step": 38840
    },
    {
      "epoch": 681.4955752212389,
      "grad_norm": 8.375,
      "learning_rate": 1.0296914161561367e-07,
      "loss": 1.1151,
      "num_input_tokens_seen": 21524920,
      "step": 38845
    },
    {
      "epoch": 681.5840707964602,
      "grad_norm": 10.5,
      "learning_rate": 1.0208093679695552e-07,
      "loss": 1.2745,
      "num_input_tokens_seen": 21527768,
      "step": 38850
    },
    {
      "epoch": 681.6725663716815,
      "grad_norm": 11.875,
      "learning_rate": 1.0119657155030493e-07,
      "loss": 1.0395,
      "num_input_tokens_seen": 21530504,
      "step": 38855
    },
    {
      "epoch": 681.7610619469026,
      "grad_norm": 9.75,
      "learning_rate": 1.003160460120417e-07,
      "loss": 1.2132,
      "num_input_tokens_seen": 21533000,
      "step": 38860
    },
    {
      "epoch": 681.8495575221239,
      "grad_norm": 7.34375,
      "learning_rate": 9.943936031795165e-08,
      "loss": 0.9092,
      "num_input_tokens_seen": 21536056,
      "step": 38865
    },
    {
      "epoch": 681.9380530973451,
      "grad_norm": 7.5625,
      "learning_rate": 9.856651460323219e-08,
      "loss": 0.9897,
      "num_input_tokens_seen": 21538616,
      "step": 38870
    },
    {
      "epoch": 682.0176991150443,
      "grad_norm": 10.875,
      "learning_rate": 9.769750900248953e-08,
      "loss": 0.9225,
      "num_input_tokens_seen": 21540760,
      "step": 38875
    },
    {
      "epoch": 682.1061946902655,
      "grad_norm": 8.75,
      "learning_rate": 9.683234364973038e-08,
      "loss": 1.133,
      "num_input_tokens_seen": 21544136,
      "step": 38880
    },
    {
      "epoch": 682.1946902654868,
      "grad_norm": 7.6875,
      "learning_rate": 9.597101867837854e-08,
      "loss": 0.9478,
      "num_input_tokens_seen": 21546920,
      "step": 38885
    },
    {
      "epoch": 682.2831858407079,
      "grad_norm": 7.96875,
      "learning_rate": 9.511353422125835e-08,
      "loss": 0.9099,
      "num_input_tokens_seen": 21550136,
      "step": 38890
    },
    {
      "epoch": 682.3716814159292,
      "grad_norm": 8.6875,
      "learning_rate": 9.42598904106029e-08,
      "loss": 1.1261,
      "num_input_tokens_seen": 21552856,
      "step": 38895
    },
    {
      "epoch": 682.4601769911504,
      "grad_norm": 7.375,
      "learning_rate": 9.341008737806245e-08,
      "loss": 0.9698,
      "num_input_tokens_seen": 21555832,
      "step": 38900
    },
    {
      "epoch": 682.5486725663717,
      "grad_norm": 7.40625,
      "learning_rate": 9.256412525467661e-08,
      "loss": 1.0989,
      "num_input_tokens_seen": 21559272,
      "step": 38905
    },
    {
      "epoch": 682.637168141593,
      "grad_norm": 10.625,
      "learning_rate": 9.172200417091326e-08,
      "loss": 1.0908,
      "num_input_tokens_seen": 21561832,
      "step": 38910
    },
    {
      "epoch": 682.7256637168142,
      "grad_norm": 10.75,
      "learning_rate": 9.088372425663239e-08,
      "loss": 1.0089,
      "num_input_tokens_seen": 21564488,
      "step": 38915
    },
    {
      "epoch": 682.8141592920354,
      "grad_norm": 8.5,
      "learning_rate": 9.004928564110837e-08,
      "loss": 1.1549,
      "num_input_tokens_seen": 21567000,
      "step": 38920
    },
    {
      "epoch": 682.9026548672566,
      "grad_norm": 8.0,
      "learning_rate": 8.92186884530244e-08,
      "loss": 1.0564,
      "num_input_tokens_seen": 21569416,
      "step": 38925
    },
    {
      "epoch": 682.9911504424779,
      "grad_norm": 9.1875,
      "learning_rate": 8.83919328204641e-08,
      "loss": 1.0436,
      "num_input_tokens_seen": 21572040,
      "step": 38930
    },
    {
      "epoch": 683.070796460177,
      "grad_norm": 5.78125,
      "learning_rate": 8.756901887093105e-08,
      "loss": 0.9267,
      "num_input_tokens_seen": 21574072,
      "step": 38935
    },
    {
      "epoch": 683.1592920353983,
      "grad_norm": 7.09375,
      "learning_rate": 8.674994673132098e-08,
      "loss": 1.0723,
      "num_input_tokens_seen": 21577368,
      "step": 38940
    },
    {
      "epoch": 683.2477876106195,
      "grad_norm": 6.59375,
      "learning_rate": 8.593471652794949e-08,
      "loss": 0.9612,
      "num_input_tokens_seen": 21580456,
      "step": 38945
    },
    {
      "epoch": 683.3362831858407,
      "grad_norm": 7.28125,
      "learning_rate": 8.512332838653548e-08,
      "loss": 1.0531,
      "num_input_tokens_seen": 21583464,
      "step": 38950
    },
    {
      "epoch": 683.4247787610619,
      "grad_norm": 7.46875,
      "learning_rate": 8.431578243220106e-08,
      "loss": 1.112,
      "num_input_tokens_seen": 21586552,
      "step": 38955
    },
    {
      "epoch": 683.5132743362832,
      "grad_norm": 7.625,
      "learning_rate": 8.351207878948552e-08,
      "loss": 0.9837,
      "num_input_tokens_seen": 21589624,
      "step": 38960
    },
    {
      "epoch": 683.6017699115044,
      "grad_norm": 7.59375,
      "learning_rate": 8.271221758232583e-08,
      "loss": 1.0194,
      "num_input_tokens_seen": 21592072,
      "step": 38965
    },
    {
      "epoch": 683.6902654867257,
      "grad_norm": 7.9375,
      "learning_rate": 8.191619893407332e-08,
      "loss": 1.205,
      "num_input_tokens_seen": 21594568,
      "step": 38970
    },
    {
      "epoch": 683.7787610619469,
      "grad_norm": 7.59375,
      "learning_rate": 8.112402296748534e-08,
      "loss": 1.0843,
      "num_input_tokens_seen": 21597000,
      "step": 38975
    },
    {
      "epoch": 683.8672566371681,
      "grad_norm": 10.8125,
      "learning_rate": 8.033568980471973e-08,
      "loss": 1.0835,
      "num_input_tokens_seen": 21599560,
      "step": 38980
    },
    {
      "epoch": 683.9557522123894,
      "grad_norm": 8.8125,
      "learning_rate": 7.955119956735146e-08,
      "loss": 1.0404,
      "num_input_tokens_seen": 21602536,
      "step": 38985
    },
    {
      "epoch": 684.0353982300885,
      "grad_norm": 11.125,
      "learning_rate": 7.877055237636155e-08,
      "loss": 1.0964,
      "num_input_tokens_seen": 21604888,
      "step": 38990
    },
    {
      "epoch": 684.1238938053098,
      "grad_norm": 9.8125,
      "learning_rate": 7.79937483521287e-08,
      "loss": 1.0716,
      "num_input_tokens_seen": 21607640,
      "step": 38995
    },
    {
      "epoch": 684.212389380531,
      "grad_norm": 7.625,
      "learning_rate": 7.722078761444873e-08,
      "loss": 0.9521,
      "num_input_tokens_seen": 21611240,
      "step": 39000
    },
    {
      "epoch": 684.212389380531,
      "eval_loss": 1.0310944318771362,
      "eval_runtime": 0.9955,
      "eval_samples_per_second": 25.113,
      "eval_steps_per_second": 13.059,
      "num_input_tokens_seen": 21611240,
      "step": 39000
    },
    {
      "epoch": 684.3008849557522,
      "grad_norm": 7.15625,
      "learning_rate": 7.645167028252631e-08,
      "loss": 1.1218,
      "num_input_tokens_seen": 21613592,
      "step": 39005
    },
    {
      "epoch": 684.3893805309734,
      "grad_norm": 8.625,
      "learning_rate": 7.568639647496379e-08,
      "loss": 1.0668,
      "num_input_tokens_seen": 21616856,
      "step": 39010
    },
    {
      "epoch": 684.4778761061947,
      "grad_norm": 9.125,
      "learning_rate": 7.492496630977508e-08,
      "loss": 1.0677,
      "num_input_tokens_seen": 21619944,
      "step": 39015
    },
    {
      "epoch": 684.566371681416,
      "grad_norm": 11.9375,
      "learning_rate": 7.416737990438571e-08,
      "loss": 1.0865,
      "num_input_tokens_seen": 21622392,
      "step": 39020
    },
    {
      "epoch": 684.6548672566372,
      "grad_norm": 9.375,
      "learning_rate": 7.341363737562445e-08,
      "loss": 1.1346,
      "num_input_tokens_seen": 21624792,
      "step": 39025
    },
    {
      "epoch": 684.7433628318585,
      "grad_norm": 8.375,
      "learning_rate": 7.266373883972887e-08,
      "loss": 1.1483,
      "num_input_tokens_seen": 21627160,
      "step": 39030
    },
    {
      "epoch": 684.8318584070796,
      "grad_norm": 8.6875,
      "learning_rate": 7.191768441233981e-08,
      "loss": 1.0656,
      "num_input_tokens_seen": 21629944,
      "step": 39035
    },
    {
      "epoch": 684.9203539823009,
      "grad_norm": 8.5,
      "learning_rate": 7.11754742085069e-08,
      "loss": 0.8615,
      "num_input_tokens_seen": 21632664,
      "step": 39040
    },
    {
      "epoch": 685.0,
      "grad_norm": 16.625,
      "learning_rate": 7.043710834269413e-08,
      "loss": 1.0084,
      "num_input_tokens_seen": 21634992,
      "step": 39045
    },
    {
      "epoch": 685.0884955752213,
      "grad_norm": 9.0625,
      "learning_rate": 6.970258692876319e-08,
      "loss": 1.0891,
      "num_input_tokens_seen": 21637792,
      "step": 39050
    },
    {
      "epoch": 685.1769911504425,
      "grad_norm": 7.5625,
      "learning_rate": 6.897191007998738e-08,
      "loss": 0.9575,
      "num_input_tokens_seen": 21640608,
      "step": 39055
    },
    {
      "epoch": 685.2654867256637,
      "grad_norm": 8.9375,
      "learning_rate": 6.824507790904599e-08,
      "loss": 1.1092,
      "num_input_tokens_seen": 21643312,
      "step": 39060
    },
    {
      "epoch": 685.3539823008849,
      "grad_norm": 11.0625,
      "learning_rate": 6.752209052802439e-08,
      "loss": 1.1366,
      "num_input_tokens_seen": 21645888,
      "step": 39065
    },
    {
      "epoch": 685.4424778761062,
      "grad_norm": 7.65625,
      "learning_rate": 6.680294804841946e-08,
      "loss": 0.8872,
      "num_input_tokens_seen": 21648512,
      "step": 39070
    },
    {
      "epoch": 685.5309734513274,
      "grad_norm": 7.6875,
      "learning_rate": 6.608765058112865e-08,
      "loss": 1.0305,
      "num_input_tokens_seen": 21651104,
      "step": 39075
    },
    {
      "epoch": 685.6194690265487,
      "grad_norm": 8.4375,
      "learning_rate": 6.537619823646368e-08,
      "loss": 0.9572,
      "num_input_tokens_seen": 21654192,
      "step": 39080
    },
    {
      "epoch": 685.70796460177,
      "grad_norm": 8.875,
      "learning_rate": 6.466859112413404e-08,
      "loss": 0.984,
      "num_input_tokens_seen": 21657200,
      "step": 39085
    },
    {
      "epoch": 685.7964601769911,
      "grad_norm": 8.3125,
      "learning_rate": 6.39648293532663e-08,
      "loss": 1.3189,
      "num_input_tokens_seen": 21660304,
      "step": 39090
    },
    {
      "epoch": 685.8849557522124,
      "grad_norm": 6.84375,
      "learning_rate": 6.32649130323848e-08,
      "loss": 1.1283,
      "num_input_tokens_seen": 21663200,
      "step": 39095
    },
    {
      "epoch": 685.9734513274336,
      "grad_norm": 9.1875,
      "learning_rate": 6.256884226943094e-08,
      "loss": 1.0265,
      "num_input_tokens_seen": 21665872,
      "step": 39100
    },
    {
      "epoch": 686.0530973451328,
      "grad_norm": 8.9375,
      "learning_rate": 6.187661717174386e-08,
      "loss": 0.973,
      "num_input_tokens_seen": 21668064,
      "step": 39105
    },
    {
      "epoch": 686.141592920354,
      "grad_norm": 11.5625,
      "learning_rate": 6.118823784607708e-08,
      "loss": 0.9828,
      "num_input_tokens_seen": 21670736,
      "step": 39110
    },
    {
      "epoch": 686.2300884955753,
      "grad_norm": 9.9375,
      "learning_rate": 6.050370439858178e-08,
      "loss": 1.0218,
      "num_input_tokens_seen": 21673632,
      "step": 39115
    },
    {
      "epoch": 686.3185840707964,
      "grad_norm": 7.71875,
      "learning_rate": 5.98230169348235e-08,
      "loss": 1.1428,
      "num_input_tokens_seen": 21676144,
      "step": 39120
    },
    {
      "epoch": 686.4070796460177,
      "grad_norm": 14.1875,
      "learning_rate": 5.914617555977664e-08,
      "loss": 1.252,
      "num_input_tokens_seen": 21678624,
      "step": 39125
    },
    {
      "epoch": 686.4955752212389,
      "grad_norm": 7.90625,
      "learning_rate": 5.8473180377816017e-08,
      "loss": 1.0784,
      "num_input_tokens_seen": 21681056,
      "step": 39130
    },
    {
      "epoch": 686.5840707964602,
      "grad_norm": 8.3125,
      "learning_rate": 5.780403149272251e-08,
      "loss": 1.0092,
      "num_input_tokens_seen": 21684032,
      "step": 39135
    },
    {
      "epoch": 686.6725663716815,
      "grad_norm": 8.5625,
      "learning_rate": 5.7138729007694126e-08,
      "loss": 1.0513,
      "num_input_tokens_seen": 21686848,
      "step": 39140
    },
    {
      "epoch": 686.7610619469026,
      "grad_norm": 7.34375,
      "learning_rate": 5.64772730253238e-08,
      "loss": 1.1486,
      "num_input_tokens_seen": 21690224,
      "step": 39145
    },
    {
      "epoch": 686.8495575221239,
      "grad_norm": 7.46875,
      "learning_rate": 5.5819663647618814e-08,
      "loss": 1.0671,
      "num_input_tokens_seen": 21693008,
      "step": 39150
    },
    {
      "epoch": 686.9380530973451,
      "grad_norm": 6.71875,
      "learning_rate": 5.5165900975989723e-08,
      "loss": 0.7933,
      "num_input_tokens_seen": 21695872,
      "step": 39155
    },
    {
      "epoch": 687.0176991150443,
      "grad_norm": 9.4375,
      "learning_rate": 5.451598511125311e-08,
      "loss": 1.0908,
      "num_input_tokens_seen": 21698400,
      "step": 39160
    },
    {
      "epoch": 687.1061946902655,
      "grad_norm": 8.75,
      "learning_rate": 5.3869916153637124e-08,
      "loss": 0.9567,
      "num_input_tokens_seen": 21701584,
      "step": 39165
    },
    {
      "epoch": 687.1946902654868,
      "grad_norm": 8.1875,
      "learning_rate": 5.322769420277318e-08,
      "loss": 1.0495,
      "num_input_tokens_seen": 21704384,
      "step": 39170
    },
    {
      "epoch": 687.2831858407079,
      "grad_norm": 8.5625,
      "learning_rate": 5.258931935769873e-08,
      "loss": 1.0965,
      "num_input_tokens_seen": 21707472,
      "step": 39175
    },
    {
      "epoch": 687.3716814159292,
      "grad_norm": 7.71875,
      "learning_rate": 5.19547917168628e-08,
      "loss": 0.9866,
      "num_input_tokens_seen": 21710176,
      "step": 39180
    },
    {
      "epoch": 687.4601769911504,
      "grad_norm": 8.5625,
      "learning_rate": 5.13241113781121e-08,
      "loss": 1.156,
      "num_input_tokens_seen": 21712720,
      "step": 39185
    },
    {
      "epoch": 687.5486725663717,
      "grad_norm": 9.875,
      "learning_rate": 5.0697278438707755e-08,
      "loss": 1.0427,
      "num_input_tokens_seen": 21715792,
      "step": 39190
    },
    {
      "epoch": 687.637168141593,
      "grad_norm": 7.125,
      "learning_rate": 5.0074292995316854e-08,
      "loss": 0.9107,
      "num_input_tokens_seen": 21718704,
      "step": 39195
    },
    {
      "epoch": 687.7256637168142,
      "grad_norm": 8.125,
      "learning_rate": 4.945515514400978e-08,
      "loss": 1.0948,
      "num_input_tokens_seen": 21721232,
      "step": 39200
    },
    {
      "epoch": 687.7256637168142,
      "eval_loss": 1.0310944318771362,
      "eval_runtime": 0.9952,
      "eval_samples_per_second": 25.12,
      "eval_steps_per_second": 13.062,
      "num_input_tokens_seen": 21721232,
      "step": 39200
    },
    {
      "epoch": 687.8141592920354,
      "grad_norm": 7.9375,
      "learning_rate": 4.883986498026571e-08,
      "loss": 1.0286,
      "num_input_tokens_seen": 21723888,
      "step": 39205
    },
    {
      "epoch": 687.9026548672566,
      "grad_norm": 8.4375,
      "learning_rate": 4.822842259896987e-08,
      "loss": 1.2327,
      "num_input_tokens_seen": 21726752,
      "step": 39210
    },
    {
      "epoch": 687.9911504424779,
      "grad_norm": 7.40625,
      "learning_rate": 4.762082809441626e-08,
      "loss": 0.951,
      "num_input_tokens_seen": 21729456,
      "step": 39215
    },
    {
      "epoch": 688.070796460177,
      "grad_norm": 8.75,
      "learning_rate": 4.7017081560302156e-08,
      "loss": 1.1646,
      "num_input_tokens_seen": 21731448,
      "step": 39220
    },
    {
      "epoch": 688.1592920353983,
      "grad_norm": 9.125,
      "learning_rate": 4.6417183089730866e-08,
      "loss": 0.9631,
      "num_input_tokens_seen": 21734088,
      "step": 39225
    },
    {
      "epoch": 688.2477876106195,
      "grad_norm": 8.9375,
      "learning_rate": 4.5821132775217265e-08,
      "loss": 1.1365,
      "num_input_tokens_seen": 21736840,
      "step": 39230
    },
    {
      "epoch": 688.3362831858407,
      "grad_norm": 7.53125,
      "learning_rate": 4.5228930708679504e-08,
      "loss": 1.0214,
      "num_input_tokens_seen": 21739464,
      "step": 39235
    },
    {
      "epoch": 688.4247787610619,
      "grad_norm": 9.0,
      "learning_rate": 4.464057698144175e-08,
      "loss": 0.9744,
      "num_input_tokens_seen": 21742456,
      "step": 39240
    },
    {
      "epoch": 688.5132743362832,
      "grad_norm": 8.9375,
      "learning_rate": 4.4056071684236974e-08,
      "loss": 1.0903,
      "num_input_tokens_seen": 21745416,
      "step": 39245
    },
    {
      "epoch": 688.6017699115044,
      "grad_norm": 7.21875,
      "learning_rate": 4.347541490719864e-08,
      "loss": 1.0335,
      "num_input_tokens_seen": 21748344,
      "step": 39250
    },
    {
      "epoch": 688.6902654867257,
      "grad_norm": 8.375,
      "learning_rate": 4.2898606739877336e-08,
      "loss": 0.8677,
      "num_input_tokens_seen": 21751496,
      "step": 39255
    },
    {
      "epoch": 688.7787610619469,
      "grad_norm": 8.8125,
      "learning_rate": 4.232564727122135e-08,
      "loss": 0.9241,
      "num_input_tokens_seen": 21754312,
      "step": 39260
    },
    {
      "epoch": 688.8672566371681,
      "grad_norm": 10.6875,
      "learning_rate": 4.1756536589585004e-08,
      "loss": 1.2121,
      "num_input_tokens_seen": 21757288,
      "step": 39265
    },
    {
      "epoch": 688.9557522123894,
      "grad_norm": 7.84375,
      "learning_rate": 4.119127478273976e-08,
      "loss": 1.1512,
      "num_input_tokens_seen": 21760088,
      "step": 39270
    },
    {
      "epoch": 689.0353982300885,
      "grad_norm": 7.40625,
      "learning_rate": 4.062986193784923e-08,
      "loss": 0.9601,
      "num_input_tokens_seen": 21762320,
      "step": 39275
    },
    {
      "epoch": 689.1238938053098,
      "grad_norm": 9.625,
      "learning_rate": 4.007229814149416e-08,
      "loss": 1.2365,
      "num_input_tokens_seen": 21764800,
      "step": 39280
    },
    {
      "epoch": 689.212389380531,
      "grad_norm": 7.34375,
      "learning_rate": 3.951858347965576e-08,
      "loss": 0.9587,
      "num_input_tokens_seen": 21767680,
      "step": 39285
    },
    {
      "epoch": 689.3008849557522,
      "grad_norm": 7.4375,
      "learning_rate": 3.896871803772684e-08,
      "loss": 0.9741,
      "num_input_tokens_seen": 21770256,
      "step": 39290
    },
    {
      "epoch": 689.3893805309734,
      "grad_norm": 8.5,
      "learning_rate": 3.842270190050068e-08,
      "loss": 1.2856,
      "num_input_tokens_seen": 21772976,
      "step": 39295
    },
    {
      "epoch": 689.4778761061947,
      "grad_norm": 13.125,
      "learning_rate": 3.7880535152179376e-08,
      "loss": 1.0105,
      "num_input_tokens_seen": 21775920,
      "step": 39300
    },
    {
      "epoch": 689.566371681416,
      "grad_norm": 8.25,
      "learning_rate": 3.734221787637382e-08,
      "loss": 1.1026,
      "num_input_tokens_seen": 21778464,
      "step": 39305
    },
    {
      "epoch": 689.6548672566372,
      "grad_norm": 10.6875,
      "learning_rate": 3.680775015609817e-08,
      "loss": 0.9251,
      "num_input_tokens_seen": 21781408,
      "step": 39310
    },
    {
      "epoch": 689.7433628318585,
      "grad_norm": 10.6875,
      "learning_rate": 3.627713207377537e-08,
      "loss": 0.9412,
      "num_input_tokens_seen": 21784432,
      "step": 39315
    },
    {
      "epoch": 689.8318584070796,
      "grad_norm": 7.28125,
      "learning_rate": 3.575036371123164e-08,
      "loss": 1.1669,
      "num_input_tokens_seen": 21787024,
      "step": 39320
    },
    {
      "epoch": 689.9203539823009,
      "grad_norm": 7.375,
      "learning_rate": 3.5227445149704776e-08,
      "loss": 1.0568,
      "num_input_tokens_seen": 21790208,
      "step": 39325
    },
    {
      "epoch": 690.0,
      "grad_norm": 8.25,
      "learning_rate": 3.470837646983027e-08,
      "loss": 0.9666,
      "num_input_tokens_seen": 21792656,
      "step": 39330
    },
    {
      "epoch": 690.0884955752213,
      "grad_norm": 8.0,
      "learning_rate": 3.419315775165799e-08,
      "loss": 1.0687,
      "num_input_tokens_seen": 21795664,
      "step": 39335
    },
    {
      "epoch": 690.1769911504425,
      "grad_norm": 8.875,
      "learning_rate": 3.368178907464103e-08,
      "loss": 1.103,
      "num_input_tokens_seen": 21798064,
      "step": 39340
    },
    {
      "epoch": 690.2654867256637,
      "grad_norm": 8.5625,
      "learning_rate": 3.317427051763855e-08,
      "loss": 0.8978,
      "num_input_tokens_seen": 21801120,
      "step": 39345
    },
    {
      "epoch": 690.3539823008849,
      "grad_norm": 6.625,
      "learning_rate": 3.267060215891571e-08,
      "loss": 1.1185,
      "num_input_tokens_seen": 21803776,
      "step": 39350
    },
    {
      "epoch": 690.4424778761062,
      "grad_norm": 7.5625,
      "learning_rate": 3.217078407614649e-08,
      "loss": 1.1325,
      "num_input_tokens_seen": 21806656,
      "step": 39355
    },
    {
      "epoch": 690.5309734513274,
      "grad_norm": 10.75,
      "learning_rate": 3.1674816346405345e-08,
      "loss": 1.1411,
      "num_input_tokens_seen": 21810192,
      "step": 39360
    },
    {
      "epoch": 690.6194690265487,
      "grad_norm": 8.4375,
      "learning_rate": 3.11826990461811e-08,
      "loss": 0.9306,
      "num_input_tokens_seen": 21813136,
      "step": 39365
    },
    {
      "epoch": 690.70796460177,
      "grad_norm": 9.8125,
      "learning_rate": 3.069443225136304e-08,
      "loss": 0.9273,
      "num_input_tokens_seen": 21815728,
      "step": 39370
    },
    {
      "epoch": 690.7964601769911,
      "grad_norm": 12.1875,
      "learning_rate": 3.021001603724372e-08,
      "loss": 1.1628,
      "num_input_tokens_seen": 21818576,
      "step": 39375
    },
    {
      "epoch": 690.8849557522124,
      "grad_norm": 9.75,
      "learning_rate": 2.9729450478532818e-08,
      "loss": 1.079,
      "num_input_tokens_seen": 21821440,
      "step": 39380
    },
    {
      "epoch": 690.9734513274336,
      "grad_norm": 7.71875,
      "learning_rate": 2.9252735649337726e-08,
      "loss": 1.0178,
      "num_input_tokens_seen": 21824032,
      "step": 39385
    },
    {
      "epoch": 691.0530973451328,
      "grad_norm": 9.9375,
      "learning_rate": 2.8779871623171863e-08,
      "loss": 1.1361,
      "num_input_tokens_seen": 21826144,
      "step": 39390
    },
    {
      "epoch": 691.141592920354,
      "grad_norm": 8.6875,
      "learning_rate": 2.8310858472957448e-08,
      "loss": 1.0963,
      "num_input_tokens_seen": 21829152,
      "step": 39395
    },
    {
      "epoch": 691.2300884955753,
      "grad_norm": 7.03125,
      "learning_rate": 2.784569627101996e-08,
      "loss": 0.9525,
      "num_input_tokens_seen": 21832720,
      "step": 39400
    },
    {
      "epoch": 691.2300884955753,
      "eval_loss": 1.0310944318771362,
      "eval_runtime": 1.0118,
      "eval_samples_per_second": 24.707,
      "eval_steps_per_second": 12.848,
      "num_input_tokens_seen": 21832720,
      "step": 39400
    },
    {
      "epoch": 691.3185840707964,
      "grad_norm": 11.5,
      "learning_rate": 2.738438508909924e-08,
      "loss": 1.0589,
      "num_input_tokens_seen": 21835376,
      "step": 39405
    },
    {
      "epoch": 691.4070796460177,
      "grad_norm": 5.84375,
      "learning_rate": 2.692692499833005e-08,
      "loss": 0.9837,
      "num_input_tokens_seen": 21838080,
      "step": 39410
    },
    {
      "epoch": 691.4955752212389,
      "grad_norm": 9.125,
      "learning_rate": 2.647331606926151e-08,
      "loss": 1.0101,
      "num_input_tokens_seen": 21840432,
      "step": 39415
    },
    {
      "epoch": 691.5840707964602,
      "grad_norm": 6.65625,
      "learning_rate": 2.6023558371843225e-08,
      "loss": 1.1045,
      "num_input_tokens_seen": 21843232,
      "step": 39420
    },
    {
      "epoch": 691.6725663716815,
      "grad_norm": 7.1875,
      "learning_rate": 2.557765197543638e-08,
      "loss": 0.8995,
      "num_input_tokens_seen": 21845952,
      "step": 39425
    },
    {
      "epoch": 691.7610619469026,
      "grad_norm": 8.875,
      "learning_rate": 2.513559694880263e-08,
      "loss": 1.0491,
      "num_input_tokens_seen": 21848752,
      "step": 39430
    },
    {
      "epoch": 691.8495575221239,
      "grad_norm": 8.0,
      "learning_rate": 2.469739336011523e-08,
      "loss": 1.0414,
      "num_input_tokens_seen": 21851536,
      "step": 39435
    },
    {
      "epoch": 691.9380530973451,
      "grad_norm": 11.0625,
      "learning_rate": 2.4263041276947894e-08,
      "loss": 1.2273,
      "num_input_tokens_seen": 21854400,
      "step": 39440
    },
    {
      "epoch": 692.0176991150443,
      "grad_norm": 7.03125,
      "learning_rate": 2.3832540766283164e-08,
      "loss": 1.0216,
      "num_input_tokens_seen": 21856552,
      "step": 39445
    },
    {
      "epoch": 692.1061946902655,
      "grad_norm": 7.75,
      "learning_rate": 2.3405891894512366e-08,
      "loss": 1.0417,
      "num_input_tokens_seen": 21859080,
      "step": 39450
    },
    {
      "epoch": 692.1946902654868,
      "grad_norm": 8.75,
      "learning_rate": 2.29830947274301e-08,
      "loss": 1.1046,
      "num_input_tokens_seen": 21861624,
      "step": 39455
    },
    {
      "epoch": 692.2831858407079,
      "grad_norm": 8.625,
      "learning_rate": 2.2564149330231432e-08,
      "loss": 0.9412,
      "num_input_tokens_seen": 21864472,
      "step": 39460
    },
    {
      "epoch": 692.3716814159292,
      "grad_norm": 10.125,
      "learning_rate": 2.2149055767528572e-08,
      "loss": 1.1646,
      "num_input_tokens_seen": 21867096,
      "step": 39465
    },
    {
      "epoch": 692.4601769911504,
      "grad_norm": 9.0625,
      "learning_rate": 2.1737814103334197e-08,
      "loss": 1.0476,
      "num_input_tokens_seen": 21870040,
      "step": 39470
    },
    {
      "epoch": 692.5486725663717,
      "grad_norm": 7.5625,
      "learning_rate": 2.1330424401064253e-08,
      "loss": 0.9998,
      "num_input_tokens_seen": 21872584,
      "step": 39475
    },
    {
      "epoch": 692.637168141593,
      "grad_norm": 6.0625,
      "learning_rate": 2.092688672354348e-08,
      "loss": 1.0811,
      "num_input_tokens_seen": 21875800,
      "step": 39480
    },
    {
      "epoch": 692.7256637168142,
      "grad_norm": 7.09375,
      "learning_rate": 2.0527201133005435e-08,
      "loss": 0.7682,
      "num_input_tokens_seen": 21879288,
      "step": 39485
    },
    {
      "epoch": 692.8141592920354,
      "grad_norm": 8.5625,
      "learning_rate": 2.0131367691084148e-08,
      "loss": 1.2854,
      "num_input_tokens_seen": 21882152,
      "step": 39490
    },
    {
      "epoch": 692.9026548672566,
      "grad_norm": 7.6875,
      "learning_rate": 1.9739386458819675e-08,
      "loss": 1.1301,
      "num_input_tokens_seen": 21884808,
      "step": 39495
    },
    {
      "epoch": 692.9911504424779,
      "grad_norm": 7.71875,
      "learning_rate": 1.9351257496666442e-08,
      "loss": 1.0792,
      "num_input_tokens_seen": 21887400,
      "step": 39500
    },
    {
      "epoch": 693.070796460177,
      "grad_norm": 6.34375,
      "learning_rate": 1.896698086447657e-08,
      "loss": 1.0223,
      "num_input_tokens_seen": 21889664,
      "step": 39505
    },
    {
      "epoch": 693.1592920353983,
      "grad_norm": 8.5625,
      "learning_rate": 1.8586556621505436e-08,
      "loss": 0.9148,
      "num_input_tokens_seen": 21892320,
      "step": 39510
    },
    {
      "epoch": 693.2477876106195,
      "grad_norm": 6.1875,
      "learning_rate": 1.820998482642833e-08,
      "loss": 0.9426,
      "num_input_tokens_seen": 21895552,
      "step": 39515
    },
    {
      "epoch": 693.3362831858407,
      "grad_norm": 11.6875,
      "learning_rate": 1.7837265537309912e-08,
      "loss": 1.003,
      "num_input_tokens_seen": 21898880,
      "step": 39520
    },
    {
      "epoch": 693.4247787610619,
      "grad_norm": 13.125,
      "learning_rate": 1.7468398811629206e-08,
      "loss": 1.2037,
      "num_input_tokens_seen": 21901712,
      "step": 39525
    },
    {
      "epoch": 693.5132743362832,
      "grad_norm": 7.625,
      "learning_rate": 1.710338470627404e-08,
      "loss": 1.2918,
      "num_input_tokens_seen": 21904128,
      "step": 39530
    },
    {
      "epoch": 693.6017699115044,
      "grad_norm": 8.3125,
      "learning_rate": 1.6742223277529945e-08,
      "loss": 1.0822,
      "num_input_tokens_seen": 21906672,
      "step": 39535
    },
    {
      "epoch": 693.6902654867257,
      "grad_norm": 8.375,
      "learning_rate": 1.6384914581094036e-08,
      "loss": 0.9755,
      "num_input_tokens_seen": 21909696,
      "step": 39540
    },
    {
      "epoch": 693.7787610619469,
      "grad_norm": 8.5,
      "learning_rate": 1.6031458672069455e-08,
      "loss": 1.0652,
      "num_input_tokens_seen": 21912256,
      "step": 39545
    },
    {
      "epoch": 693.8672566371681,
      "grad_norm": 8.3125,
      "learning_rate": 1.5681855604962602e-08,
      "loss": 1.027,
      "num_input_tokens_seen": 21915120,
      "step": 39550
    },
    {
      "epoch": 693.9557522123894,
      "grad_norm": 8.25,
      "learning_rate": 1.5336105433683135e-08,
      "loss": 1.1349,
      "num_input_tokens_seen": 21917792,
      "step": 39555
    },
    {
      "epoch": 694.0353982300885,
      "grad_norm": 14.125,
      "learning_rate": 1.499420821155506e-08,
      "loss": 1.1544,
      "num_input_tokens_seen": 21920168,
      "step": 39560
    },
    {
      "epoch": 694.1238938053098,
      "grad_norm": 8.625,
      "learning_rate": 1.4656163991302874e-08,
      "loss": 1.1088,
      "num_input_tokens_seen": 21922824,
      "step": 39565
    },
    {
      "epoch": 694.212389380531,
      "grad_norm": 9.875,
      "learning_rate": 1.4321972825051544e-08,
      "loss": 0.9453,
      "num_input_tokens_seen": 21926024,
      "step": 39570
    },
    {
      "epoch": 694.3008849557522,
      "grad_norm": 9.0625,
      "learning_rate": 1.3991634764345951e-08,
      "loss": 1.0721,
      "num_input_tokens_seen": 21928328,
      "step": 39575
    },
    {
      "epoch": 694.3893805309734,
      "grad_norm": 8.4375,
      "learning_rate": 1.3665149860120352e-08,
      "loss": 1.1264,
      "num_input_tokens_seen": 21931176,
      "step": 39580
    },
    {
      "epoch": 694.4778761061947,
      "grad_norm": 6.71875,
      "learning_rate": 1.3342518162728912e-08,
      "loss": 1.1986,
      "num_input_tokens_seen": 21933608,
      "step": 39585
    },
    {
      "epoch": 694.566371681416,
      "grad_norm": 7.625,
      "learning_rate": 1.30237397219235e-08,
      "loss": 0.8844,
      "num_input_tokens_seen": 21936632,
      "step": 39590
    },
    {
      "epoch": 694.6548672566372,
      "grad_norm": 8.25,
      "learning_rate": 1.2708814586862016e-08,
      "loss": 1.1048,
      "num_input_tokens_seen": 21939704,
      "step": 39595
    },
    {
      "epoch": 694.7433628318585,
      "grad_norm": 7.71875,
      "learning_rate": 1.2397742806111168e-08,
      "loss": 1.0582,
      "num_input_tokens_seen": 21942280,
      "step": 39600
    },
    {
      "epoch": 694.7433628318585,
      "eval_loss": 1.0310944318771362,
      "eval_runtime": 1.0172,
      "eval_samples_per_second": 24.577,
      "eval_steps_per_second": 12.78,
      "num_input_tokens_seen": 21942280,
      "step": 39600
    },
    {
      "epoch": 694.8318584070796,
      "grad_norm": 8.5,
      "learning_rate": 1.209052442764369e-08,
      "loss": 0.9865,
      "num_input_tokens_seen": 21945032,
      "step": 39605
    },
    {
      "epoch": 694.9203539823009,
      "grad_norm": 6.28125,
      "learning_rate": 1.17871594988328e-08,
      "loss": 0.9618,
      "num_input_tokens_seen": 21947880,
      "step": 39610
    },
    {
      "epoch": 695.0,
      "grad_norm": 20.375,
      "learning_rate": 1.1487648066466072e-08,
      "loss": 1.02,
      "num_input_tokens_seen": 21950592,
      "step": 39615
    },
    {
      "epoch": 695.0884955752213,
      "grad_norm": 7.65625,
      "learning_rate": 1.1191990176728784e-08,
      "loss": 1.0673,
      "num_input_tokens_seen": 21953456,
      "step": 39620
    },
    {
      "epoch": 695.1769911504425,
      "grad_norm": 9.125,
      "learning_rate": 1.0900185875215018e-08,
      "loss": 1.0342,
      "num_input_tokens_seen": 21956432,
      "step": 39625
    },
    {
      "epoch": 695.2654867256637,
      "grad_norm": 9.1875,
      "learning_rate": 1.0612235206924891e-08,
      "loss": 1.0038,
      "num_input_tokens_seen": 21959472,
      "step": 39630
    },
    {
      "epoch": 695.3539823008849,
      "grad_norm": 6.875,
      "learning_rate": 1.0328138216264549e-08,
      "loss": 0.8835,
      "num_input_tokens_seen": 21961632,
      "step": 39635
    },
    {
      "epoch": 695.4424778761062,
      "grad_norm": 9.6875,
      "learning_rate": 1.004789494704339e-08,
      "loss": 1.1567,
      "num_input_tokens_seen": 21964240,
      "step": 39640
    },
    {
      "epoch": 695.5309734513274,
      "grad_norm": 8.8125,
      "learning_rate": 9.771505442482397e-09,
      "loss": 1.1227,
      "num_input_tokens_seen": 21967120,
      "step": 39645
    },
    {
      "epoch": 695.6194690265487,
      "grad_norm": 7.4375,
      "learning_rate": 9.498969745200259e-09,
      "loss": 0.9705,
      "num_input_tokens_seen": 21969936,
      "step": 39650
    },
    {
      "epoch": 695.70796460177,
      "grad_norm": 7.90625,
      "learning_rate": 9.230287897230017e-09,
      "loss": 1.123,
      "num_input_tokens_seen": 21972608,
      "step": 39655
    },
    {
      "epoch": 695.7964601769911,
      "grad_norm": 5.53125,
      "learning_rate": 8.965459940002419e-09,
      "loss": 1.0427,
      "num_input_tokens_seen": 21975232,
      "step": 39660
    },
    {
      "epoch": 695.8849557522124,
      "grad_norm": 8.25,
      "learning_rate": 8.704485914357019e-09,
      "loss": 1.037,
      "num_input_tokens_seen": 21978400,
      "step": 39665
    },
    {
      "epoch": 695.9734513274336,
      "grad_norm": 8.8125,
      "learning_rate": 8.447365860539402e-09,
      "loss": 1.112,
      "num_input_tokens_seen": 21981664,
      "step": 39670
    },
    {
      "epoch": 696.0530973451328,
      "grad_norm": 8.0,
      "learning_rate": 8.194099818201184e-09,
      "loss": 0.85,
      "num_input_tokens_seen": 21983856,
      "step": 39675
    },
    {
      "epoch": 696.141592920354,
      "grad_norm": 7.3125,
      "learning_rate": 7.944687826400011e-09,
      "loss": 0.9295,
      "num_input_tokens_seen": 21987232,
      "step": 39680
    },
    {
      "epoch": 696.2300884955753,
      "grad_norm": 7.4375,
      "learning_rate": 7.699129923599557e-09,
      "loss": 1.0953,
      "num_input_tokens_seen": 21989440,
      "step": 39685
    },
    {
      "epoch": 696.3185840707964,
      "grad_norm": 7.40625,
      "learning_rate": 7.457426147663982e-09,
      "loss": 1.0763,
      "num_input_tokens_seen": 21992976,
      "step": 39690
    },
    {
      "epoch": 696.4070796460177,
      "grad_norm": 9.625,
      "learning_rate": 7.219576535871797e-09,
      "loss": 1.0339,
      "num_input_tokens_seen": 21995776,
      "step": 39695
    },
    {
      "epoch": 696.4955752212389,
      "grad_norm": 8.3125,
      "learning_rate": 6.985581124896445e-09,
      "loss": 1.0953,
      "num_input_tokens_seen": 21998432,
      "step": 39700
    },
    {
      "epoch": 696.5840707964602,
      "grad_norm": 9.3125,
      "learning_rate": 6.755439950828501e-09,
      "loss": 1.1219,
      "num_input_tokens_seen": 22001072,
      "step": 39705
    },
    {
      "epoch": 696.6725663716815,
      "grad_norm": 11.6875,
      "learning_rate": 6.5291530491562444e-09,
      "loss": 1.1794,
      "num_input_tokens_seen": 22003616,
      "step": 39710
    },
    {
      "epoch": 696.7610619469026,
      "grad_norm": 7.96875,
      "learning_rate": 6.3067204547739845e-09,
      "loss": 0.9421,
      "num_input_tokens_seen": 22006480,
      "step": 39715
    },
    {
      "epoch": 696.8495575221239,
      "grad_norm": 8.375,
      "learning_rate": 6.088142201987612e-09,
      "loss": 1.0114,
      "num_input_tokens_seen": 22009520,
      "step": 39720
    },
    {
      "epoch": 696.9380530973451,
      "grad_norm": 8.5625,
      "learning_rate": 5.873418324503499e-09,
      "loss": 1.0603,
      "num_input_tokens_seen": 22012224,
      "step": 39725
    },
    {
      "epoch": 697.0176991150443,
      "grad_norm": 8.9375,
      "learning_rate": 5.6625488554340465e-09,
      "loss": 1.0838,
      "num_input_tokens_seen": 22014416,
      "step": 39730
    },
    {
      "epoch": 697.1061946902655,
      "grad_norm": 7.21875,
      "learning_rate": 5.455533827297688e-09,
      "loss": 0.9236,
      "num_input_tokens_seen": 22017440,
      "step": 39735
    },
    {
      "epoch": 697.1946902654868,
      "grad_norm": 9.25,
      "learning_rate": 5.252373272018885e-09,
      "loss": 0.9275,
      "num_input_tokens_seen": 22020464,
      "step": 39740
    },
    {
      "epoch": 697.2831858407079,
      "grad_norm": 8.5,
      "learning_rate": 5.053067220925356e-09,
      "loss": 1.2923,
      "num_input_tokens_seen": 22023232,
      "step": 39745
    },
    {
      "epoch": 697.3716814159292,
      "grad_norm": 9.6875,
      "learning_rate": 4.857615704759177e-09,
      "loss": 1.1198,
      "num_input_tokens_seen": 22026016,
      "step": 39750
    },
    {
      "epoch": 697.4601769911504,
      "grad_norm": 9.5,
      "learning_rate": 4.666018753654577e-09,
      "loss": 1.1176,
      "num_input_tokens_seen": 22028704,
      "step": 39755
    },
    {
      "epoch": 697.5486725663717,
      "grad_norm": 9.875,
      "learning_rate": 4.478276397162917e-09,
      "loss": 1.1563,
      "num_input_tokens_seen": 22031088,
      "step": 39760
    },
    {
      "epoch": 697.637168141593,
      "grad_norm": 11.875,
      "learning_rate": 4.294388664233262e-09,
      "loss": 1.0803,
      "num_input_tokens_seen": 22033792,
      "step": 39765
    },
    {
      "epoch": 697.7256637168142,
      "grad_norm": 9.0625,
      "learning_rate": 4.114355583223484e-09,
      "loss": 0.9855,
      "num_input_tokens_seen": 22036928,
      "step": 39770
    },
    {
      "epoch": 697.8141592920354,
      "grad_norm": 7.28125,
      "learning_rate": 3.9381771818974845e-09,
      "loss": 0.9659,
      "num_input_tokens_seen": 22039776,
      "step": 39775
    },
    {
      "epoch": 697.9026548672566,
      "grad_norm": 11.375,
      "learning_rate": 3.765853487427973e-09,
      "loss": 1.0267,
      "num_input_tokens_seen": 22042432,
      "step": 39780
    },
    {
      "epoch": 697.9911504424779,
      "grad_norm": 7.1875,
      "learning_rate": 3.5973845263825857e-09,
      "loss": 0.9825,
      "num_input_tokens_seen": 22045424,
      "step": 39785
    },
    {
      "epoch": 698.070796460177,
      "grad_norm": 5.0625,
      "learning_rate": 3.4327703247488684e-09,
      "loss": 0.8519,
      "num_input_tokens_seen": 22047560,
      "step": 39790
    },
    {
      "epoch": 698.1592920353983,
      "grad_norm": 8.25,
      "learning_rate": 3.2720109079037443e-09,
      "loss": 1.1258,
      "num_input_tokens_seen": 22050120,
      "step": 39795
    },
    {
      "epoch": 698.2477876106195,
      "grad_norm": 8.25,
      "learning_rate": 3.1151063006468193e-09,
      "loss": 1.0302,
      "num_input_tokens_seen": 22053128,
      "step": 39800
    },
    {
      "epoch": 698.2477876106195,
      "eval_loss": 1.0310944318771362,
      "eval_runtime": 1.0048,
      "eval_samples_per_second": 24.881,
      "eval_steps_per_second": 12.938,
      "num_input_tokens_seen": 22053128,
      "step": 39800
    },
    {
      "epoch": 698.3362831858407,
      "grad_norm": 5.6875,
      "learning_rate": 2.962056527169854e-09,
      "loss": 0.8212,
      "num_input_tokens_seen": 22056024,
      "step": 39805
    },
    {
      "epoch": 698.4247787610619,
      "grad_norm": 8.5625,
      "learning_rate": 2.8128616110761898e-09,
      "loss": 1.1084,
      "num_input_tokens_seen": 22058504,
      "step": 39810
    },
    {
      "epoch": 698.5132743362832,
      "grad_norm": 9.0,
      "learning_rate": 2.6675215753724223e-09,
      "loss": 1.0394,
      "num_input_tokens_seen": 22061176,
      "step": 39815
    },
    {
      "epoch": 698.6017699115044,
      "grad_norm": 7.3125,
      "learning_rate": 2.5260364424739557e-09,
      "loss": 1.1022,
      "num_input_tokens_seen": 22064072,
      "step": 39820
    },
    {
      "epoch": 698.6902654867257,
      "grad_norm": 9.3125,
      "learning_rate": 2.3884062341994475e-09,
      "loss": 1.1032,
      "num_input_tokens_seen": 22067400,
      "step": 39825
    },
    {
      "epoch": 698.7787610619469,
      "grad_norm": 7.5625,
      "learning_rate": 2.25463097177081e-09,
      "loss": 1.1242,
      "num_input_tokens_seen": 22069912,
      "step": 39830
    },
    {
      "epoch": 698.8672566371681,
      "grad_norm": 8.0625,
      "learning_rate": 2.1247106758215397e-09,
      "loss": 1.0822,
      "num_input_tokens_seen": 22072456,
      "step": 39835
    },
    {
      "epoch": 698.9557522123894,
      "grad_norm": 8.1875,
      "learning_rate": 1.998645366382834e-09,
      "loss": 1.0733,
      "num_input_tokens_seen": 22075560,
      "step": 39840
    },
    {
      "epoch": 699.0353982300885,
      "grad_norm": 9.625,
      "learning_rate": 1.876435062897475e-09,
      "loss": 0.9502,
      "num_input_tokens_seen": 22078352,
      "step": 39845
    },
    {
      "epoch": 699.1238938053098,
      "grad_norm": 8.375,
      "learning_rate": 1.758079784211497e-09,
      "loss": 0.7858,
      "num_input_tokens_seen": 22081216,
      "step": 39850
    },
    {
      "epoch": 699.212389380531,
      "grad_norm": 8.25,
      "learning_rate": 1.6435795485797434e-09,
      "loss": 1.1894,
      "num_input_tokens_seen": 22083952,
      "step": 39855
    },
    {
      "epoch": 699.3008849557522,
      "grad_norm": 8.9375,
      "learning_rate": 1.5329343736547596e-09,
      "loss": 1.0199,
      "num_input_tokens_seen": 22086864,
      "step": 39860
    },
    {
      "epoch": 699.3893805309734,
      "grad_norm": 7.875,
      "learning_rate": 1.4261442765006739e-09,
      "loss": 1.1236,
      "num_input_tokens_seen": 22089584,
      "step": 39865
    },
    {
      "epoch": 699.4778761061947,
      "grad_norm": 9.1875,
      "learning_rate": 1.3232092735876445e-09,
      "loss": 1.1375,
      "num_input_tokens_seen": 22092384,
      "step": 39870
    },
    {
      "epoch": 699.566371681416,
      "grad_norm": 8.4375,
      "learning_rate": 1.2241293807918607e-09,
      "loss": 1.0768,
      "num_input_tokens_seen": 22095152,
      "step": 39875
    },
    {
      "epoch": 699.6548672566372,
      "grad_norm": 8.0,
      "learning_rate": 1.128904613387216e-09,
      "loss": 0.9638,
      "num_input_tokens_seen": 22097904,
      "step": 39880
    },
    {
      "epoch": 699.7433628318585,
      "grad_norm": 7.9375,
      "learning_rate": 1.0375349860591853e-09,
      "loss": 0.9703,
      "num_input_tokens_seen": 22100720,
      "step": 39885
    },
    {
      "epoch": 699.8318584070796,
      "grad_norm": 10.625,
      "learning_rate": 9.5002051290205e-10,
      "loss": 1.184,
      "num_input_tokens_seen": 22103184,
      "step": 39890
    },
    {
      "epoch": 699.9203539823009,
      "grad_norm": 8.4375,
      "learning_rate": 8.663612074077954e-10,
      "loss": 1.0703,
      "num_input_tokens_seen": 22105936,
      "step": 39895
    },
    {
      "epoch": 700.0,
      "grad_norm": 30.375,
      "learning_rate": 7.865570824799884e-10,
      "loss": 1.3927,
      "num_input_tokens_seen": 22108104,
      "step": 39900
    },
    {
      "epoch": 700.0884955752213,
      "grad_norm": 9.3125,
      "learning_rate": 7.106081504254514e-10,
      "loss": 1.2332,
      "num_input_tokens_seen": 22110856,
      "step": 39905
    },
    {
      "epoch": 700.1769911504425,
      "grad_norm": 11.5625,
      "learning_rate": 6.385144229570372e-10,
      "loss": 1.1341,
      "num_input_tokens_seen": 22113608,
      "step": 39910
    },
    {
      "epoch": 700.2654867256637,
      "grad_norm": 10.0,
      "learning_rate": 5.70275911190854e-10,
      "loss": 1.0557,
      "num_input_tokens_seen": 22116072,
      "step": 39915
    },
    {
      "epoch": 700.3539823008849,
      "grad_norm": 7.40625,
      "learning_rate": 5.058926256490403e-10,
      "loss": 0.9841,
      "num_input_tokens_seen": 22118840,
      "step": 39920
    },
    {
      "epoch": 700.4424778761062,
      "grad_norm": 7.75,
      "learning_rate": 4.4536457626254134e-10,
      "loss": 1.0484,
      "num_input_tokens_seen": 22121640,
      "step": 39925
    },
    {
      "epoch": 700.5309734513274,
      "grad_norm": 8.25,
      "learning_rate": 3.88691772365557e-10,
      "loss": 0.8099,
      "num_input_tokens_seen": 22124568,
      "step": 39930
    },
    {
      "epoch": 700.6194690265487,
      "grad_norm": 9.1875,
      "learning_rate": 3.358742226955425e-10,
      "loss": 1.0403,
      "num_input_tokens_seen": 22127400,
      "step": 39935
    },
    {
      "epoch": 700.70796460177,
      "grad_norm": 10.3125,
      "learning_rate": 2.8691193539875925e-10,
      "loss": 0.9812,
      "num_input_tokens_seen": 22130328,
      "step": 39940
    },
    {
      "epoch": 700.7964601769911,
      "grad_norm": 12.625,
      "learning_rate": 2.418049180274995e-10,
      "loss": 1.0974,
      "num_input_tokens_seen": 22133512,
      "step": 39945
    },
    {
      "epoch": 700.8849557522124,
      "grad_norm": 7.3125,
      "learning_rate": 2.005531775373104e-10,
      "loss": 1.1063,
      "num_input_tokens_seen": 22136136,
      "step": 39950
    },
    {
      "epoch": 700.9734513274336,
      "grad_norm": 6.96875,
      "learning_rate": 1.6315672028699435e-10,
      "loss": 1.1024,
      "num_input_tokens_seen": 22139512,
      "step": 39955
    },
    {
      "epoch": 701.0530973451328,
      "grad_norm": 7.625,
      "learning_rate": 1.2961555204693555e-10,
      "loss": 0.9561,
      "num_input_tokens_seen": 22141648,
      "step": 39960
    },
    {
      "epoch": 701.141592920354,
      "grad_norm": 7.3125,
      "learning_rate": 9.992967798799768e-11,
      "loss": 0.9276,
      "num_input_tokens_seen": 22144800,
      "step": 39965
    },
    {
      "epoch": 701.2300884955753,
      "grad_norm": 10.0,
      "learning_rate": 7.409910268707521e-11,
      "loss": 1.1954,
      "num_input_tokens_seen": 22147616,
      "step": 39970
    },
    {
      "epoch": 701.3185840707964,
      "grad_norm": 9.75,
      "learning_rate": 5.212383012986877e-11,
      "loss": 1.1826,
      "num_input_tokens_seen": 22150912,
      "step": 39975
    },
    {
      "epoch": 701.4070796460177,
      "grad_norm": 7.5,
      "learning_rate": 3.400386370533415e-11,
      "loss": 1.0434,
      "num_input_tokens_seen": 22153504,
      "step": 39980
    },
    {
      "epoch": 701.4955752212389,
      "grad_norm": 13.3125,
      "learning_rate": 1.9739206205682258e-11,
      "loss": 1.0067,
      "num_input_tokens_seen": 22156544,
      "step": 39985
    },
    {
      "epoch": 701.5840707964602,
      "grad_norm": 7.625,
      "learning_rate": 9.329859829154685e-12,
      "loss": 1.0387,
      "num_input_tokens_seen": 22159536,
      "step": 39990
    },
    {
      "epoch": 701.6725663716815,
      "grad_norm": 6.125,
      "learning_rate": 2.7758261855748148e-12,
      "loss": 1.06,
      "num_input_tokens_seen": 22162128,
      "step": 39995
    },
    {
      "epoch": 701.7610619469026,
      "grad_norm": 8.0,
      "learning_rate": 7.710628524559838e-14,
      "loss": 0.9108,
      "num_input_tokens_seen": 22164464,
      "step": 40000
    },
    {
      "epoch": 701.7610619469026,
      "eval_loss": 1.0310944318771362,
      "eval_runtime": 1.0156,
      "eval_samples_per_second": 24.616,
      "eval_steps_per_second": 12.8,
      "num_input_tokens_seen": 22164464,
      "step": 40000
    },
    {
      "epoch": 701.7610619469026,
      "num_input_tokens_seen": 22164464,
      "step": 40000,
      "total_flos": 9.98091202692907e+17,
      "train_loss": 1.0564270602226258,
      "train_runtime": 11305.8108,
      "train_samples_per_second": 14.152,
      "train_steps_per_second": 3.538
    }
  ],
  "logging_steps": 5,
  "max_steps": 40000,
  "num_input_tokens_seen": 22164464,
  "num_train_epochs": 715,
  "save_steps": 200,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 9.98091202692907e+17,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}