{
  "best_metric": 1.311687707901001,
  "best_model_checkpoint": "./output/checkpoints/2024-06-11_13-53-54/checkpoint-10",
  "epoch": 3.0,
  "eval_steps": 1,
  "global_step": 54,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.05555555555555555,
      "grad_norm": 1.9413901567459106,
      "learning_rate": 6.666666666666667e-05,
      "loss": 3.544,
      "step": 1
    },
    {
      "epoch": 0.05555555555555555,
      "eval_loss": 3.981394052505493,
      "eval_runtime": 12.5428,
      "eval_samples_per_second": 11.401,
      "eval_steps_per_second": 0.718,
      "step": 1
    },
    {
      "epoch": 0.1111111111111111,
      "grad_norm": 2.0220563411712646,
      "learning_rate": 0.00013333333333333334,
      "loss": 3.6373,
      "step": 2
    },
    {
      "epoch": 0.1111111111111111,
      "eval_loss": 3.788512706756592,
      "eval_runtime": 12.6976,
      "eval_samples_per_second": 11.262,
      "eval_steps_per_second": 0.709,
      "step": 2
    },
    {
      "epoch": 0.16666666666666666,
      "grad_norm": 1.90468430519104,
      "learning_rate": 0.0002,
      "loss": 3.3119,
      "step": 3
    },
    {
      "epoch": 0.16666666666666666,
      "eval_loss": 3.026510238647461,
      "eval_runtime": 12.655,
      "eval_samples_per_second": 11.3,
      "eval_steps_per_second": 0.711,
      "step": 3
    },
    {
      "epoch": 0.2222222222222222,
      "grad_norm": 2.224761724472046,
      "learning_rate": 0.0002666666666666667,
      "loss": 2.5408,
      "step": 4
    },
    {
      "epoch": 0.2222222222222222,
      "eval_loss": 2.1147303581237793,
      "eval_runtime": 12.7118,
      "eval_samples_per_second": 11.249,
      "eval_steps_per_second": 0.708,
      "step": 4
    },
    {
      "epoch": 0.2777777777777778,
      "grad_norm": 1.1557661294937134,
      "learning_rate": 0.0003333333333333334,
      "loss": 1.7003,
      "step": 5
    },
    {
      "epoch": 0.2777777777777778,
      "eval_loss": 1.702366590499878,
      "eval_runtime": 12.7156,
      "eval_samples_per_second": 11.246,
      "eval_steps_per_second": 0.708,
      "step": 5
    },
    {
      "epoch": 0.3333333333333333,
      "grad_norm": 1.0826658010482788,
      "learning_rate": 0.0004,
      "loss": 1.3464,
      "step": 6
    },
    {
      "epoch": 0.3333333333333333,
      "eval_loss": 1.4100245237350464,
      "eval_runtime": 12.7488,
      "eval_samples_per_second": 11.217,
      "eval_steps_per_second": 0.706,
      "step": 6
    },
    {
      "epoch": 0.3888888888888889,
      "grad_norm": 0.7294781804084778,
      "learning_rate": 0.0003916666666666667,
      "loss": 1.0757,
      "step": 7
    },
    {
      "epoch": 0.3888888888888889,
      "eval_loss": 1.2797588109970093,
      "eval_runtime": 12.7015,
      "eval_samples_per_second": 11.259,
      "eval_steps_per_second": 0.709,
      "step": 7
    },
    {
      "epoch": 0.4444444444444444,
      "grad_norm": 0.6977319121360779,
      "learning_rate": 0.00038333333333333334,
      "loss": 0.9387,
      "step": 8
    },
    {
      "epoch": 0.4444444444444444,
      "eval_loss": 1.233304738998413,
      "eval_runtime": 12.7924,
      "eval_samples_per_second": 11.179,
      "eval_steps_per_second": 0.704,
      "step": 8
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.25461432337760925,
      "learning_rate": 0.000375,
      "loss": 0.8435,
      "step": 9
    },
    {
      "epoch": 0.5,
      "eval_loss": 1.256897211074829,
      "eval_runtime": 12.7345,
      "eval_samples_per_second": 11.229,
      "eval_steps_per_second": 0.707,
      "step": 9
    },
    {
      "epoch": 0.5555555555555556,
      "grad_norm": 0.20771068334579468,
      "learning_rate": 0.00036666666666666667,
      "loss": 0.7791,
      "step": 10
    },
    {
      "epoch": 0.5555555555555556,
      "eval_loss": 1.311687707901001,
      "eval_runtime": 12.7288,
      "eval_samples_per_second": 11.234,
      "eval_steps_per_second": 0.707,
      "step": 10
    },
    {
      "epoch": 0.6111111111111112,
      "grad_norm": 0.24797727167606354,
      "learning_rate": 0.0003583333333333334,
      "loss": 0.7485,
      "step": 11
    },
    {
      "epoch": 0.6111111111111112,
      "eval_loss": 1.3500157594680786,
      "eval_runtime": 12.7478,
      "eval_samples_per_second": 11.218,
      "eval_steps_per_second": 0.706,
      "step": 11
    },
    {
      "epoch": 0.6666666666666666,
      "grad_norm": 0.23104745149612427,
      "learning_rate": 0.00035,
      "loss": 0.6892,
      "step": 12
    },
    {
      "epoch": 0.6666666666666666,
      "eval_loss": 1.3617342710494995,
      "eval_runtime": 12.6893,
      "eval_samples_per_second": 11.269,
      "eval_steps_per_second": 0.709,
      "step": 12
    },
    {
      "epoch": 0.7222222222222222,
      "grad_norm": 0.16125920414924622,
      "learning_rate": 0.00034166666666666666,
      "loss": 0.6602,
      "step": 13
    },
    {
      "epoch": 0.7222222222222222,
      "eval_loss": 1.3614617586135864,
      "eval_runtime": 12.6941,
      "eval_samples_per_second": 11.265,
      "eval_steps_per_second": 0.709,
      "step": 13
    },
    {
      "epoch": 0.7777777777777778,
      "grad_norm": 0.1296168714761734,
      "learning_rate": 0.0003333333333333334,
      "loss": 0.6505,
      "step": 14
    },
    {
      "epoch": 0.7777777777777778,
      "eval_loss": 1.3570654392242432,
      "eval_runtime": 12.7115,
      "eval_samples_per_second": 11.25,
      "eval_steps_per_second": 0.708,
      "step": 14
    },
    {
      "epoch": 0.8333333333333334,
      "grad_norm": 0.14436963200569153,
      "learning_rate": 0.00032500000000000004,
      "loss": 0.6267,
      "step": 15
    },
    {
      "epoch": 0.8333333333333334,
      "eval_loss": 1.3531259298324585,
      "eval_runtime": 12.7384,
      "eval_samples_per_second": 11.226,
      "eval_steps_per_second": 0.707,
      "step": 15
    },
    {
      "epoch": 0.8888888888888888,
      "grad_norm": 0.1280059963464737,
      "learning_rate": 0.00031666666666666665,
      "loss": 0.6341,
      "step": 16
    },
    {
      "epoch": 0.8888888888888888,
      "eval_loss": 1.3554112911224365,
      "eval_runtime": 12.8038,
      "eval_samples_per_second": 11.169,
      "eval_steps_per_second": 0.703,
      "step": 16
    },
    {
      "epoch": 0.9444444444444444,
      "grad_norm": 0.09258498251438141,
      "learning_rate": 0.00030833333333333337,
      "loss": 0.6277,
      "step": 17
    },
    {
      "epoch": 0.9444444444444444,
      "eval_loss": 1.3708301782608032,
      "eval_runtime": 12.7261,
      "eval_samples_per_second": 11.237,
      "eval_steps_per_second": 0.707,
      "step": 17
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.14155693352222443,
      "learning_rate": 0.00030000000000000003,
      "loss": 0.6253,
      "step": 18
    },
    {
      "epoch": 1.0,
      "eval_loss": 1.3944224119186401,
      "eval_runtime": 12.6915,
      "eval_samples_per_second": 11.267,
      "eval_steps_per_second": 0.709,
      "step": 18
    },
    {
      "epoch": 1.0555555555555556,
      "grad_norm": 0.07174329459667206,
      "learning_rate": 0.0002916666666666667,
      "loss": 0.5724,
      "step": 19
    },
    {
      "epoch": 1.0555555555555556,
      "eval_loss": 1.4162791967391968,
      "eval_runtime": 12.8162,
      "eval_samples_per_second": 11.158,
      "eval_steps_per_second": 0.702,
      "step": 19
    },
    {
      "epoch": 1.1111111111111112,
      "grad_norm": 0.06882944703102112,
      "learning_rate": 0.00028333333333333335,
      "loss": 0.6061,
      "step": 20
    },
    {
      "epoch": 1.1111111111111112,
      "eval_loss": 1.4311778545379639,
      "eval_runtime": 12.7151,
      "eval_samples_per_second": 11.246,
      "eval_steps_per_second": 0.708,
      "step": 20
    },
    {
      "epoch": 1.1666666666666667,
      "grad_norm": 0.07908931374549866,
      "learning_rate": 0.000275,
      "loss": 0.5886,
      "step": 21
    },
    {
      "epoch": 1.1666666666666667,
      "eval_loss": 1.4312688112258911,
      "eval_runtime": 12.7335,
      "eval_samples_per_second": 11.23,
      "eval_steps_per_second": 0.707,
      "step": 21
    },
    {
      "epoch": 1.2222222222222223,
      "grad_norm": 0.07461100816726685,
      "learning_rate": 0.0002666666666666667,
      "loss": 0.5774,
      "step": 22
    },
    {
      "epoch": 1.2222222222222223,
      "eval_loss": 1.4189623594284058,
      "eval_runtime": 12.7737,
      "eval_samples_per_second": 11.195,
      "eval_steps_per_second": 0.705,
      "step": 22
    },
    {
      "epoch": 1.2777777777777777,
      "grad_norm": 0.06924545019865036,
      "learning_rate": 0.00025833333333333334,
      "loss": 0.5735,
      "step": 23
    },
    {
      "epoch": 1.2777777777777777,
      "eval_loss": 1.3990516662597656,
      "eval_runtime": 12.8077,
      "eval_samples_per_second": 11.165,
      "eval_steps_per_second": 0.703,
      "step": 23
    },
    {
      "epoch": 1.3333333333333333,
      "grad_norm": 0.056572116911411285,
      "learning_rate": 0.00025,
      "loss": 0.563,
      "step": 24
    },
    {
      "epoch": 1.3333333333333333,
      "eval_loss": 1.3818256855010986,
      "eval_runtime": 12.7694,
      "eval_samples_per_second": 11.199,
      "eval_steps_per_second": 0.705,
      "step": 24
    },
    {
      "epoch": 1.3888888888888888,
      "grad_norm": 0.05974334850907326,
      "learning_rate": 0.00024166666666666667,
      "loss": 0.5334,
      "step": 25
    },
    {
      "epoch": 1.3888888888888888,
      "eval_loss": 1.3698084354400635,
      "eval_runtime": 12.7833,
      "eval_samples_per_second": 11.187,
      "eval_steps_per_second": 0.704,
      "step": 25
    },
    {
      "epoch": 1.4444444444444444,
      "grad_norm": 0.05873117223381996,
      "learning_rate": 0.00023333333333333336,
      "loss": 0.5687,
      "step": 26
    },
    {
      "epoch": 1.4444444444444444,
      "eval_loss": 1.3634446859359741,
      "eval_runtime": 12.7332,
      "eval_samples_per_second": 11.23,
      "eval_steps_per_second": 0.707,
      "step": 26
    },
    {
      "epoch": 1.5,
      "grad_norm": 0.0712776780128479,
      "learning_rate": 0.00022500000000000002,
      "loss": 0.5681,
      "step": 27
    },
    {
      "epoch": 1.5,
      "eval_loss": 1.3670543432235718,
      "eval_runtime": 12.7169,
      "eval_samples_per_second": 11.245,
      "eval_steps_per_second": 0.708,
      "step": 27
    },
    {
      "epoch": 1.5555555555555556,
      "grad_norm": 0.06001581624150276,
      "learning_rate": 0.00021666666666666666,
      "loss": 0.5831,
      "step": 28
    },
    {
      "epoch": 1.5555555555555556,
      "eval_loss": 1.3755391836166382,
      "eval_runtime": 12.7091,
      "eval_samples_per_second": 11.252,
      "eval_steps_per_second": 0.708,
      "step": 28
    },
    {
      "epoch": 1.6111111111111112,
      "grad_norm": 0.059290554374456406,
      "learning_rate": 0.00020833333333333335,
      "loss": 0.5163,
      "step": 29
    },
    {
      "epoch": 1.6111111111111112,
      "eval_loss": 1.3861252069473267,
      "eval_runtime": 12.7394,
      "eval_samples_per_second": 11.225,
      "eval_steps_per_second": 0.706,
      "step": 29
    },
    {
      "epoch": 1.6666666666666665,
      "grad_norm": 0.05152224004268646,
      "learning_rate": 0.0002,
      "loss": 0.5735,
      "step": 30
    },
    {
      "epoch": 1.6666666666666665,
      "eval_loss": 1.3946057558059692,
      "eval_runtime": 12.7631,
      "eval_samples_per_second": 11.204,
      "eval_steps_per_second": 0.705,
      "step": 30
    },
    {
      "epoch": 1.7222222222222223,
      "grad_norm": 0.053750913590192795,
      "learning_rate": 0.00019166666666666667,
      "loss": 0.5616,
      "step": 31
    },
    {
      "epoch": 1.7222222222222223,
      "eval_loss": 1.4017627239227295,
      "eval_runtime": 12.7518,
      "eval_samples_per_second": 11.214,
      "eval_steps_per_second": 0.706,
      "step": 31
    },
    {
      "epoch": 1.7777777777777777,
      "grad_norm": 0.061749089509248734,
      "learning_rate": 0.00018333333333333334,
      "loss": 0.5418,
      "step": 32
    },
    {
      "epoch": 1.7777777777777777,
      "eval_loss": 1.4021672010421753,
      "eval_runtime": 13.2584,
      "eval_samples_per_second": 10.786,
      "eval_steps_per_second": 0.679,
      "step": 32
    },
    {
      "epoch": 1.8333333333333335,
      "grad_norm": 0.055768076330423355,
      "learning_rate": 0.000175,
      "loss": 0.5447,
      "step": 33
    },
    {
      "epoch": 1.8333333333333335,
      "eval_loss": 1.4005753993988037,
      "eval_runtime": 12.8325,
      "eval_samples_per_second": 11.144,
      "eval_steps_per_second": 0.701,
      "step": 33
    },
    {
      "epoch": 1.8888888888888888,
      "grad_norm": 0.058024097234010696,
      "learning_rate": 0.0001666666666666667,
      "loss": 0.5612,
      "step": 34
    },
    {
      "epoch": 1.8888888888888888,
      "eval_loss": 1.3969610929489136,
      "eval_runtime": 12.7822,
      "eval_samples_per_second": 11.187,
      "eval_steps_per_second": 0.704,
      "step": 34
    },
    {
      "epoch": 1.9444444444444444,
      "grad_norm": 0.055260155349969864,
      "learning_rate": 0.00015833333333333332,
      "loss": 0.5652,
      "step": 35
    },
    {
      "epoch": 1.9444444444444444,
      "eval_loss": 1.3906365633010864,
      "eval_runtime": 12.7076,
      "eval_samples_per_second": 11.253,
      "eval_steps_per_second": 0.708,
      "step": 35
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.12786947190761566,
      "learning_rate": 0.00015000000000000001,
      "loss": 0.5689,
      "step": 36
    },
    {
      "epoch": 2.0,
      "eval_loss": 1.3780609369277954,
      "eval_runtime": 12.722,
      "eval_samples_per_second": 11.24,
      "eval_steps_per_second": 0.707,
      "step": 36
    },
    {
      "epoch": 2.0555555555555554,
      "grad_norm": 0.05128234624862671,
      "learning_rate": 0.00014166666666666668,
      "loss": 0.5399,
      "step": 37
    },
    {
      "epoch": 2.0555555555555554,
      "eval_loss": 1.3723970651626587,
      "eval_runtime": 12.7179,
      "eval_samples_per_second": 11.244,
      "eval_steps_per_second": 0.708,
      "step": 37
    },
    {
      "epoch": 2.111111111111111,
      "grad_norm": 0.06030972674489021,
      "learning_rate": 0.00013333333333333334,
      "loss": 0.534,
      "step": 38
    },
    {
      "epoch": 2.111111111111111,
      "eval_loss": 1.3728123903274536,
      "eval_runtime": 12.7597,
      "eval_samples_per_second": 11.207,
      "eval_steps_per_second": 0.705,
      "step": 38
    },
    {
      "epoch": 2.1666666666666665,
      "grad_norm": 0.049170780926942825,
      "learning_rate": 0.000125,
      "loss": 0.533,
      "step": 39
    },
    {
      "epoch": 2.1666666666666665,
      "eval_loss": 1.375289797782898,
      "eval_runtime": 12.8384,
      "eval_samples_per_second": 11.138,
      "eval_steps_per_second": 0.701,
      "step": 39
    },
    {
      "epoch": 2.2222222222222223,
      "grad_norm": 0.05079694464802742,
      "learning_rate": 0.00011666666666666668,
      "loss": 0.5336,
      "step": 40
    },
    {
      "epoch": 2.2222222222222223,
      "eval_loss": 1.3813451528549194,
      "eval_runtime": 12.7475,
      "eval_samples_per_second": 11.218,
      "eval_steps_per_second": 0.706,
      "step": 40
    },
    {
      "epoch": 2.2777777777777777,
      "grad_norm": 0.046478718519210815,
      "learning_rate": 0.00010833333333333333,
      "loss": 0.5386,
      "step": 41
    },
    {
      "epoch": 2.2777777777777777,
      "eval_loss": 1.3882359266281128,
      "eval_runtime": 12.7759,
      "eval_samples_per_second": 11.193,
      "eval_steps_per_second": 0.704,
      "step": 41
    },
    {
      "epoch": 2.3333333333333335,
      "grad_norm": 0.05194352567195892,
      "learning_rate": 0.0001,
      "loss": 0.5277,
      "step": 42
    },
    {
      "epoch": 2.3333333333333335,
      "eval_loss": 1.396583080291748,
      "eval_runtime": 12.776,
      "eval_samples_per_second": 11.193,
      "eval_steps_per_second": 0.704,
      "step": 42
    },
    {
      "epoch": 2.388888888888889,
      "grad_norm": 0.048547472804784775,
      "learning_rate": 9.166666666666667e-05,
      "loss": 0.5222,
      "step": 43
    },
    {
      "epoch": 2.388888888888889,
      "eval_loss": 1.4045826196670532,
      "eval_runtime": 12.7747,
      "eval_samples_per_second": 11.194,
      "eval_steps_per_second": 0.705,
      "step": 43
    },
    {
      "epoch": 2.4444444444444446,
      "grad_norm": 0.04673220217227936,
      "learning_rate": 8.333333333333334e-05,
      "loss": 0.5182,
      "step": 44
    },
    {
      "epoch": 2.4444444444444446,
      "eval_loss": 1.4115573167800903,
      "eval_runtime": 12.7319,
      "eval_samples_per_second": 11.232,
      "eval_steps_per_second": 0.707,
      "step": 44
    },
    {
      "epoch": 2.5,
      "grad_norm": 0.04925357550382614,
      "learning_rate": 7.500000000000001e-05,
      "loss": 0.544,
      "step": 45
    },
    {
      "epoch": 2.5,
      "eval_loss": 1.417358160018921,
      "eval_runtime": 12.8413,
      "eval_samples_per_second": 11.136,
      "eval_steps_per_second": 0.701,
      "step": 45
    },
    {
      "epoch": 2.5555555555555554,
      "grad_norm": 0.05455848202109337,
      "learning_rate": 6.666666666666667e-05,
      "loss": 0.5309,
      "step": 46
    },
    {
      "epoch": 2.5555555555555554,
      "eval_loss": 1.420274257659912,
      "eval_runtime": 12.7701,
      "eval_samples_per_second": 11.198,
      "eval_steps_per_second": 0.705,
      "step": 46
    },
    {
      "epoch": 2.611111111111111,
      "grad_norm": 0.0490451417863369,
      "learning_rate": 5.833333333333334e-05,
      "loss": 0.5514,
      "step": 47
    },
    {
      "epoch": 2.611111111111111,
      "eval_loss": 1.4224988222122192,
      "eval_runtime": 12.8347,
      "eval_samples_per_second": 11.142,
      "eval_steps_per_second": 0.701,
      "step": 47
    },
    {
      "epoch": 2.6666666666666665,
      "grad_norm": 0.049362048506736755,
      "learning_rate": 5e-05,
      "loss": 0.5202,
      "step": 48
    },
    {
      "epoch": 2.6666666666666665,
      "eval_loss": 1.423671007156372,
      "eval_runtime": 12.7791,
      "eval_samples_per_second": 11.19,
      "eval_steps_per_second": 0.704,
      "step": 48
    },
    {
      "epoch": 2.7222222222222223,
      "grad_norm": 0.05601557716727257,
      "learning_rate": 4.166666666666667e-05,
      "loss": 0.5429,
      "step": 49
    },
    {
      "epoch": 2.7222222222222223,
      "eval_loss": 1.422905683517456,
      "eval_runtime": 12.7552,
      "eval_samples_per_second": 11.211,
      "eval_steps_per_second": 0.706,
      "step": 49
    },
    {
      "epoch": 2.7777777777777777,
      "grad_norm": 0.052850477397441864,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 0.5306,
      "step": 50
    },
    {
      "epoch": 2.7777777777777777,
      "eval_loss": 1.4214112758636475,
      "eval_runtime": 12.8332,
      "eval_samples_per_second": 11.143,
      "eval_steps_per_second": 0.701,
      "step": 50
    },
    {
      "epoch": 2.8333333333333335,
      "grad_norm": 0.06296616047620773,
      "learning_rate": 2.5e-05,
      "loss": 0.535,
      "step": 51
    },
    {
      "epoch": 2.8333333333333335,
      "eval_loss": 1.4192702770233154,
      "eval_runtime": 12.7525,
      "eval_samples_per_second": 11.213,
      "eval_steps_per_second": 0.706,
      "step": 51
    },
    {
      "epoch": 2.888888888888889,
      "grad_norm": 0.06152864173054695,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 0.5389,
      "step": 52
    },
    {
      "epoch": 2.888888888888889,
      "eval_loss": 1.4175350666046143,
      "eval_runtime": 12.72,
      "eval_samples_per_second": 11.242,
      "eval_steps_per_second": 0.708,
      "step": 52
    },
    {
      "epoch": 2.9444444444444446,
      "grad_norm": 0.04936552420258522,
      "learning_rate": 8.333333333333334e-06,
      "loss": 0.5438,
      "step": 53
    },
    {
      "epoch": 2.9444444444444446,
      "eval_loss": 1.4166295528411865,
      "eval_runtime": 12.785,
      "eval_samples_per_second": 11.185,
      "eval_steps_per_second": 0.704,
      "step": 53
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.11227822303771973,
      "learning_rate": 0.0,
      "loss": 0.5125,
      "step": 54
    },
    {
      "epoch": 3.0,
      "eval_loss": 1.416577696800232,
      "eval_runtime": 12.7343,
      "eval_samples_per_second": 11.229,
      "eval_steps_per_second": 0.707,
      "step": 54
    },
    {
      "epoch": 3.0,
      "step": 54,
      "total_flos": 1.908935806471373e+16,
      "train_loss": 0.828608477557147,
      "train_runtime": 1175.7023,
      "train_samples_per_second": 1.401,
      "train_steps_per_second": 0.046
    }
  ],
  "logging_steps": 1,
  "max_steps": 54,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 10,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.908935806471373e+16,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}