{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 1000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "grad_norm": 4.737985134124756,
      "learning_rate": 5e-06,
      "loss": 1.5912,
      "step": 10
    },
    {
      "epoch": 0.02,
      "grad_norm": 7.149003982543945,
      "learning_rate": 1e-05,
      "loss": 3.2011,
      "step": 20
    },
    {
      "epoch": 0.03,
      "grad_norm": 6.096139430999756,
      "learning_rate": 1.5e-05,
      "loss": 1.2445,
      "step": 30
    },
    {
      "epoch": 0.04,
      "grad_norm": 6.4711103439331055,
      "learning_rate": 2e-05,
      "loss": 1.4005,
      "step": 40
    },
    {
      "epoch": 0.05,
      "grad_norm": 6.055766582489014,
      "learning_rate": 2.5e-05,
      "loss": 0.9311,
      "step": 50
    },
    {
      "epoch": 0.06,
      "grad_norm": 23.06619644165039,
      "learning_rate": 3e-05,
      "loss": 2.2437,
      "step": 60
    },
    {
      "epoch": 0.07,
      "grad_norm": 9.455103874206543,
      "learning_rate": 3.5e-05,
      "loss": 2.7137,
      "step": 70
    },
    {
      "epoch": 0.08,
      "grad_norm": 8.64001750946045,
      "learning_rate": 4e-05,
      "loss": 2.6398,
      "step": 80
    },
    {
      "epoch": 0.09,
      "grad_norm": 58.382110595703125,
      "learning_rate": 4.5e-05,
      "loss": 2.2014,
      "step": 90
    },
    {
      "epoch": 0.1,
      "grad_norm": 5.963547706604004,
      "learning_rate": 5e-05,
      "loss": 2.0685,
      "step": 100
    },
    {
      "epoch": 0.11,
      "grad_norm": 5.909413814544678,
      "learning_rate": 4.99847706754774e-05,
      "loss": 3.1558,
      "step": 110
    },
    {
      "epoch": 0.12,
      "grad_norm": 4.110403537750244,
      "learning_rate": 4.993910125649561e-05,
      "loss": 2.2536,
      "step": 120
    },
    {
      "epoch": 0.13,
      "grad_norm": 17.61237907409668,
      "learning_rate": 4.9863047384206835e-05,
      "loss": 3.2304,
      "step": 130
    },
    {
      "epoch": 0.14,
      "grad_norm": 42.55618667602539,
      "learning_rate": 4.975670171853926e-05,
      "loss": 1.8168,
      "step": 140
    },
    {
      "epoch": 0.15,
      "grad_norm": 7.033435344696045,
      "learning_rate": 4.962019382530521e-05,
      "loss": 0.7345,
      "step": 150
    },
    {
      "epoch": 0.16,
      "grad_norm": 6.939785480499268,
      "learning_rate": 4.9453690018345144e-05,
      "loss": 1.0468,
      "step": 160
    },
    {
      "epoch": 0.17,
      "grad_norm": 7.7263078689575195,
      "learning_rate": 4.925739315689991e-05,
      "loss": 1.6574,
      "step": 170
    },
    {
      "epoch": 0.18,
      "grad_norm": 14.468883514404297,
      "learning_rate": 4.9031542398457974e-05,
      "loss": 1.6383,
      "step": 180
    },
    {
      "epoch": 0.19,
      "grad_norm": 5.9138946533203125,
      "learning_rate": 4.877641290737884e-05,
      "loss": 0.8235,
      "step": 190
    },
    {
      "epoch": 0.2,
      "grad_norm": 11.40308666229248,
      "learning_rate": 4.849231551964771e-05,
      "loss": 0.489,
      "step": 200
    },
    {
      "epoch": 0.21,
      "grad_norm": 20.220449447631836,
      "learning_rate": 4.817959636416969e-05,
      "loss": 1.4448,
      "step": 210
    },
    {
      "epoch": 0.22,
      "grad_norm": 29.097707748413086,
      "learning_rate": 4.783863644106502e-05,
      "loss": 1.7983,
      "step": 220
    },
    {
      "epoch": 0.23,
      "grad_norm": 16.577442169189453,
      "learning_rate": 4.7469851157479177e-05,
      "loss": 1.8216,
      "step": 230
    },
    {
      "epoch": 0.24,
      "grad_norm": 3.516134262084961,
      "learning_rate": 4.707368982147318e-05,
      "loss": 2.0506,
      "step": 240
    },
    {
      "epoch": 0.25,
      "grad_norm": 3.0684404373168945,
      "learning_rate": 4.665063509461097e-05,
      "loss": 1.258,
      "step": 250
    },
    {
      "epoch": 0.26,
      "grad_norm": 4.914421558380127,
      "learning_rate": 4.620120240391065e-05,
      "loss": 0.3911,
      "step": 260
    },
    {
      "epoch": 0.27,
      "grad_norm": 6.704916954040527,
      "learning_rate": 4.572593931387604e-05,
      "loss": 0.6957,
      "step": 270
    },
    {
      "epoch": 0.28,
      "grad_norm": 5.172149181365967,
      "learning_rate": 4.522542485937369e-05,
      "loss": 1.3392,
      "step": 280
    },
    {
      "epoch": 0.29,
      "grad_norm": 2.9674534797668457,
      "learning_rate": 4.4700268840168045e-05,
      "loss": 0.6846,
      "step": 290
    },
    {
      "epoch": 0.3,
      "grad_norm": 2.472653388977051,
      "learning_rate": 4.415111107797445e-05,
      "loss": 0.2735,
      "step": 300
    },
    {
      "epoch": 0.31,
      "grad_norm": 2.405486583709717,
      "learning_rate": 4.357862063693486e-05,
      "loss": 0.2887,
      "step": 310
    },
    {
      "epoch": 0.32,
      "grad_norm": 2.07614803314209,
      "learning_rate": 4.2983495008466276e-05,
      "loss": 0.3026,
      "step": 320
    },
    {
      "epoch": 0.33,
      "grad_norm": 7.051891803741455,
      "learning_rate": 4.2366459261474933e-05,
      "loss": 0.688,
      "step": 330
    },
    {
      "epoch": 0.34,
      "grad_norm": 3.9787468910217285,
      "learning_rate": 4.172826515897146e-05,
      "loss": 2.5623,
      "step": 340
    },
    {
      "epoch": 0.35,
      "grad_norm": 5.217522621154785,
      "learning_rate": 4.1069690242163484e-05,
      "loss": 1.2036,
      "step": 350
    },
    {
      "epoch": 0.36,
      "grad_norm": 5.582896709442139,
      "learning_rate": 4.039153688314145e-05,
      "loss": 1.3976,
      "step": 360
    },
    {
      "epoch": 0.37,
      "grad_norm": 4.498598098754883,
      "learning_rate": 3.969463130731183e-05,
      "loss": 1.6016,
      "step": 370
    },
    {
      "epoch": 0.38,
      "grad_norm": 3.115513324737549,
      "learning_rate": 3.897982258676867e-05,
      "loss": 0.7646,
      "step": 380
    },
    {
      "epoch": 0.39,
      "grad_norm": 3.488847494125366,
      "learning_rate": 3.824798160583012e-05,
      "loss": 0.5091,
      "step": 390
    },
    {
      "epoch": 0.4,
      "grad_norm": 5.372682094573975,
      "learning_rate": 3.7500000000000003e-05,
      "loss": 1.3678,
      "step": 400
    },
    {
      "epoch": 0.41,
      "grad_norm": 5.9022674560546875,
      "learning_rate": 3.673678906964727e-05,
      "loss": 2.1068,
      "step": 410
    },
    {
      "epoch": 0.42,
      "grad_norm": 3.6478700637817383,
      "learning_rate": 3.5959278669726935e-05,
      "loss": 1.4413,
      "step": 420
    },
    {
      "epoch": 0.43,
      "grad_norm": 7.891127109527588,
      "learning_rate": 3.516841607689501e-05,
      "loss": 1.2757,
      "step": 430
    },
    {
      "epoch": 0.44,
      "grad_norm": 5.28378438949585,
      "learning_rate": 3.436516483539781e-05,
      "loss": 0.7885,
      "step": 440
    },
    {
      "epoch": 0.45,
      "grad_norm": 6.047487735748291,
      "learning_rate": 3.355050358314172e-05,
      "loss": 2.3279,
      "step": 450
    },
    {
      "epoch": 0.46,
      "grad_norm": 2.6575264930725098,
      "learning_rate": 3.272542485937369e-05,
      "loss": 0.5281,
      "step": 460
    },
    {
      "epoch": 0.47,
      "grad_norm": 2.3986458778381348,
      "learning_rate": 3.1890933895424976e-05,
      "loss": 0.2615,
      "step": 470
    },
    {
      "epoch": 0.48,
      "grad_norm": 1.6229203939437866,
      "learning_rate": 3.104804738999169e-05,
      "loss": 0.4922,
      "step": 480
    },
    {
      "epoch": 0.49,
      "grad_norm": 1.5451277494430542,
      "learning_rate": 3.0197792270443982e-05,
      "loss": 0.2118,
      "step": 490
    },
    {
      "epoch": 0.5,
      "grad_norm": 1.311353087425232,
      "learning_rate": 2.9341204441673266e-05,
      "loss": 0.1915,
      "step": 500
    },
    {
      "epoch": 0.5,
      "eval_loss": 1.5571472644805908,
      "eval_runtime": 5999.5519,
      "eval_samples_per_second": 5.286,
      "eval_steps_per_second": 5.286,
      "step": 500
    },
    {
      "epoch": 0.51,
      "grad_norm": 1.1946927309036255,
      "learning_rate": 2.8479327524001636e-05,
      "loss": 0.1914,
      "step": 510
    },
    {
      "epoch": 0.52,
      "grad_norm": 1.0991920232772827,
      "learning_rate": 2.761321158169134e-05,
      "loss": 0.2036,
      "step": 520
    },
    {
      "epoch": 0.53,
      "grad_norm": 1.178324580192566,
      "learning_rate": 2.674391184360313e-05,
      "loss": 0.1903,
      "step": 530
    },
    {
      "epoch": 0.54,
      "grad_norm": 1.6252118349075317,
      "learning_rate": 2.587248741756253e-05,
      "loss": 0.1897,
      "step": 540
    },
    {
      "epoch": 0.55,
      "grad_norm": 2.4392871856689453,
      "learning_rate": 2.5e-05,
      "loss": 0.2962,
      "step": 550
    },
    {
      "epoch": 0.56,
      "grad_norm": 4.543885707855225,
      "learning_rate": 2.4127512582437485e-05,
      "loss": 0.7395,
      "step": 560
    },
    {
      "epoch": 0.57,
      "grad_norm": 1.537638783454895,
      "learning_rate": 2.3256088156396868e-05,
      "loss": 0.2917,
      "step": 570
    },
    {
      "epoch": 0.58,
      "grad_norm": 1.9524015188217163,
      "learning_rate": 2.238678841830867e-05,
      "loss": 0.2657,
      "step": 580
    },
    {
      "epoch": 0.59,
      "grad_norm": 3.946528673171997,
      "learning_rate": 2.1520672475998373e-05,
      "loss": 0.9463,
      "step": 590
    },
    {
      "epoch": 0.6,
      "grad_norm": 6.766805171966553,
      "learning_rate": 2.0658795558326743e-05,
      "loss": 2.8137,
      "step": 600
    },
    {
      "epoch": 0.61,
      "grad_norm": 5.235180377960205,
      "learning_rate": 1.980220772955602e-05,
      "loss": 1.1135,
      "step": 610
    },
    {
      "epoch": 0.62,
      "grad_norm": 11.928383827209473,
      "learning_rate": 1.895195261000831e-05,
      "loss": 1.6387,
      "step": 620
    },
    {
      "epoch": 0.63,
      "grad_norm": 1.4929327964782715,
      "learning_rate": 1.8109066104575023e-05,
      "loss": 0.5561,
      "step": 630
    },
    {
      "epoch": 0.64,
      "grad_norm": 1.8914227485656738,
      "learning_rate": 1.7274575140626318e-05,
      "loss": 0.2111,
      "step": 640
    },
    {
      "epoch": 0.65,
      "grad_norm": 3.6962478160858154,
      "learning_rate": 1.6449496416858284e-05,
      "loss": 0.6296,
      "step": 650
    },
    {
      "epoch": 0.66,
      "grad_norm": 4.319825649261475,
      "learning_rate": 1.56348351646022e-05,
      "loss": 1.7621,
      "step": 660
    },
    {
      "epoch": 0.67,
      "grad_norm": 5.848764896392822,
      "learning_rate": 1.4831583923104999e-05,
      "loss": 1.7205,
      "step": 670
    },
    {
      "epoch": 0.68,
      "grad_norm": 7.215566158294678,
      "learning_rate": 1.4040721330273062e-05,
      "loss": 1.4845,
      "step": 680
    },
    {
      "epoch": 0.69,
      "grad_norm": 4.199477195739746,
      "learning_rate": 1.3263210930352737e-05,
      "loss": 2.1168,
      "step": 690
    },
    {
      "epoch": 0.7,
      "grad_norm": 6.580512046813965,
      "learning_rate": 1.2500000000000006e-05,
      "loss": 1.7481,
      "step": 700
    },
    {
      "epoch": 0.71,
      "grad_norm": 7.199362754821777,
      "learning_rate": 1.175201839416988e-05,
      "loss": 1.6295,
      "step": 710
    },
    {
      "epoch": 0.72,
      "grad_norm": 3.5996954441070557,
      "learning_rate": 1.1020177413231334e-05,
      "loss": 1.266,
      "step": 720
    },
    {
      "epoch": 0.73,
      "grad_norm": 7.7472968101501465,
      "learning_rate": 1.0305368692688174e-05,
      "loss": 0.5478,
      "step": 730
    },
    {
      "epoch": 0.74,
      "grad_norm": 3.7776882648468018,
      "learning_rate": 9.608463116858542e-06,
      "loss": 1.1993,
      "step": 740
    },
    {
      "epoch": 0.75,
      "grad_norm": 2.427421808242798,
      "learning_rate": 8.930309757836517e-06,
      "loss": 1.6423,
      "step": 750
    },
    {
      "epoch": 0.76,
      "grad_norm": 3.11273193359375,
      "learning_rate": 8.271734841028553e-06,
      "loss": 0.4206,
      "step": 760
    },
    {
      "epoch": 0.77,
      "grad_norm": 1.6959909200668335,
      "learning_rate": 7.633540738525066e-06,
      "loss": 0.257,
      "step": 770
    },
    {
      "epoch": 0.78,
      "grad_norm": 4.651413440704346,
      "learning_rate": 7.016504991533726e-06,
      "loss": 0.3458,
      "step": 780
    },
    {
      "epoch": 0.79,
      "grad_norm": 4.119582653045654,
      "learning_rate": 6.421379363065142e-06,
      "loss": 1.721,
      "step": 790
    },
    {
      "epoch": 0.8,
      "grad_norm": 4.266563415527344,
      "learning_rate": 5.848888922025553e-06,
      "loss": 1.2284,
      "step": 800
    },
    {
      "epoch": 0.81,
      "grad_norm": 2.50252366065979,
      "learning_rate": 5.299731159831953e-06,
      "loss": 1.1696,
      "step": 810
    },
    {
      "epoch": 0.82,
      "grad_norm": 2.9079418182373047,
      "learning_rate": 4.7745751406263165e-06,
      "loss": 0.953,
      "step": 820
    },
    {
      "epoch": 0.83,
      "grad_norm": 3.4728331565856934,
      "learning_rate": 4.274060686123959e-06,
      "loss": 1.7072,
      "step": 830
    },
    {
      "epoch": 0.84,
      "grad_norm": 1.1848517656326294,
      "learning_rate": 3.798797596089351e-06,
      "loss": 0.1604,
      "step": 840
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.647183895111084,
      "learning_rate": 3.3493649053890326e-06,
      "loss": 0.0397,
      "step": 850
    },
    {
      "epoch": 0.86,
      "grad_norm": 7.117775917053223,
      "learning_rate": 2.9263101785268254e-06,
      "loss": 0.8525,
      "step": 860
    },
    {
      "epoch": 0.87,
      "grad_norm": 4.795046329498291,
      "learning_rate": 2.5301488425208296e-06,
      "loss": 5.167,
      "step": 870
    },
    {
      "epoch": 0.88,
      "grad_norm": 4.327600955963135,
      "learning_rate": 2.1613635589349756e-06,
      "loss": 4.8475,
      "step": 880
    },
    {
      "epoch": 0.89,
      "grad_norm": 4.947289943695068,
      "learning_rate": 1.8204036358303173e-06,
      "loss": 5.0221,
      "step": 890
    },
    {
      "epoch": 0.9,
      "grad_norm": 3.9482979774475098,
      "learning_rate": 1.5076844803522922e-06,
      "loss": 4.9018,
      "step": 900
    },
    {
      "epoch": 0.91,
      "grad_norm": 6.357035160064697,
      "learning_rate": 1.2235870926211619e-06,
      "loss": 4.4224,
      "step": 910
    },
    {
      "epoch": 0.92,
      "grad_norm": 3.0725436210632324,
      "learning_rate": 9.684576015420278e-07,
      "loss": 0.8727,
      "step": 920
    },
    {
      "epoch": 0.93,
      "grad_norm": 2.637556552886963,
      "learning_rate": 7.426068431000882e-07,
      "loss": 0.5169,
      "step": 930
    },
    {
      "epoch": 0.94,
      "grad_norm": 2.796225070953369,
      "learning_rate": 5.463099816548579e-07,
      "loss": 0.6463,
      "step": 940
    },
    {
      "epoch": 0.95,
      "grad_norm": 2.2664058208465576,
      "learning_rate": 3.7980617469479953e-07,
      "loss": 0.2752,
      "step": 950
    },
    {
      "epoch": 0.96,
      "grad_norm": 4.001941680908203,
      "learning_rate": 2.4329828146074095e-07,
      "loss": 0.5726,
      "step": 960
    },
    {
      "epoch": 0.97,
      "grad_norm": 3.260882616043091,
      "learning_rate": 1.3695261579316777e-07,
      "loss": 1.4077,
      "step": 970
    },
    {
      "epoch": 0.98,
      "grad_norm": 5.255369663238525,
      "learning_rate": 6.089874350439506e-08,
      "loss": 1.4728,
      "step": 980
    },
    {
      "epoch": 0.99,
      "grad_norm": 3.699906826019287,
      "learning_rate": 1.522932452260595e-08,
      "loss": 1.2553,
      "step": 990
    },
    {
      "epoch": 1.0,
      "grad_norm": 3.7628183364868164,
      "learning_rate": 0.0,
      "loss": 0.6231,
      "step": 1000
    },
    {
      "epoch": 1.0,
      "eval_loss": 1.401175856590271,
      "eval_runtime": 6003.7558,
      "eval_samples_per_second": 5.282,
      "eval_steps_per_second": 5.282,
      "step": 1000
    }
  ],
  "logging_steps": 10,
  "max_steps": 1000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9223372036854775807,
  "save_steps": 500,
  "total_flos": 2.5061788483584e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}