{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.000169891366606153,
  "eval_steps": 10,
  "global_step": 70,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 2.4270195229450424e-06,
      "grad_norm": 1.988567590713501,
      "learning_rate": 6.666666649834951e-06,
      "loss": 2.6244,
      "step": 1
    },
    {
      "epoch": 2.4270195229450424e-06,
      "eval_loss": 3.429434299468994,
      "eval_runtime": 46.9267,
      "eval_samples_per_second": 0.448,
      "eval_steps_per_second": 0.448,
      "step": 1
    },
    {
      "epoch": 4.854039045890085e-06,
      "grad_norm": 3.1377198696136475,
      "learning_rate": 1.3333333299669903e-05,
      "loss": 3.3996,
      "step": 2
    },
    {
      "epoch": 7.281058568835128e-06,
      "grad_norm": 2.6447088718414307,
      "learning_rate": 1.9999999494757503e-05,
      "loss": 3.4643,
      "step": 3
    },
    {
      "epoch": 9.70807809178017e-06,
      "grad_norm": 2.257084369659424,
      "learning_rate": 2.6666666599339806e-05,
      "loss": 3.1453,
      "step": 4
    },
    {
      "epoch": 1.2135097614725214e-05,
      "grad_norm": 2.7705202102661133,
      "learning_rate": 3.333333370392211e-05,
      "loss": 3.69,
      "step": 5
    },
    {
      "epoch": 1.4562117137670256e-05,
      "grad_norm": 2.6461124420166016,
      "learning_rate": 3.9999998989515007e-05,
      "loss": 3.7107,
      "step": 6
    },
    {
      "epoch": 1.69891366606153e-05,
      "grad_norm": 3.9613280296325684,
      "learning_rate": 4.6666664275107905e-05,
      "loss": 4.2353,
      "step": 7
    },
    {
      "epoch": 1.941615618356034e-05,
      "grad_norm": 2.75366473197937,
      "learning_rate": 5.333333319867961e-05,
      "loss": 2.5853,
      "step": 8
    },
    {
      "epoch": 2.1843175706505383e-05,
      "grad_norm": 2.350839853286743,
      "learning_rate": 6.000000212225132e-05,
      "loss": 3.4014,
      "step": 9
    },
    {
      "epoch": 2.4270195229450427e-05,
      "grad_norm": 2.133833169937134,
      "learning_rate": 6.666666740784422e-05,
      "loss": 3.0469,
      "step": 10
    },
    {
      "epoch": 2.4270195229450427e-05,
      "eval_loss": 3.3984174728393555,
      "eval_runtime": 47.0885,
      "eval_samples_per_second": 0.446,
      "eval_steps_per_second": 0.446,
      "step": 10
    },
    {
      "epoch": 2.6697214752395468e-05,
      "grad_norm": 4.320294380187988,
      "learning_rate": 7.333333633141592e-05,
      "loss": 3.1139,
      "step": 11
    },
    {
      "epoch": 2.9124234275340512e-05,
      "grad_norm": 2.433712959289551,
      "learning_rate": 7.999999797903001e-05,
      "loss": 3.3114,
      "step": 12
    },
    {
      "epoch": 3.155125379828555e-05,
      "grad_norm": 2.948375701904297,
      "learning_rate": 8.666666690260172e-05,
      "loss": 3.4952,
      "step": 13
    },
    {
      "epoch": 3.39782733212306e-05,
      "grad_norm": 3.58933424949646,
      "learning_rate": 9.333332855021581e-05,
      "loss": 3.791,
      "step": 14
    },
    {
      "epoch": 3.640529284417564e-05,
      "grad_norm": 3.299588918685913,
      "learning_rate": 9.999999747378752e-05,
      "loss": 4.9645,
      "step": 15
    },
    {
      "epoch": 3.883231236712068e-05,
      "grad_norm": 2.313577890396118,
      "learning_rate": 0.00010666666639735922,
      "loss": 1.6593,
      "step": 16
    },
    {
      "epoch": 4.125933189006572e-05,
      "grad_norm": 2.2489659786224365,
      "learning_rate": 0.00011333332804497331,
      "loss": 1.8195,
      "step": 17
    },
    {
      "epoch": 4.3686351413010766e-05,
      "grad_norm": 2.143282175064087,
      "learning_rate": 0.00012000000424450263,
      "loss": 2.8584,
      "step": 18
    },
    {
      "epoch": 4.611337093595581e-05,
      "grad_norm": 6.221863746643066,
      "learning_rate": 0.00012666666589211673,
      "loss": 5.1277,
      "step": 19
    },
    {
      "epoch": 4.8540390458900854e-05,
      "grad_norm": 4.732879638671875,
      "learning_rate": 0.00013333333481568843,
      "loss": 3.8138,
      "step": 20
    },
    {
      "epoch": 4.8540390458900854e-05,
      "eval_loss": 3.1118509769439697,
      "eval_runtime": 47.1901,
      "eval_samples_per_second": 0.445,
      "eval_steps_per_second": 0.445,
      "step": 20
    },
    {
      "epoch": 5.096740998184589e-05,
      "grad_norm": 4.0478925704956055,
      "learning_rate": 0.0001399999891873449,
      "loss": 1.6834,
      "step": 21
    },
    {
      "epoch": 5.3394429504790936e-05,
      "grad_norm": 3.7350783348083496,
      "learning_rate": 0.00014666667266283184,
      "loss": 2.5659,
      "step": 22
    },
    {
      "epoch": 5.582144902773598e-05,
      "grad_norm": 3.017839193344116,
      "learning_rate": 0.00015333332703448832,
      "loss": 2.5162,
      "step": 23
    },
    {
      "epoch": 5.8248468550681024e-05,
      "grad_norm": 4.125138759613037,
      "learning_rate": 0.00015999999595806003,
      "loss": 3.7915,
      "step": 24
    },
    {
      "epoch": 6.067548807362606e-05,
      "grad_norm": 2.2610907554626465,
      "learning_rate": 0.00016666666488163173,
      "loss": 1.2178,
      "step": 25
    },
    {
      "epoch": 6.31025075965711e-05,
      "grad_norm": 4.597209453582764,
      "learning_rate": 0.00017333333380520344,
      "loss": 2.5644,
      "step": 26
    },
    {
      "epoch": 6.552952711951614e-05,
      "grad_norm": 3.9226925373077393,
      "learning_rate": 0.00017999998817685992,
      "loss": 2.5344,
      "step": 27
    },
    {
      "epoch": 6.79565466424612e-05,
      "grad_norm": 3.756793260574341,
      "learning_rate": 0.00018666665710043162,
      "loss": 2.3925,
      "step": 28
    },
    {
      "epoch": 7.038356616540623e-05,
      "grad_norm": 3.9993410110473633,
      "learning_rate": 0.00019333332602400333,
      "loss": 1.8358,
      "step": 29
    },
    {
      "epoch": 7.281058568835128e-05,
      "grad_norm": 6.380167484283447,
      "learning_rate": 0.00019999999494757503,
      "loss": 3.9897,
      "step": 30
    },
    {
      "epoch": 7.281058568835128e-05,
      "eval_loss": 2.611670970916748,
      "eval_runtime": 46.9072,
      "eval_samples_per_second": 0.448,
      "eval_steps_per_second": 0.448,
      "step": 30
    },
    {
      "epoch": 7.523760521129632e-05,
      "grad_norm": 4.4075164794921875,
      "learning_rate": 0.00019999999494757503,
      "loss": 2.755,
      "step": 31
    },
    {
      "epoch": 7.766462473424136e-05,
      "grad_norm": 5.7215046882629395,
      "learning_rate": 0.00019999999494757503,
      "loss": 3.0434,
      "step": 32
    },
    {
      "epoch": 8.009164425718641e-05,
      "grad_norm": 4.067835330963135,
      "learning_rate": 0.00019999999494757503,
      "loss": 2.2226,
      "step": 33
    },
    {
      "epoch": 8.251866378013144e-05,
      "grad_norm": 4.797438621520996,
      "learning_rate": 0.00019999999494757503,
      "loss": 1.5217,
      "step": 34
    },
    {
      "epoch": 8.49456833030765e-05,
      "grad_norm": 10.192020416259766,
      "learning_rate": 0.00019999999494757503,
      "loss": 2.7074,
      "step": 35
    },
    {
      "epoch": 8.737270282602153e-05,
      "grad_norm": 4.287100791931152,
      "learning_rate": 0.00019999999494757503,
      "loss": 1.7217,
      "step": 36
    },
    {
      "epoch": 8.979972234896657e-05,
      "grad_norm": 5.257225513458252,
      "learning_rate": 0.00019999999494757503,
      "loss": 2.0953,
      "step": 37
    },
    {
      "epoch": 9.222674187191162e-05,
      "grad_norm": 8.44377613067627,
      "learning_rate": 0.00019999999494757503,
      "loss": 2.3192,
      "step": 38
    },
    {
      "epoch": 9.465376139485666e-05,
      "grad_norm": 6.8432297706604,
      "learning_rate": 0.00019999999494757503,
      "loss": 2.459,
      "step": 39
    },
    {
      "epoch": 9.708078091780171e-05,
      "grad_norm": 5.657060623168945,
      "learning_rate": 0.00019999999494757503,
      "loss": 2.0209,
      "step": 40
    },
    {
      "epoch": 9.708078091780171e-05,
      "eval_loss": 2.3423023223876953,
      "eval_runtime": 46.9221,
      "eval_samples_per_second": 0.448,
      "eval_steps_per_second": 0.448,
      "step": 40
    },
    {
      "epoch": 9.950780044074675e-05,
      "grad_norm": 6.757113456726074,
      "learning_rate": 0.00019999999494757503,
      "loss": 3.2373,
      "step": 41
    },
    {
      "epoch": 0.00010193481996369178,
      "grad_norm": 3.9233195781707764,
      "learning_rate": 0.00019999999494757503,
      "loss": 1.269,
      "step": 42
    },
    {
      "epoch": 0.00010436183948663683,
      "grad_norm": 4.863090991973877,
      "learning_rate": 0.00019999999494757503,
      "loss": 2.1246,
      "step": 43
    },
    {
      "epoch": 0.00010678885900958187,
      "grad_norm": 5.010152816772461,
      "learning_rate": 0.00019999999494757503,
      "loss": 2.4953,
      "step": 44
    },
    {
      "epoch": 0.00010921587853252691,
      "grad_norm": 5.084195613861084,
      "learning_rate": 0.00019999999494757503,
      "loss": 2.3639,
      "step": 45
    },
    {
      "epoch": 0.00011164289805547196,
      "grad_norm": 4.2990193367004395,
      "learning_rate": 0.00019999999494757503,
      "loss": 1.5201,
      "step": 46
    },
    {
      "epoch": 0.000114069917578417,
      "grad_norm": 3.991379976272583,
      "learning_rate": 0.00019999999494757503,
      "loss": 1.3434,
      "step": 47
    },
    {
      "epoch": 0.00011649693710136205,
      "grad_norm": 3.651796579360962,
      "learning_rate": 0.00019999999494757503,
      "loss": 0.9977,
      "step": 48
    },
    {
      "epoch": 0.00011892395662430709,
      "grad_norm": 3.8026585578918457,
      "learning_rate": 0.00019999999494757503,
      "loss": 1.662,
      "step": 49
    },
    {
      "epoch": 0.00012135097614725212,
      "grad_norm": 4.1620612144470215,
      "learning_rate": 0.00019999999494757503,
      "loss": 2.3788,
      "step": 50
    },
    {
      "epoch": 0.00012135097614725212,
      "eval_loss": 2.2624521255493164,
      "eval_runtime": 46.8803,
      "eval_samples_per_second": 0.448,
      "eval_steps_per_second": 0.448,
      "step": 50
    },
    {
      "epoch": 0.00012377799567019716,
      "grad_norm": 3.4755067825317383,
      "learning_rate": 0.00019999999494757503,
      "loss": 1.2644,
      "step": 51
    },
    {
      "epoch": 0.0001262050151931422,
      "grad_norm": 3.064790725708008,
      "learning_rate": 0.00019999999494757503,
      "loss": 0.816,
      "step": 52
    },
    {
      "epoch": 0.00012863203471608726,
      "grad_norm": 6.472192764282227,
      "learning_rate": 0.00019999999494757503,
      "loss": 3.0402,
      "step": 53
    },
    {
      "epoch": 0.00013105905423903229,
      "grad_norm": 6.017213821411133,
      "learning_rate": 0.00019999999494757503,
      "loss": 1.9637,
      "step": 54
    },
    {
      "epoch": 0.00013348607376197734,
      "grad_norm": 2.969481945037842,
      "learning_rate": 0.00019999999494757503,
      "loss": 1.0158,
      "step": 55
    },
    {
      "epoch": 0.0001359130932849224,
      "grad_norm": 4.294132709503174,
      "learning_rate": 0.00019999999494757503,
      "loss": 2.329,
      "step": 56
    },
    {
      "epoch": 0.00013834011280786744,
      "grad_norm": 2.0407297611236572,
      "learning_rate": 0.00019999999494757503,
      "loss": 0.8446,
      "step": 57
    },
    {
      "epoch": 0.00014076713233081246,
      "grad_norm": 6.152376174926758,
      "learning_rate": 0.00019999999494757503,
      "loss": 1.7769,
      "step": 58
    },
    {
      "epoch": 0.0001431941518537575,
      "grad_norm": 5.564684867858887,
      "learning_rate": 0.00019999999494757503,
      "loss": 1.8718,
      "step": 59
    },
    {
      "epoch": 0.00014562117137670256,
      "grad_norm": 3.1191859245300293,
      "learning_rate": 0.00019999999494757503,
      "loss": 1.1267,
      "step": 60
    },
    {
      "epoch": 0.00014562117137670256,
      "eval_loss": 2.2274293899536133,
      "eval_runtime": 46.8696,
      "eval_samples_per_second": 0.448,
      "eval_steps_per_second": 0.448,
      "step": 60
    },
    {
      "epoch": 0.0001480481908996476,
      "grad_norm": 2.424983024597168,
      "learning_rate": 0.00019999999494757503,
      "loss": 1.631,
      "step": 61
    },
    {
      "epoch": 0.00015047521042259264,
      "grad_norm": 6.923261642456055,
      "learning_rate": 0.00019999999494757503,
      "loss": 3.3649,
      "step": 62
    },
    {
      "epoch": 0.0001529022299455377,
      "grad_norm": 3.311091184616089,
      "learning_rate": 0.00019999999494757503,
      "loss": 2.1499,
      "step": 63
    },
    {
      "epoch": 0.0001553292494684827,
      "grad_norm": 4.510260581970215,
      "learning_rate": 0.00019999999494757503,
      "loss": 1.5414,
      "step": 64
    },
    {
      "epoch": 0.00015775626899142776,
      "grad_norm": 4.838099956512451,
      "learning_rate": 0.00019999999494757503,
      "loss": 2.9635,
      "step": 65
    },
    {
      "epoch": 0.00016018328851437281,
      "grad_norm": 3.5294570922851562,
      "learning_rate": 0.00019999999494757503,
      "loss": 1.4734,
      "step": 66
    },
    {
      "epoch": 0.00016261030803731787,
      "grad_norm": 3.9977903366088867,
      "learning_rate": 0.00019999999494757503,
      "loss": 1.6908,
      "step": 67
    },
    {
      "epoch": 0.0001650373275602629,
      "grad_norm": 4.988344192504883,
      "learning_rate": 0.00019999999494757503,
      "loss": 0.6794,
      "step": 68
    },
    {
      "epoch": 0.00016746434708320794,
      "grad_norm": 1.6817117929458618,
      "learning_rate": 0.00019999999494757503,
      "loss": 1.4359,
      "step": 69
    },
    {
      "epoch": 0.000169891366606153,
      "grad_norm": 5.336698532104492,
      "learning_rate": 0.00019999999494757503,
      "loss": 3.0466,
      "step": 70
    },
    {
      "epoch": 0.000169891366606153,
      "eval_loss": 2.2040059566497803,
      "eval_runtime": 47.1606,
      "eval_samples_per_second": 0.445,
      "eval_steps_per_second": 0.445,
      "step": 70
    }
  ],
  "logging_steps": 1,
  "max_steps": 1236084,
  "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": 1528526725447680.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}