{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.995722277064824,
  "global_step": 569,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 0.0,
      "loss": 1.6786,
      "step": 1
    },
    {
      "epoch": 0.05,
      "learning_rate": 9.998088142969587e-06,
      "loss": 1.64,
      "step": 10
    },
    {
      "epoch": 0.11,
      "learning_rate": 9.985017573980262e-06,
      "loss": 1.06,
      "step": 20
    },
    {
      "epoch": 0.16,
      "learning_rate": 9.956012654497073e-06,
      "loss": 0.486,
      "step": 30
    },
    {
      "epoch": 0.21,
      "learning_rate": 9.911850333228427e-06,
      "loss": 0.2374,
      "step": 40
    },
    {
      "epoch": 0.26,
      "learning_rate": 9.865696363132769e-06,
      "loss": 0.1837,
      "step": 50
    },
    {
      "epoch": 0.32,
      "learning_rate": 9.7946219754852e-06,
      "loss": 0.1587,
      "step": 60
    },
    {
      "epoch": 0.37,
      "learning_rate": 9.708883781112711e-06,
      "loss": 0.1517,
      "step": 70
    },
    {
      "epoch": 0.42,
      "learning_rate": 9.629910009876223e-06,
      "loss": 0.1553,
      "step": 80
    },
    {
      "epoch": 0.47,
      "learning_rate": 9.51846738818602e-06,
      "loss": 0.1469,
      "step": 90
    },
    {
      "epoch": 0.53,
      "learning_rate": 9.39320554720971e-06,
      "loss": 0.145,
      "step": 100
    },
    {
      "epoch": 0.53,
      "eval_loss": 0.05455470830202103,
      "eval_runtime": 85.5684,
      "eval_samples_per_second": 11.383,
      "eval_steps_per_second": 1.426,
      "step": 100
    },
    {
      "epoch": 0.58,
      "learning_rate": 9.297503309182422e-06,
      "loss": 0.1549,
      "step": 110
    },
    {
      "epoch": 0.63,
      "learning_rate": 9.149650063920841e-06,
      "loss": 0.1435,
      "step": 120
    },
    {
      "epoch": 0.68,
      "learning_rate": 8.989105585268073e-06,
      "loss": 0.1414,
      "step": 130
    },
    {
      "epoch": 0.74,
      "learning_rate": 8.816360880276967e-06,
      "loss": 0.1421,
      "step": 140
    },
    {
      "epoch": 0.79,
      "learning_rate": 8.631944269006895e-06,
      "loss": 0.1391,
      "step": 150
    },
    {
      "epoch": 0.84,
      "learning_rate": 8.436419768716853e-06,
      "loss": 0.1386,
      "step": 160
    },
    {
      "epoch": 0.9,
      "learning_rate": 8.230385368882732e-06,
      "loss": 0.1382,
      "step": 170
    },
    {
      "epoch": 0.95,
      "learning_rate": 8.014471202314443e-06,
      "loss": 0.1382,
      "step": 180
    },
    {
      "epoch": 1.0,
      "learning_rate": 7.789337617966275e-06,
      "loss": 0.1357,
      "step": 190
    },
    {
      "epoch": 1.05,
      "learning_rate": 7.5794033237905e-06,
      "loss": 0.1408,
      "step": 200
    },
    {
      "epoch": 1.05,
      "eval_loss": 0.048922911286354065,
      "eval_runtime": 85.5902,
      "eval_samples_per_second": 11.38,
      "eval_steps_per_second": 1.425,
      "step": 200
    },
    {
      "epoch": 1.11,
      "learning_rate": 7.338671397287409e-06,
      "loss": 0.1341,
      "step": 210
    },
    {
      "epoch": 1.16,
      "learning_rate": 7.1158757783214904e-06,
      "loss": 0.1391,
      "step": 220
    },
    {
      "epoch": 1.21,
      "learning_rate": 6.862201608610134e-06,
      "loss": 0.1343,
      "step": 230
    },
    {
      "epoch": 1.26,
      "learning_rate": 6.602832106793113e-06,
      "loss": 0.1316,
      "step": 240
    },
    {
      "epoch": 1.32,
      "learning_rate": 6.338560525031794e-06,
      "loss": 0.1316,
      "step": 250
    },
    {
      "epoch": 1.37,
      "learning_rate": 6.0701951079422615e-06,
      "loss": 0.1313,
      "step": 260
    },
    {
      "epoch": 1.42,
      "learning_rate": 5.825844147403353e-06,
      "loss": 0.1347,
      "step": 270
    },
    {
      "epoch": 1.47,
      "learning_rate": 5.579447229838992e-06,
      "loss": 0.1337,
      "step": 280
    },
    {
      "epoch": 1.53,
      "learning_rate": 5.3040158758857886e-06,
      "loss": 0.1315,
      "step": 290
    },
    {
      "epoch": 1.58,
      "learning_rate": 5.027654723907197e-06,
      "loss": 0.1309,
      "step": 300
    },
    {
      "epoch": 1.58,
      "eval_loss": 0.04729650914669037,
      "eval_runtime": 85.6206,
      "eval_samples_per_second": 11.376,
      "eval_steps_per_second": 1.425,
      "step": 300
    },
    {
      "epoch": 1.63,
      "learning_rate": 4.751208993096637e-06,
      "loss": 0.1307,
      "step": 310
    },
    {
      "epoch": 1.68,
      "learning_rate": 4.475524161322288e-06,
      "loss": 0.1305,
      "step": 320
    },
    {
      "epoch": 1.74,
      "learning_rate": 4.2014433793290435e-06,
      "loss": 0.1298,
      "step": 330
    },
    {
      "epoch": 1.79,
      "learning_rate": 3.92980489205774e-06,
      "loss": 0.1308,
      "step": 340
    },
    {
      "epoch": 1.84,
      "learning_rate": 3.6614394749682057e-06,
      "loss": 0.13,
      "step": 350
    },
    {
      "epoch": 1.9,
      "learning_rate": 3.3971678932068875e-06,
      "loss": 0.1296,
      "step": 360
    },
    {
      "epoch": 1.95,
      "learning_rate": 3.1377983913898673e-06,
      "loss": 0.1289,
      "step": 370
    },
    {
      "epoch": 2.0,
      "learning_rate": 2.8841242216785116e-06,
      "loss": 0.129,
      "step": 380
    },
    {
      "epoch": 2.05,
      "learning_rate": 2.6369212177078306e-06,
      "loss": 0.1291,
      "step": 390
    },
    {
      "epoch": 2.11,
      "learning_rate": 2.3969454217874325e-06,
      "loss": 0.1277,
      "step": 400
    },
    {
      "epoch": 2.11,
      "eval_loss": 0.046792980283498764,
      "eval_runtime": 85.593,
      "eval_samples_per_second": 11.379,
      "eval_steps_per_second": 1.425,
      "step": 400
    },
    {
      "epoch": 2.16,
      "learning_rate": 2.164930772631996e-06,
      "loss": 0.1276,
      "step": 410
    },
    {
      "epoch": 2.21,
      "learning_rate": 1.94158686069306e-06,
      "loss": 0.1275,
      "step": 420
    },
    {
      "epoch": 2.26,
      "learning_rate": 1.7275967579572427e-06,
      "loss": 0.1296,
      "step": 430
    },
    {
      "epoch": 2.32,
      "learning_rate": 1.5236149288481428e-06,
      "loss": 0.1275,
      "step": 440
    },
    {
      "epoch": 2.37,
      "learning_rate": 1.3302652286212397e-06,
      "loss": 0.1287,
      "step": 450
    },
    {
      "epoch": 2.42,
      "learning_rate": 1.148138995373459e-06,
      "loss": 0.1267,
      "step": 460
    },
    {
      "epoch": 2.47,
      "learning_rate": 9.777932415027608e-07,
      "loss": 0.1282,
      "step": 470
    },
    {
      "epoch": 2.53,
      "learning_rate": 8.197489501489924e-07,
      "loss": 0.1249,
      "step": 480
    },
    {
      "epoch": 2.58,
      "learning_rate": 6.744894818261311e-07,
      "loss": 0.127,
      "step": 490
    },
    {
      "epoch": 2.63,
      "learning_rate": 5.424590961190474e-07,
      "loss": 0.1253,
      "step": 500
    },
    {
      "epoch": 2.63,
      "eval_loss": 0.046507786959409714,
      "eval_runtime": 85.5909,
      "eval_samples_per_second": 11.38,
      "eval_steps_per_second": 1.425,
      "step": 500
    },
    {
      "epoch": 2.69,
      "learning_rate": 4.240615929660341e-07,
      "loss": 0.1279,
      "step": 510
    },
    {
      "epoch": 2.74,
      "learning_rate": 3.1965907768255035e-07,
      "loss": 0.128,
      "step": 520
    },
    {
      "epoch": 2.79,
      "learning_rate": 2.2957085350325092e-07,
      "loss": 0.1288,
      "step": 530
    },
    {
      "epoch": 2.84,
      "learning_rate": 1.540724450293035e-07,
      "loss": 0.1264,
      "step": 540
    },
    {
      "epoch": 2.9,
      "learning_rate": 9.339475556770006e-08,
      "loss": 0.1274,
      "step": 550
    },
    {
      "epoch": 2.95,
      "learning_rate": 4.77233609397082e-08,
      "loss": 0.1277,
      "step": 560
    },
    {
      "epoch": 3.0,
      "step": 569,
      "total_flos": 1.6024405879092675e+18,
      "train_loss": 0.18567189055712655,
      "train_runtime": 15471.5941,
      "train_samples_per_second": 4.707,
      "train_steps_per_second": 0.037
    }
  ],
  "max_steps": 569,
  "num_train_epochs": 4,
  "total_flos": 1.6024405879092675e+18,
  "trial_name": null,
  "trial_params": null
}