{
  "best_metric": 1.0333940982818604,
  "best_model_checkpoint": "./output/checkpoints/2024-06-10_16-11-56/checkpoint-40",
  "epoch": 1.0,
  "eval_steps": 1,
  "global_step": 49,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.02040816326530612,
      "grad_norm": 20.516210556030273,
      "learning_rate": 8e-05,
      "loss": 4.7411,
      "step": 1
    },
    {
      "epoch": 0.02040816326530612,
      "eval_loss": 4.733951568603516,
      "eval_runtime": 7.8124,
      "eval_samples_per_second": 7.424,
      "eval_steps_per_second": 0.512,
      "step": 1
    },
    {
      "epoch": 0.04081632653061224,
      "grad_norm": 20.680587768554688,
      "learning_rate": 0.00016,
      "loss": 4.755,
      "step": 2
    },
    {
      "epoch": 0.04081632653061224,
      "eval_loss": 2.225498676300049,
      "eval_runtime": 7.7691,
      "eval_samples_per_second": 7.465,
      "eval_steps_per_second": 0.515,
      "step": 2
    },
    {
      "epoch": 0.061224489795918366,
      "grad_norm": 11.66711711883545,
      "learning_rate": 0.00024,
      "loss": 2.2003,
      "step": 3
    },
    {
      "epoch": 0.061224489795918366,
      "eval_loss": 1.4929580688476562,
      "eval_runtime": 7.6881,
      "eval_samples_per_second": 7.544,
      "eval_steps_per_second": 0.52,
      "step": 3
    },
    {
      "epoch": 0.08163265306122448,
      "grad_norm": 3.128970146179199,
      "learning_rate": 0.00032,
      "loss": 1.6439,
      "step": 4
    },
    {
      "epoch": 0.08163265306122448,
      "eval_loss": 1.3402283191680908,
      "eval_runtime": 7.708,
      "eval_samples_per_second": 7.525,
      "eval_steps_per_second": 0.519,
      "step": 4
    },
    {
      "epoch": 0.10204081632653061,
      "grad_norm": 3.4351813793182373,
      "learning_rate": 0.0004,
      "loss": 1.3756,
      "step": 5
    },
    {
      "epoch": 0.10204081632653061,
      "eval_loss": 1.227684497833252,
      "eval_runtime": 7.7491,
      "eval_samples_per_second": 7.485,
      "eval_steps_per_second": 0.516,
      "step": 5
    },
    {
      "epoch": 0.12244897959183673,
      "grad_norm": 2.636197328567505,
      "learning_rate": 0.00039090909090909096,
      "loss": 1.2884,
      "step": 6
    },
    {
      "epoch": 0.12244897959183673,
      "eval_loss": 1.25971519947052,
      "eval_runtime": 7.8734,
      "eval_samples_per_second": 7.367,
      "eval_steps_per_second": 0.508,
      "step": 6
    },
    {
      "epoch": 0.14285714285714285,
      "grad_norm": 1.7103031873703003,
      "learning_rate": 0.00038181818181818184,
      "loss": 1.3103,
      "step": 7
    },
    {
      "epoch": 0.14285714285714285,
      "eval_loss": 1.1791292428970337,
      "eval_runtime": 7.7644,
      "eval_samples_per_second": 7.47,
      "eval_steps_per_second": 0.515,
      "step": 7
    },
    {
      "epoch": 0.16326530612244897,
      "grad_norm": 0.9646056294441223,
      "learning_rate": 0.00037272727272727273,
      "loss": 1.1536,
      "step": 8
    },
    {
      "epoch": 0.16326530612244897,
      "eval_loss": 1.3148419857025146,
      "eval_runtime": 7.8433,
      "eval_samples_per_second": 7.395,
      "eval_steps_per_second": 0.51,
      "step": 8
    },
    {
      "epoch": 0.1836734693877551,
      "grad_norm": 1.4100276231765747,
      "learning_rate": 0.00036363636363636367,
      "loss": 1.1972,
      "step": 9
    },
    {
      "epoch": 0.1836734693877551,
      "eval_loss": 1.2365388870239258,
      "eval_runtime": 7.8198,
      "eval_samples_per_second": 7.417,
      "eval_steps_per_second": 0.512,
      "step": 9
    },
    {
      "epoch": 0.20408163265306123,
      "grad_norm": 0.9489567875862122,
      "learning_rate": 0.00035454545454545455,
      "loss": 1.1192,
      "step": 10
    },
    {
      "epoch": 0.20408163265306123,
      "eval_loss": 1.1423839330673218,
      "eval_runtime": 7.8348,
      "eval_samples_per_second": 7.403,
      "eval_steps_per_second": 0.511,
      "step": 10
    },
    {
      "epoch": 0.22448979591836735,
      "grad_norm": 0.364205926656723,
      "learning_rate": 0.0003454545454545455,
      "loss": 1.0749,
      "step": 11
    },
    {
      "epoch": 0.22448979591836735,
      "eval_loss": 1.108851432800293,
      "eval_runtime": 7.8222,
      "eval_samples_per_second": 7.415,
      "eval_steps_per_second": 0.511,
      "step": 11
    },
    {
      "epoch": 0.24489795918367346,
      "grad_norm": 0.4010220170021057,
      "learning_rate": 0.0003363636363636364,
      "loss": 1.0388,
      "step": 12
    },
    {
      "epoch": 0.24489795918367346,
      "eval_loss": 1.1047261953353882,
      "eval_runtime": 7.768,
      "eval_samples_per_second": 7.466,
      "eval_steps_per_second": 0.515,
      "step": 12
    },
    {
      "epoch": 0.2653061224489796,
      "grad_norm": 0.26062682271003723,
      "learning_rate": 0.0003272727272727273,
      "loss": 1.0671,
      "step": 13
    },
    {
      "epoch": 0.2653061224489796,
      "eval_loss": 1.138075828552246,
      "eval_runtime": 7.8236,
      "eval_samples_per_second": 7.413,
      "eval_steps_per_second": 0.511,
      "step": 13
    },
    {
      "epoch": 0.2857142857142857,
      "grad_norm": 0.5915237665176392,
      "learning_rate": 0.0003181818181818182,
      "loss": 1.0291,
      "step": 14
    },
    {
      "epoch": 0.2857142857142857,
      "eval_loss": 1.0997107028961182,
      "eval_runtime": 7.8857,
      "eval_samples_per_second": 7.355,
      "eval_steps_per_second": 0.507,
      "step": 14
    },
    {
      "epoch": 0.30612244897959184,
      "grad_norm": 0.29103440046310425,
      "learning_rate": 0.0003090909090909091,
      "loss": 0.9446,
      "step": 15
    },
    {
      "epoch": 0.30612244897959184,
      "eval_loss": 1.0747274160385132,
      "eval_runtime": 7.7626,
      "eval_samples_per_second": 7.472,
      "eval_steps_per_second": 0.515,
      "step": 15
    },
    {
      "epoch": 0.32653061224489793,
      "grad_norm": 0.49479591846466064,
      "learning_rate": 0.00030000000000000003,
      "loss": 0.9729,
      "step": 16
    },
    {
      "epoch": 0.32653061224489793,
      "eval_loss": 1.0710903406143188,
      "eval_runtime": 7.742,
      "eval_samples_per_second": 7.492,
      "eval_steps_per_second": 0.517,
      "step": 16
    },
    {
      "epoch": 0.3469387755102041,
      "grad_norm": 0.33930784463882446,
      "learning_rate": 0.0002909090909090909,
      "loss": 1.0033,
      "step": 17
    },
    {
      "epoch": 0.3469387755102041,
      "eval_loss": 1.0736569166183472,
      "eval_runtime": 7.7392,
      "eval_samples_per_second": 7.494,
      "eval_steps_per_second": 0.517,
      "step": 17
    },
    {
      "epoch": 0.3673469387755102,
      "grad_norm": 0.14431457221508026,
      "learning_rate": 0.00028181818181818185,
      "loss": 0.9444,
      "step": 18
    },
    {
      "epoch": 0.3673469387755102,
      "eval_loss": 1.093230962753296,
      "eval_runtime": 7.7853,
      "eval_samples_per_second": 7.45,
      "eval_steps_per_second": 0.514,
      "step": 18
    },
    {
      "epoch": 0.3877551020408163,
      "grad_norm": 0.34998685121536255,
      "learning_rate": 0.00027272727272727274,
      "loss": 0.9754,
      "step": 19
    },
    {
      "epoch": 0.3877551020408163,
      "eval_loss": 1.0835151672363281,
      "eval_runtime": 7.7735,
      "eval_samples_per_second": 7.461,
      "eval_steps_per_second": 0.515,
      "step": 19
    },
    {
      "epoch": 0.40816326530612246,
      "grad_norm": 0.24697448313236237,
      "learning_rate": 0.0002636363636363636,
      "loss": 0.9219,
      "step": 20
    },
    {
      "epoch": 0.40816326530612246,
      "eval_loss": 1.0648112297058105,
      "eval_runtime": 7.7263,
      "eval_samples_per_second": 7.507,
      "eval_steps_per_second": 0.518,
      "step": 20
    },
    {
      "epoch": 0.42857142857142855,
      "grad_norm": 0.14658571779727936,
      "learning_rate": 0.00025454545454545456,
      "loss": 0.9978,
      "step": 21
    },
    {
      "epoch": 0.42857142857142855,
      "eval_loss": 1.0575088262557983,
      "eval_runtime": 7.7823,
      "eval_samples_per_second": 7.453,
      "eval_steps_per_second": 0.514,
      "step": 21
    },
    {
      "epoch": 0.4489795918367347,
      "grad_norm": 0.16511592268943787,
      "learning_rate": 0.00024545454545454545,
      "loss": 0.9567,
      "step": 22
    },
    {
      "epoch": 0.4489795918367347,
      "eval_loss": 1.055058479309082,
      "eval_runtime": 7.768,
      "eval_samples_per_second": 7.467,
      "eval_steps_per_second": 0.515,
      "step": 22
    },
    {
      "epoch": 0.46938775510204084,
      "grad_norm": 0.1519726663827896,
      "learning_rate": 0.0002363636363636364,
      "loss": 0.9364,
      "step": 23
    },
    {
      "epoch": 0.46938775510204084,
      "eval_loss": 1.053310751914978,
      "eval_runtime": 7.7426,
      "eval_samples_per_second": 7.491,
      "eval_steps_per_second": 0.517,
      "step": 23
    },
    {
      "epoch": 0.4897959183673469,
      "grad_norm": 0.1743665635585785,
      "learning_rate": 0.0002272727272727273,
      "loss": 0.9467,
      "step": 24
    },
    {
      "epoch": 0.4897959183673469,
      "eval_loss": 1.0527348518371582,
      "eval_runtime": 7.7567,
      "eval_samples_per_second": 7.477,
      "eval_steps_per_second": 0.516,
      "step": 24
    },
    {
      "epoch": 0.5102040816326531,
      "grad_norm": 0.13117213547229767,
      "learning_rate": 0.00021818181818181818,
      "loss": 0.9849,
      "step": 25
    },
    {
      "epoch": 0.5102040816326531,
      "eval_loss": 1.0531996488571167,
      "eval_runtime": 7.8342,
      "eval_samples_per_second": 7.403,
      "eval_steps_per_second": 0.511,
      "step": 25
    },
    {
      "epoch": 0.5306122448979592,
      "grad_norm": 0.15746016800403595,
      "learning_rate": 0.0002090909090909091,
      "loss": 0.9442,
      "step": 26
    },
    {
      "epoch": 0.5306122448979592,
      "eval_loss": 1.0544291734695435,
      "eval_runtime": 7.7804,
      "eval_samples_per_second": 7.455,
      "eval_steps_per_second": 0.514,
      "step": 26
    },
    {
      "epoch": 0.5510204081632653,
      "grad_norm": 0.210020512342453,
      "learning_rate": 0.0002,
      "loss": 0.8907,
      "step": 27
    },
    {
      "epoch": 0.5510204081632653,
      "eval_loss": 1.0516557693481445,
      "eval_runtime": 7.7543,
      "eval_samples_per_second": 7.48,
      "eval_steps_per_second": 0.516,
      "step": 27
    },
    {
      "epoch": 0.5714285714285714,
      "grad_norm": 0.1952683925628662,
      "learning_rate": 0.00019090909090909092,
      "loss": 0.9028,
      "step": 28
    },
    {
      "epoch": 0.5714285714285714,
      "eval_loss": 1.0470374822616577,
      "eval_runtime": 7.7964,
      "eval_samples_per_second": 7.439,
      "eval_steps_per_second": 0.513,
      "step": 28
    },
    {
      "epoch": 0.5918367346938775,
      "grad_norm": 0.12711568176746368,
      "learning_rate": 0.00018181818181818183,
      "loss": 0.9236,
      "step": 29
    },
    {
      "epoch": 0.5918367346938775,
      "eval_loss": 1.0439784526824951,
      "eval_runtime": 7.7312,
      "eval_samples_per_second": 7.502,
      "eval_steps_per_second": 0.517,
      "step": 29
    },
    {
      "epoch": 0.6122448979591837,
      "grad_norm": 0.1451597958803177,
      "learning_rate": 0.00017272727272727275,
      "loss": 0.872,
      "step": 30
    },
    {
      "epoch": 0.6122448979591837,
      "eval_loss": 1.0427075624465942,
      "eval_runtime": 7.7694,
      "eval_samples_per_second": 7.465,
      "eval_steps_per_second": 0.515,
      "step": 30
    },
    {
      "epoch": 0.6326530612244898,
      "grad_norm": 0.13339708745479584,
      "learning_rate": 0.00016363636363636366,
      "loss": 0.9643,
      "step": 31
    },
    {
      "epoch": 0.6326530612244898,
      "eval_loss": 1.0416866540908813,
      "eval_runtime": 7.7544,
      "eval_samples_per_second": 7.48,
      "eval_steps_per_second": 0.516,
      "step": 31
    },
    {
      "epoch": 0.6530612244897959,
      "grad_norm": 0.12144018709659576,
      "learning_rate": 0.00015454545454545454,
      "loss": 0.9042,
      "step": 32
    },
    {
      "epoch": 0.6530612244897959,
      "eval_loss": 1.040906548500061,
      "eval_runtime": 7.7547,
      "eval_samples_per_second": 7.479,
      "eval_steps_per_second": 0.516,
      "step": 32
    },
    {
      "epoch": 0.673469387755102,
      "grad_norm": 0.12495947629213333,
      "learning_rate": 0.00014545454545454546,
      "loss": 1.0005,
      "step": 33
    },
    {
      "epoch": 0.673469387755102,
      "eval_loss": 1.040427803993225,
      "eval_runtime": 7.8674,
      "eval_samples_per_second": 7.372,
      "eval_steps_per_second": 0.508,
      "step": 33
    },
    {
      "epoch": 0.6938775510204082,
      "grad_norm": 0.11797568202018738,
      "learning_rate": 0.00013636363636363637,
      "loss": 0.8941,
      "step": 34
    },
    {
      "epoch": 0.6938775510204082,
      "eval_loss": 1.0402781963348389,
      "eval_runtime": 7.815,
      "eval_samples_per_second": 7.422,
      "eval_steps_per_second": 0.512,
      "step": 34
    },
    {
      "epoch": 0.7142857142857143,
      "grad_norm": 0.13842004537582397,
      "learning_rate": 0.00012727272727272728,
      "loss": 0.9387,
      "step": 35
    },
    {
      "epoch": 0.7142857142857143,
      "eval_loss": 1.0394043922424316,
      "eval_runtime": 7.8286,
      "eval_samples_per_second": 7.409,
      "eval_steps_per_second": 0.511,
      "step": 35
    },
    {
      "epoch": 0.7346938775510204,
      "grad_norm": 0.13169452548027039,
      "learning_rate": 0.0001181818181818182,
      "loss": 0.892,
      "step": 36
    },
    {
      "epoch": 0.7346938775510204,
      "eval_loss": 1.0382304191589355,
      "eval_runtime": 7.7509,
      "eval_samples_per_second": 7.483,
      "eval_steps_per_second": 0.516,
      "step": 36
    },
    {
      "epoch": 0.7551020408163265,
      "grad_norm": 0.12079770863056183,
      "learning_rate": 0.00010909090909090909,
      "loss": 0.9791,
      "step": 37
    },
    {
      "epoch": 0.7551020408163265,
      "eval_loss": 1.0367978811264038,
      "eval_runtime": 7.7475,
      "eval_samples_per_second": 7.486,
      "eval_steps_per_second": 0.516,
      "step": 37
    },
    {
      "epoch": 0.7755102040816326,
      "grad_norm": 0.13031338155269623,
      "learning_rate": 0.0001,
      "loss": 0.9312,
      "step": 38
    },
    {
      "epoch": 0.7755102040816326,
      "eval_loss": 1.0354607105255127,
      "eval_runtime": 7.8232,
      "eval_samples_per_second": 7.414,
      "eval_steps_per_second": 0.511,
      "step": 38
    },
    {
      "epoch": 0.7959183673469388,
      "grad_norm": 0.1265690177679062,
      "learning_rate": 9.090909090909092e-05,
      "loss": 0.9093,
      "step": 39
    },
    {
      "epoch": 0.7959183673469388,
      "eval_loss": 1.0342280864715576,
      "eval_runtime": 7.8639,
      "eval_samples_per_second": 7.375,
      "eval_steps_per_second": 0.509,
      "step": 39
    },
    {
      "epoch": 0.8163265306122449,
      "grad_norm": 0.12322778254747391,
      "learning_rate": 8.181818181818183e-05,
      "loss": 0.9181,
      "step": 40
    },
    {
      "epoch": 0.8163265306122449,
      "eval_loss": 1.0333940982818604,
      "eval_runtime": 7.8466,
      "eval_samples_per_second": 7.392,
      "eval_steps_per_second": 0.51,
      "step": 40
    },
    {
      "epoch": 0.8367346938775511,
      "grad_norm": 0.13320668041706085,
      "learning_rate": 7.272727272727273e-05,
      "loss": 0.8642,
      "step": 41
    },
    {
      "epoch": 0.8367346938775511,
      "eval_loss": 1.0329786539077759,
      "eval_runtime": 7.7007,
      "eval_samples_per_second": 7.532,
      "eval_steps_per_second": 0.519,
      "step": 41
    },
    {
      "epoch": 0.8571428571428571,
      "grad_norm": 0.1269012689590454,
      "learning_rate": 6.363636363636364e-05,
      "loss": 0.9378,
      "step": 42
    },
    {
      "epoch": 0.8571428571428571,
      "eval_loss": 1.0328903198242188,
      "eval_runtime": 7.8286,
      "eval_samples_per_second": 7.409,
      "eval_steps_per_second": 0.511,
      "step": 42
    },
    {
      "epoch": 0.8775510204081632,
      "grad_norm": 0.13395856320858002,
      "learning_rate": 5.4545454545454546e-05,
      "loss": 0.8944,
      "step": 43
    },
    {
      "epoch": 0.8775510204081632,
      "eval_loss": 1.0329813957214355,
      "eval_runtime": 7.741,
      "eval_samples_per_second": 7.493,
      "eval_steps_per_second": 0.517,
      "step": 43
    },
    {
      "epoch": 0.8979591836734694,
      "grad_norm": 0.13417017459869385,
      "learning_rate": 4.545454545454546e-05,
      "loss": 0.8439,
      "step": 44
    },
    {
      "epoch": 0.8979591836734694,
      "eval_loss": 1.033360242843628,
      "eval_runtime": 7.8192,
      "eval_samples_per_second": 7.418,
      "eval_steps_per_second": 0.512,
      "step": 44
    },
    {
      "epoch": 0.9183673469387755,
      "grad_norm": 0.12144576758146286,
      "learning_rate": 3.6363636363636364e-05,
      "loss": 0.8755,
      "step": 45
    },
    {
      "epoch": 0.9183673469387755,
      "eval_loss": 1.033748745918274,
      "eval_runtime": 7.7165,
      "eval_samples_per_second": 7.516,
      "eval_steps_per_second": 0.518,
      "step": 45
    },
    {
      "epoch": 0.9387755102040817,
      "grad_norm": 0.1293192207813263,
      "learning_rate": 2.7272727272727273e-05,
      "loss": 0.882,
      "step": 46
    },
    {
      "epoch": 0.9387755102040817,
      "eval_loss": 1.034084677696228,
      "eval_runtime": 7.6836,
      "eval_samples_per_second": 7.549,
      "eval_steps_per_second": 0.521,
      "step": 46
    },
    {
      "epoch": 0.9591836734693877,
      "grad_norm": 0.13552704453468323,
      "learning_rate": 1.8181818181818182e-05,
      "loss": 0.9249,
      "step": 47
    },
    {
      "epoch": 0.9591836734693877,
      "eval_loss": 1.0341947078704834,
      "eval_runtime": 7.7207,
      "eval_samples_per_second": 7.512,
      "eval_steps_per_second": 0.518,
      "step": 47
    },
    {
      "epoch": 0.9795918367346939,
      "grad_norm": 0.11890257894992828,
      "learning_rate": 9.090909090909091e-06,
      "loss": 0.9159,
      "step": 48
    },
    {
      "epoch": 0.9795918367346939,
      "eval_loss": 1.0341988801956177,
      "eval_runtime": 7.7862,
      "eval_samples_per_second": 7.449,
      "eval_steps_per_second": 0.514,
      "step": 48
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.25241100788116455,
      "learning_rate": 0.0,
      "loss": 0.8313,
      "step": 49
    },
    {
      "epoch": 1.0,
      "eval_loss": 1.0341778993606567,
      "eval_runtime": 7.7578,
      "eval_samples_per_second": 7.476,
      "eval_steps_per_second": 0.516,
      "step": 49
    },
    {
      "epoch": 1.0,
      "step": 49,
      "total_flos": 3.583414438251725e+16,
      "train_loss": 1.1717130614786733,
      "train_runtime": 1068.0664,
      "train_samples_per_second": 1.444,
      "train_steps_per_second": 0.046
    }
  ],
  "logging_steps": 1,
  "max_steps": 49,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "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": 3.583414438251725e+16,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}