{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.0202020202020203,
  "eval_steps": 20,
  "global_step": 400,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.050505050505050504,
      "grad_norm": 0.1097809225320816,
      "learning_rate": 2.9999999999999997e-05,
      "loss": 1.2164,
      "step": 10
    },
    {
      "epoch": 0.10101010101010101,
      "grad_norm": 0.16844922304153442,
      "learning_rate": 5.9999999999999995e-05,
      "loss": 1.1774,
      "step": 20
    },
    {
      "epoch": 0.10101010101010101,
      "eval_loss": 1.1594151258468628,
      "eval_runtime": 295.5081,
      "eval_samples_per_second": 2.382,
      "eval_steps_per_second": 0.298,
      "step": 20
    },
    {
      "epoch": 0.15151515151515152,
      "grad_norm": 0.3893352150917053,
      "learning_rate": 8.999999999999999e-05,
      "loss": 0.9942,
      "step": 30
    },
    {
      "epoch": 0.20202020202020202,
      "grad_norm": 0.4363686740398407,
      "learning_rate": 0.00011999999999999999,
      "loss": 0.7259,
      "step": 40
    },
    {
      "epoch": 0.20202020202020202,
      "eval_loss": 0.6028452515602112,
      "eval_runtime": 294.8084,
      "eval_samples_per_second": 2.388,
      "eval_steps_per_second": 0.298,
      "step": 40
    },
    {
      "epoch": 0.25252525252525254,
      "grad_norm": 0.40602898597717285,
      "learning_rate": 0.00015,
      "loss": 0.4454,
      "step": 50
    },
    {
      "epoch": 0.30303030303030304,
      "grad_norm": 0.2177702635526657,
      "learning_rate": 0.00017999999999999998,
      "loss": 0.5139,
      "step": 60
    },
    {
      "epoch": 0.30303030303030304,
      "eval_loss": 0.3834875822067261,
      "eval_runtime": 295.1108,
      "eval_samples_per_second": 2.386,
      "eval_steps_per_second": 0.298,
      "step": 60
    },
    {
      "epoch": 0.35353535353535354,
      "grad_norm": 0.1777501404285431,
      "learning_rate": 0.00020999999999999998,
      "loss": 0.4096,
      "step": 70
    },
    {
      "epoch": 0.40404040404040403,
      "grad_norm": 0.1749625951051712,
      "learning_rate": 0.00023999999999999998,
      "loss": 0.2589,
      "step": 80
    },
    {
      "epoch": 0.40404040404040403,
      "eval_loss": 0.28596043586730957,
      "eval_runtime": 294.9045,
      "eval_samples_per_second": 2.387,
      "eval_steps_per_second": 0.298,
      "step": 80
    },
    {
      "epoch": 0.45454545454545453,
      "grad_norm": 0.2074197232723236,
      "learning_rate": 0.00027,
      "loss": 0.1932,
      "step": 90
    },
    {
      "epoch": 0.5050505050505051,
      "grad_norm": 0.36931025981903076,
      "learning_rate": 0.0003,
      "loss": 0.2177,
      "step": 100
    },
    {
      "epoch": 0.5050505050505051,
      "eval_loss": 0.25518086552619934,
      "eval_runtime": 295.271,
      "eval_samples_per_second": 2.384,
      "eval_steps_per_second": 0.298,
      "step": 100
    },
    {
      "epoch": 0.5555555555555556,
      "grad_norm": 0.20555801689624786,
      "learning_rate": 0.00029,
      "loss": 0.299,
      "step": 110
    },
    {
      "epoch": 0.6060606060606061,
      "grad_norm": 0.14223843812942505,
      "learning_rate": 0.00028,
      "loss": 0.2508,
      "step": 120
    },
    {
      "epoch": 0.6060606060606061,
      "eval_loss": 0.2106790542602539,
      "eval_runtime": 294.889,
      "eval_samples_per_second": 2.387,
      "eval_steps_per_second": 0.298,
      "step": 120
    },
    {
      "epoch": 0.6565656565656566,
      "grad_norm": 0.14103205502033234,
      "learning_rate": 0.00027,
      "loss": 0.1666,
      "step": 130
    },
    {
      "epoch": 0.7070707070707071,
      "grad_norm": 0.1655016392469406,
      "learning_rate": 0.00026,
      "loss": 0.1511,
      "step": 140
    },
    {
      "epoch": 0.7070707070707071,
      "eval_loss": 0.20343247056007385,
      "eval_runtime": 294.7497,
      "eval_samples_per_second": 2.388,
      "eval_steps_per_second": 0.299,
      "step": 140
    },
    {
      "epoch": 0.7575757575757576,
      "grad_norm": 0.33550018072128296,
      "learning_rate": 0.00025,
      "loss": 0.1798,
      "step": 150
    },
    {
      "epoch": 0.8080808080808081,
      "grad_norm": 0.18753309547901154,
      "learning_rate": 0.00023999999999999998,
      "loss": 0.245,
      "step": 160
    },
    {
      "epoch": 0.8080808080808081,
      "eval_loss": 0.19205082952976227,
      "eval_runtime": 294.8122,
      "eval_samples_per_second": 2.388,
      "eval_steps_per_second": 0.298,
      "step": 160
    },
    {
      "epoch": 0.8585858585858586,
      "grad_norm": 0.18931691348552704,
      "learning_rate": 0.00023,
      "loss": 0.2125,
      "step": 170
    },
    {
      "epoch": 0.9090909090909091,
      "grad_norm": 0.1642046868801117,
      "learning_rate": 0.00021999999999999995,
      "loss": 0.1479,
      "step": 180
    },
    {
      "epoch": 0.9090909090909091,
      "eval_loss": 0.1822979599237442,
      "eval_runtime": 295.0975,
      "eval_samples_per_second": 2.386,
      "eval_steps_per_second": 0.298,
      "step": 180
    },
    {
      "epoch": 0.9595959595959596,
      "grad_norm": 0.14441154897212982,
      "learning_rate": 0.00020999999999999998,
      "loss": 0.1399,
      "step": 190
    },
    {
      "epoch": 1.0101010101010102,
      "grad_norm": 0.20781415700912476,
      "learning_rate": 0.00019999999999999998,
      "loss": 0.17,
      "step": 200
    },
    {
      "epoch": 1.0101010101010102,
      "eval_loss": 0.18901540338993073,
      "eval_runtime": 294.9368,
      "eval_samples_per_second": 2.387,
      "eval_steps_per_second": 0.298,
      "step": 200
    },
    {
      "epoch": 1.0606060606060606,
      "grad_norm": 0.18098795413970947,
      "learning_rate": 0.00018999999999999998,
      "loss": 0.2102,
      "step": 210
    },
    {
      "epoch": 1.1111111111111112,
      "grad_norm": 0.12826304137706757,
      "learning_rate": 0.00017999999999999998,
      "loss": 0.2085,
      "step": 220
    },
    {
      "epoch": 1.1111111111111112,
      "eval_loss": 0.17017190158367157,
      "eval_runtime": 295.0936,
      "eval_samples_per_second": 2.386,
      "eval_steps_per_second": 0.298,
      "step": 220
    },
    {
      "epoch": 1.1616161616161615,
      "grad_norm": 0.13167431950569153,
      "learning_rate": 0.00016999999999999999,
      "loss": 0.1399,
      "step": 230
    },
    {
      "epoch": 1.2121212121212122,
      "grad_norm": 0.13884004950523376,
      "learning_rate": 0.00015999999999999999,
      "loss": 0.1311,
      "step": 240
    },
    {
      "epoch": 1.2121212121212122,
      "eval_loss": 0.16837279498577118,
      "eval_runtime": 294.8077,
      "eval_samples_per_second": 2.388,
      "eval_steps_per_second": 0.298,
      "step": 240
    },
    {
      "epoch": 1.2626262626262625,
      "grad_norm": 0.17478157579898834,
      "learning_rate": 0.00015,
      "loss": 0.1559,
      "step": 250
    },
    {
      "epoch": 1.3131313131313131,
      "grad_norm": 0.16669002175331116,
      "learning_rate": 0.00014,
      "loss": 0.205,
      "step": 260
    },
    {
      "epoch": 1.3131313131313131,
      "eval_loss": 0.16558928787708282,
      "eval_runtime": 294.9081,
      "eval_samples_per_second": 2.387,
      "eval_steps_per_second": 0.298,
      "step": 260
    },
    {
      "epoch": 1.3636363636363638,
      "grad_norm": 0.14852865040302277,
      "learning_rate": 0.00013,
      "loss": 0.1826,
      "step": 270
    },
    {
      "epoch": 1.4141414141414141,
      "grad_norm": 0.14237315952777863,
      "learning_rate": 0.00011999999999999999,
      "loss": 0.1329,
      "step": 280
    },
    {
      "epoch": 1.4141414141414141,
      "eval_loss": 0.15846213698387146,
      "eval_runtime": 295.0174,
      "eval_samples_per_second": 2.386,
      "eval_steps_per_second": 0.298,
      "step": 280
    },
    {
      "epoch": 1.4646464646464645,
      "grad_norm": 0.1401522010564804,
      "learning_rate": 0.00010999999999999998,
      "loss": 0.1205,
      "step": 290
    },
    {
      "epoch": 1.5151515151515151,
      "grad_norm": 0.18983517587184906,
      "learning_rate": 9.999999999999999e-05,
      "loss": 0.1607,
      "step": 300
    },
    {
      "epoch": 1.5151515151515151,
      "eval_loss": 0.16163967549800873,
      "eval_runtime": 294.807,
      "eval_samples_per_second": 2.388,
      "eval_steps_per_second": 0.299,
      "step": 300
    },
    {
      "epoch": 1.5656565656565657,
      "grad_norm": 0.1500634402036667,
      "learning_rate": 8.999999999999999e-05,
      "loss": 0.2121,
      "step": 310
    },
    {
      "epoch": 1.6161616161616161,
      "grad_norm": 0.15094490349292755,
      "learning_rate": 7.999999999999999e-05,
      "loss": 0.1713,
      "step": 320
    },
    {
      "epoch": 1.6161616161616161,
      "eval_loss": 0.1540195345878601,
      "eval_runtime": 295.157,
      "eval_samples_per_second": 2.385,
      "eval_steps_per_second": 0.298,
      "step": 320
    },
    {
      "epoch": 1.6666666666666665,
      "grad_norm": 0.15443100035190582,
      "learning_rate": 7e-05,
      "loss": 0.129,
      "step": 330
    },
    {
      "epoch": 1.7171717171717171,
      "grad_norm": 0.1568954586982727,
      "learning_rate": 5.9999999999999995e-05,
      "loss": 0.1196,
      "step": 340
    },
    {
      "epoch": 1.7171717171717171,
      "eval_loss": 0.15283828973770142,
      "eval_runtime": 295.3671,
      "eval_samples_per_second": 2.383,
      "eval_steps_per_second": 0.298,
      "step": 340
    },
    {
      "epoch": 1.7676767676767677,
      "grad_norm": 0.19448110461235046,
      "learning_rate": 4.9999999999999996e-05,
      "loss": 0.1603,
      "step": 350
    },
    {
      "epoch": 1.8181818181818183,
      "grad_norm": 0.12574005126953125,
      "learning_rate": 3.9999999999999996e-05,
      "loss": 0.1722,
      "step": 360
    },
    {
      "epoch": 1.8181818181818183,
      "eval_loss": 0.15110255777835846,
      "eval_runtime": 295.1101,
      "eval_samples_per_second": 2.386,
      "eval_steps_per_second": 0.298,
      "step": 360
    },
    {
      "epoch": 1.8686868686868687,
      "grad_norm": 0.11745467782020569,
      "learning_rate": 2.9999999999999997e-05,
      "loss": 0.1669,
      "step": 370
    },
    {
      "epoch": 1.9191919191919191,
      "grad_norm": 0.1326703280210495,
      "learning_rate": 1.9999999999999998e-05,
      "loss": 0.1243,
      "step": 380
    },
    {
      "epoch": 1.9191919191919191,
      "eval_loss": 0.1487378031015396,
      "eval_runtime": 295.0041,
      "eval_samples_per_second": 2.386,
      "eval_steps_per_second": 0.298,
      "step": 380
    },
    {
      "epoch": 1.9696969696969697,
      "grad_norm": 0.1454104483127594,
      "learning_rate": 9.999999999999999e-06,
      "loss": 0.1148,
      "step": 390
    },
    {
      "epoch": 2.0202020202020203,
      "grad_norm": 0.1772913634777069,
      "learning_rate": 0.0,
      "loss": 0.1633,
      "step": 400
    },
    {
      "epoch": 2.0202020202020203,
      "eval_loss": 0.14871937036514282,
      "eval_runtime": 295.1345,
      "eval_samples_per_second": 2.385,
      "eval_steps_per_second": 0.298,
      "step": 400
    }
  ],
  "logging_steps": 10,
  "max_steps": 400,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 20,
  "total_flos": 2.2256168195948544e+17,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}