{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9974102848686645,
  "eval_steps": 100,
  "global_step": 337,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.014798372179060304,
      "grad_norm": 2.5319928724744716,
      "learning_rate": 2.9411764705882355e-06,
      "loss": 1.0934,
      "mean_token_accuracy": 0.710963543585844,
      "step": 5
    },
    {
      "epoch": 0.029596744358120607,
      "grad_norm": 1.4244723552718639,
      "learning_rate": 5.882352941176471e-06,
      "loss": 1.0769,
      "mean_token_accuracy": 0.7137204635526403,
      "step": 10
    },
    {
      "epoch": 0.04439511653718091,
      "grad_norm": 1.2482381526688695,
      "learning_rate": 8.823529411764707e-06,
      "loss": 0.9837,
      "mean_token_accuracy": 0.729854775606871,
      "step": 15
    },
    {
      "epoch": 0.059193488716241215,
      "grad_norm": 0.8840186116418455,
      "learning_rate": 1.1764705882352942e-05,
      "loss": 0.959,
      "mean_token_accuracy": 0.7300877725865283,
      "step": 20
    },
    {
      "epoch": 0.07399186089530152,
      "grad_norm": 0.7992527652377878,
      "learning_rate": 1.4705882352941179e-05,
      "loss": 0.8946,
      "mean_token_accuracy": 0.7439422970769893,
      "step": 25
    },
    {
      "epoch": 0.08879023307436182,
      "grad_norm": 0.7193278191451224,
      "learning_rate": 1.7647058823529414e-05,
      "loss": 0.8869,
      "mean_token_accuracy": 0.7441348945078509,
      "step": 30
    },
    {
      "epoch": 0.10358860525342212,
      "grad_norm": 0.620194874837715,
      "learning_rate": 1.9999462497359468e-05,
      "loss": 0.846,
      "mean_token_accuracy": 0.753654310037237,
      "step": 35
    },
    {
      "epoch": 0.11838697743248243,
      "grad_norm": 0.605036208438505,
      "learning_rate": 1.9980655971335944e-05,
      "loss": 0.8244,
      "mean_token_accuracy": 0.7586147304733527,
      "step": 40
    },
    {
      "epoch": 0.13318534961154274,
      "grad_norm": 0.5325835774353814,
      "learning_rate": 1.993503206718859e-05,
      "loss": 0.8234,
      "mean_token_accuracy": 0.7580317217224875,
      "step": 45
    },
    {
      "epoch": 0.14798372179060304,
      "grad_norm": 0.5922728516285096,
      "learning_rate": 1.986271337340182e-05,
      "loss": 0.8061,
      "mean_token_accuracy": 0.7622985291883017,
      "step": 50
    },
    {
      "epoch": 0.16278209396966334,
      "grad_norm": 0.5378220329694758,
      "learning_rate": 1.976389420563607e-05,
      "loss": 0.8147,
      "mean_token_accuracy": 0.759805779032566,
      "step": 55
    },
    {
      "epoch": 0.17758046614872364,
      "grad_norm": 0.5259675117637312,
      "learning_rate": 1.9638840084614182e-05,
      "loss": 0.8023,
      "mean_token_accuracy": 0.7627232625492987,
      "step": 60
    },
    {
      "epoch": 0.19237883832778394,
      "grad_norm": 0.5762209896140157,
      "learning_rate": 1.9487887022684336e-05,
      "loss": 0.792,
      "mean_token_accuracy": 0.7641335052616005,
      "step": 65
    },
    {
      "epoch": 0.20717721050684423,
      "grad_norm": 0.4808227147945767,
      "learning_rate": 1.9311440620976597e-05,
      "loss": 0.7952,
      "mean_token_accuracy": 0.7634592815770592,
      "step": 70
    },
    {
      "epoch": 0.22197558268590456,
      "grad_norm": 0.5001394930884526,
      "learning_rate": 1.9109974979578852e-05,
      "loss": 0.7766,
      "mean_token_accuracy": 0.768052524785652,
      "step": 75
    },
    {
      "epoch": 0.23677395486496486,
      "grad_norm": 0.5439973784585546,
      "learning_rate": 1.8884031423660492e-05,
      "loss": 0.7858,
      "mean_token_accuracy": 0.7659040802193098,
      "step": 80
    },
    {
      "epoch": 0.25157232704402516,
      "grad_norm": 0.4626471578447052,
      "learning_rate": 1.8634217048966638e-05,
      "loss": 0.7898,
      "mean_token_accuracy": 0.7641630963096735,
      "step": 85
    },
    {
      "epoch": 0.2663706992230855,
      "grad_norm": 0.4953360320003589,
      "learning_rate": 1.836120309059107e-05,
      "loss": 0.7653,
      "mean_token_accuracy": 0.7711456471218759,
      "step": 90
    },
    {
      "epoch": 0.28116907140214575,
      "grad_norm": 0.4920623807025407,
      "learning_rate": 1.8065723119410885e-05,
      "loss": 0.7729,
      "mean_token_accuracy": 0.7689427685113397,
      "step": 95
    },
    {
      "epoch": 0.2959674435812061,
      "grad_norm": 0.5103694322535611,
      "learning_rate": 1.77485710710289e-05,
      "loss": 0.7769,
      "mean_token_accuracy": 0.7670994908657515,
      "step": 100
    },
    {
      "epoch": 0.2959674435812061,
      "eval_loss": 0.7992530465126038,
      "eval_mean_token_accuracy": 0.7520903997377723,
      "eval_runtime": 37.942,
      "eval_samples_per_second": 3.4,
      "eval_steps_per_second": 0.237,
      "step": 100
    },
    {
      "epoch": 0.31076581576026635,
      "grad_norm": 0.5055132167951369,
      "learning_rate": 1.741059911251997e-05,
      "loss": 0.7743,
      "mean_token_accuracy": 0.7680788037040351,
      "step": 105
    },
    {
      "epoch": 0.3255641879393267,
      "grad_norm": 0.506430933020723,
      "learning_rate": 1.7052715352713076e-05,
      "loss": 0.7701,
      "mean_token_accuracy": 0.768703717503791,
      "step": 110
    },
    {
      "epoch": 0.340362560118387,
      "grad_norm": 0.5043957879151227,
      "learning_rate": 1.667588140216154e-05,
      "loss": 0.7809,
      "mean_token_accuracy": 0.7663542286457965,
      "step": 115
    },
    {
      "epoch": 0.3551609322974473,
      "grad_norm": 0.4950305678035583,
      "learning_rate": 1.628110978935756e-05,
      "loss": 0.7858,
      "mean_token_accuracy": 0.7642025057223836,
      "step": 120
    },
    {
      "epoch": 0.3699593044765076,
      "grad_norm": 0.48560370957454324,
      "learning_rate": 1.586946124013354e-05,
      "loss": 0.7721,
      "mean_token_accuracy": 0.7674092863960766,
      "step": 125
    },
    {
      "epoch": 0.38475767665556787,
      "grad_norm": 0.4630554031524683,
      "learning_rate": 1.5442041827560274e-05,
      "loss": 0.7701,
      "mean_token_accuracy": 0.7682268770584184,
      "step": 130
    },
    {
      "epoch": 0.3995560488346282,
      "grad_norm": 0.48760057057176687,
      "learning_rate": 1.5000000000000002e-05,
      "loss": 0.7784,
      "mean_token_accuracy": 0.7661367681955669,
      "step": 135
    },
    {
      "epoch": 0.41435442101368847,
      "grad_norm": 0.5733716326658232,
      "learning_rate": 1.4544523495299843e-05,
      "loss": 0.7479,
      "mean_token_accuracy": 0.7748393560242525,
      "step": 140
    },
    {
      "epoch": 0.4291527931927488,
      "grad_norm": 0.4674485766276026,
      "learning_rate": 1.4076836149416889e-05,
      "loss": 0.7624,
      "mean_token_accuracy": 0.7700554565613973,
      "step": 145
    },
    {
      "epoch": 0.4439511653718091,
      "grad_norm": 0.45437614184568614,
      "learning_rate": 1.3598194608050011e-05,
      "loss": 0.7615,
      "mean_token_accuracy": 0.771047339890566,
      "step": 150
    },
    {
      "epoch": 0.4587495375508694,
      "grad_norm": 0.5161211912106276,
      "learning_rate": 1.3109884950114007e-05,
      "loss": 0.7483,
      "mean_token_accuracy": 0.7738057195374108,
      "step": 155
    },
    {
      "epoch": 0.4735479097299297,
      "grad_norm": 0.48299986888606256,
      "learning_rate": 1.2613219232128608e-05,
      "loss": 0.7785,
      "mean_token_accuracy": 0.7653209447855766,
      "step": 160
    },
    {
      "epoch": 0.48834628190899,
      "grad_norm": 0.4972165634793764,
      "learning_rate": 1.2109531962807333e-05,
      "loss": 0.7584,
      "mean_token_accuracy": 0.7717189958506077,
      "step": 165
    },
    {
      "epoch": 0.5031446540880503,
      "grad_norm": 0.5038295262718444,
      "learning_rate": 1.1600176517318742e-05,
      "loss": 0.7697,
      "mean_token_accuracy": 0.7677580608314103,
      "step": 170
    },
    {
      "epoch": 0.5179430262671106,
      "grad_norm": 0.506972102629947,
      "learning_rate": 1.1086521500854746e-05,
      "loss": 0.7644,
      "mean_token_accuracy": 0.7690021577852479,
      "step": 175
    },
    {
      "epoch": 0.532741398446171,
      "grad_norm": 0.4607643963694159,
      "learning_rate": 1.0569947071276847e-05,
      "loss": 0.7508,
      "mean_token_accuracy": 0.7737279353788342,
      "step": 180
    },
    {
      "epoch": 0.5475397706252312,
      "grad_norm": 0.4690988373460707,
      "learning_rate": 1.0051841230721065e-05,
      "loss": 0.7571,
      "mean_token_accuracy": 0.7718751333144616,
      "step": 185
    },
    {
      "epoch": 0.5623381428042915,
      "grad_norm": 0.5230684589508905,
      "learning_rate": 9.533596096125826e-06,
      "loss": 0.7473,
      "mean_token_accuracy": 0.7735613336731262,
      "step": 190
    },
    {
      "epoch": 0.5771365149833518,
      "grad_norm": 0.4965020439000996,
      "learning_rate": 9.016604158703654e-06,
      "loss": 0.7366,
      "mean_token_accuracy": 0.7761915520691296,
      "step": 195
    },
    {
      "epoch": 0.5919348871624122,
      "grad_norm": 0.47470212551853064,
      "learning_rate": 8.502254542407186e-06,
      "loss": 0.7509,
      "mean_token_accuracy": 0.7725352138788509,
      "step": 200
    },
    {
      "epoch": 0.5919348871624122,
      "eval_loss": 0.7736743092536926,
      "eval_mean_token_accuracy": 0.7563969146133266,
      "eval_runtime": 36.8476,
      "eval_samples_per_second": 3.501,
      "eval_steps_per_second": 0.244,
      "step": 200
    },
    {
      "epoch": 0.6067332593414725,
      "grad_norm": 0.460020351716667,
      "learning_rate": 7.991929271442817e-06,
      "loss": 0.7406,
      "mean_token_accuracy": 0.7756438027345837,
      "step": 205
    },
    {
      "epoch": 0.6215316315205327,
      "grad_norm": 0.4298188029300264,
      "learning_rate": 7.48699955686089e-06,
      "loss": 0.7481,
      "mean_token_accuracy": 0.7735235687761747,
      "step": 210
    },
    {
      "epoch": 0.636330003699593,
      "grad_norm": 0.4966460976529198,
      "learning_rate": 6.988822112200157e-06,
      "loss": 0.775,
      "mean_token_accuracy": 0.7659616415859062,
      "step": 215
    },
    {
      "epoch": 0.6511283758786534,
      "grad_norm": 0.41853465241085613,
      "learning_rate": 6.498735508086094e-06,
      "loss": 0.7498,
      "mean_token_accuracy": 0.7732758807008967,
      "step": 220
    },
    {
      "epoch": 0.6659267480577137,
      "grad_norm": 0.4702312396884364,
      "learning_rate": 6.018056575578075e-06,
      "loss": 0.7592,
      "mean_token_accuracy": 0.7700622405709057,
      "step": 225
    },
    {
      "epoch": 0.680725120236774,
      "grad_norm": 0.42975452969024625,
      "learning_rate": 5.548076867929331e-06,
      "loss": 0.7378,
      "mean_token_accuracy": 0.7767314400830869,
      "step": 230
    },
    {
      "epoch": 0.6955234924158342,
      "grad_norm": 0.4108583246355971,
      "learning_rate": 5.090059190266779e-06,
      "loss": 0.7406,
      "mean_token_accuracy": 0.7765217843280351,
      "step": 235
    },
    {
      "epoch": 0.7103218645948945,
      "grad_norm": 0.45304370273166905,
      "learning_rate": 4.645234206515171e-06,
      "loss": 0.7316,
      "mean_token_accuracy": 0.7780036833059428,
      "step": 240
    },
    {
      "epoch": 0.7251202367739549,
      "grad_norm": 0.4232322187564541,
      "learning_rate": 4.214797132682597e-06,
      "loss": 0.744,
      "mean_token_accuracy": 0.7749563602812362,
      "step": 245
    },
    {
      "epoch": 0.7399186089530152,
      "grad_norm": 0.46727693453523017,
      "learning_rate": 3.799904525392251e-06,
      "loss": 0.7444,
      "mean_token_accuracy": 0.7747609637386101,
      "step": 250
    },
    {
      "epoch": 0.7547169811320755,
      "grad_norm": 0.4558283663119779,
      "learning_rate": 3.401671174289469e-06,
      "loss": 0.7487,
      "mean_token_accuracy": 0.7735000671597567,
      "step": 255
    },
    {
      "epoch": 0.7695153533111357,
      "grad_norm": 0.4473149950802802,
      "learning_rate": 3.021167106673928e-06,
      "loss": 0.7445,
      "mean_token_accuracy": 0.7743566228811749,
      "step": 260
    },
    {
      "epoch": 0.7843137254901961,
      "grad_norm": 0.4461233002812556,
      "learning_rate": 2.6594147124053983e-06,
      "loss": 0.7611,
      "mean_token_accuracy": 0.7693132343191504,
      "step": 265
    },
    {
      "epoch": 0.7991120976692564,
      "grad_norm": 0.4288144300029119,
      "learning_rate": 2.317385996808195e-06,
      "loss": 0.7445,
      "mean_token_accuracy": 0.7748136727030183,
      "step": 270
    },
    {
      "epoch": 0.8139104698483167,
      "grad_norm": 0.42462095484083673,
      "learning_rate": 1.9959999689556407e-06,
      "loss": 0.7434,
      "mean_token_accuracy": 0.7749245362781049,
      "step": 275
    },
    {
      "epoch": 0.8287088420273769,
      "grad_norm": 0.4063322872281588,
      "learning_rate": 1.6961201723520248e-06,
      "loss": 0.7449,
      "mean_token_accuracy": 0.7749680039942058,
      "step": 280
    },
    {
      "epoch": 0.8435072142064373,
      "grad_norm": 0.397264999375389,
      "learning_rate": 1.4185523646469822e-06,
      "loss": 0.7453,
      "mean_token_accuracy": 0.7743507033929469,
      "step": 285
    },
    {
      "epoch": 0.8583055863854976,
      "grad_norm": 0.4451350635097519,
      "learning_rate": 1.1640423526166987e-06,
      "loss": 0.742,
      "mean_token_accuracy": 0.7751791694228443,
      "step": 290
    },
    {
      "epoch": 0.8731039585645579,
      "grad_norm": 0.43301511965025696,
      "learning_rate": 9.332739882292752e-07,
      "loss": 0.7389,
      "mean_token_accuracy": 0.7757602051567191,
      "step": 295
    },
    {
      "epoch": 0.8879023307436182,
      "grad_norm": 0.4096833414120662,
      "learning_rate": 7.268673311786378e-07,
      "loss": 0.731,
      "mean_token_accuracy": 0.7785745911127538,
      "step": 300
    },
    {
      "epoch": 0.8879023307436182,
      "eval_loss": 0.7652355432510376,
      "eval_mean_token_accuracy": 0.7601538929893327,
      "eval_runtime": 37.6456,
      "eval_samples_per_second": 3.427,
      "eval_steps_per_second": 0.239,
      "step": 300
    },
    {
      "epoch": 0.9027007029226785,
      "grad_norm": 0.40132760343408064,
      "learning_rate": 5.453769828241872e-07,
      "loss": 0.7391,
      "mean_token_accuracy": 0.7762386779688674,
      "step": 305
    },
    {
      "epoch": 0.9174990751017388,
      "grad_norm": 0.3932064251891102,
      "learning_rate": 3.8929059601275463e-07,
      "loss": 0.712,
      "mean_token_accuracy": 0.7837027776073648,
      "step": 310
    },
    {
      "epoch": 0.9322974472807991,
      "grad_norm": 0.404346485308974,
      "learning_rate": 2.5902756478688674e-07,
      "loss": 0.74,
      "mean_token_accuracy": 0.7763130349280738,
      "step": 315
    },
    {
      "epoch": 0.9470958194598594,
      "grad_norm": 0.4014282081908607,
      "learning_rate": 1.5493789750014032e-07,
      "loss": 0.7373,
      "mean_token_accuracy": 0.7765861246677741,
      "step": 320
    },
    {
      "epoch": 0.9618941916389198,
      "grad_norm": 0.411881298783951,
      "learning_rate": 7.730127636723539e-08,
      "loss": 0.7236,
      "mean_token_accuracy": 0.7803048063818399,
      "step": 325
    },
    {
      "epoch": 0.97669256381798,
      "grad_norm": 0.38863002242821104,
      "learning_rate": 2.6326305976001054e-08,
      "loss": 0.7384,
      "mean_token_accuracy": 0.7764187201948197,
      "step": 330
    },
    {
      "epoch": 0.9914909359970403,
      "grad_norm": 0.4208440998084639,
      "learning_rate": 2.149952780321485e-09,
      "loss": 0.7368,
      "mean_token_accuracy": 0.7763385910458351,
      "step": 335
    },
    {
      "epoch": 0.9974102848686645,
      "mean_token_accuracy": 0.7691396070282631,
      "step": 337,
      "total_flos": 76774385909760.0,
      "train_loss": 0.7816258675974271,
      "train_runtime": 9729.0212,
      "train_samples_per_second": 2.223,
      "train_steps_per_second": 0.035
    }
  ],
  "logging_steps": 5,
  "max_steps": 337,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": false,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 76774385909760.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}