{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 4.0,
  "eval_steps": 500,
  "global_step": 420,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.047619047619047616,
      "grad_norm": 1.4257669918070304,
      "learning_rate": 5.9523809523809525e-06,
      "loss": 0.5346,
      "mean_token_accuracy": 0.8659344732761383,
      "step": 5
    },
    {
      "epoch": 0.09523809523809523,
      "grad_norm": 0.7751540276308181,
      "learning_rate": 1.1904761904761905e-05,
      "loss": 0.4535,
      "mean_token_accuracy": 0.8758046269416809,
      "step": 10
    },
    {
      "epoch": 0.14285714285714285,
      "grad_norm": 0.3688156015884892,
      "learning_rate": 1.785714285714286e-05,
      "loss": 0.4071,
      "mean_token_accuracy": 0.8853001713752746,
      "step": 15
    },
    {
      "epoch": 0.19047619047619047,
      "grad_norm": 0.2571090114484444,
      "learning_rate": 2.380952380952381e-05,
      "loss": 0.3811,
      "mean_token_accuracy": 0.8898558735847473,
      "step": 20
    },
    {
      "epoch": 0.23809523809523808,
      "grad_norm": 0.2445335136114927,
      "learning_rate": 2.9761904761904762e-05,
      "loss": 0.3714,
      "mean_token_accuracy": 0.8922576308250427,
      "step": 25
    },
    {
      "epoch": 0.2857142857142857,
      "grad_norm": 0.25497240173448055,
      "learning_rate": 3.571428571428572e-05,
      "loss": 0.3621,
      "mean_token_accuracy": 0.8938062608242034,
      "step": 30
    },
    {
      "epoch": 0.3333333333333333,
      "grad_norm": 0.22758565353000912,
      "learning_rate": 4.166666666666667e-05,
      "loss": 0.3572,
      "mean_token_accuracy": 0.8950600028038025,
      "step": 35
    },
    {
      "epoch": 0.38095238095238093,
      "grad_norm": 0.21108211135784025,
      "learning_rate": 4.761904761904762e-05,
      "loss": 0.3463,
      "mean_token_accuracy": 0.8972749292850495,
      "step": 40
    },
    {
      "epoch": 0.42857142857142855,
      "grad_norm": 0.2207672207804811,
      "learning_rate": 4.999300659501837e-05,
      "loss": 0.3443,
      "mean_token_accuracy": 0.8975539267063141,
      "step": 45
    },
    {
      "epoch": 0.47619047619047616,
      "grad_norm": 0.21926886992248296,
      "learning_rate": 4.995028486261366e-05,
      "loss": 0.3378,
      "mean_token_accuracy": 0.899033111333847,
      "step": 50
    },
    {
      "epoch": 0.5238095238095238,
      "grad_norm": 0.224235174049935,
      "learning_rate": 4.986880029700503e-05,
      "loss": 0.3403,
      "mean_token_accuracy": 0.8979385972023011,
      "step": 55
    },
    {
      "epoch": 0.5714285714285714,
      "grad_norm": 0.21434998232271477,
      "learning_rate": 4.974869359006539e-05,
      "loss": 0.3297,
      "mean_token_accuracy": 0.9011363804340362,
      "step": 60
    },
    {
      "epoch": 0.6190476190476191,
      "grad_norm": 0.22126533026421416,
      "learning_rate": 4.959017211895173e-05,
      "loss": 0.3329,
      "mean_token_accuracy": 0.8999340653419494,
      "step": 65
    },
    {
      "epoch": 0.6666666666666666,
      "grad_norm": 0.2122647717862151,
      "learning_rate": 4.9393509588046036e-05,
      "loss": 0.3306,
      "mean_token_accuracy": 0.9009318828582764,
      "step": 70
    },
    {
      "epoch": 0.7142857142857143,
      "grad_norm": 0.20764297035844217,
      "learning_rate": 4.915904555637527e-05,
      "loss": 0.33,
      "mean_token_accuracy": 0.9004346609115601,
      "step": 75
    },
    {
      "epoch": 0.7619047619047619,
      "grad_norm": 0.20816077188373128,
      "learning_rate": 4.8887184851326155e-05,
      "loss": 0.3202,
      "mean_token_accuracy": 0.9034851253032684,
      "step": 80
    },
    {
      "epoch": 0.8095238095238095,
      "grad_norm": 0.22148078953488418,
      "learning_rate": 4.8578396869667095e-05,
      "loss": 0.3252,
      "mean_token_accuracy": 0.9017885386943817,
      "step": 85
    },
    {
      "epoch": 0.8571428571428571,
      "grad_norm": 0.21406123617550696,
      "learning_rate": 4.823321476708417e-05,
      "loss": 0.3207,
      "mean_token_accuracy": 0.9030977070331574,
      "step": 90
    },
    {
      "epoch": 0.9047619047619048,
      "grad_norm": 0.20063414494959075,
      "learning_rate": 4.7852234537630435e-05,
      "loss": 0.3151,
      "mean_token_accuracy": 0.9040939033031463,
      "step": 95
    },
    {
      "epoch": 0.9523809523809523,
      "grad_norm": 0.20157178980104304,
      "learning_rate": 4.7436113984678024e-05,
      "loss": 0.3133,
      "mean_token_accuracy": 0.9044515311717987,
      "step": 100
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.21946571056086206,
      "learning_rate": 4.6985571585149876e-05,
      "loss": 0.3088,
      "mean_token_accuracy": 0.9057298839092255,
      "step": 105
    },
    {
      "epoch": 1.0476190476190477,
      "grad_norm": 0.22243550466680892,
      "learning_rate": 4.650138524899193e-05,
      "loss": 0.2589,
      "mean_token_accuracy": 0.9184723615646362,
      "step": 110
    },
    {
      "epoch": 1.0952380952380953,
      "grad_norm": 0.21782596690450573,
      "learning_rate": 4.5984390976027946e-05,
      "loss": 0.2467,
      "mean_token_accuracy": 0.9217934429645538,
      "step": 115
    },
    {
      "epoch": 1.1428571428571428,
      "grad_norm": 0.20047942498513227,
      "learning_rate": 4.5435481412515755e-05,
      "loss": 0.2458,
      "mean_token_accuracy": 0.9213689804077149,
      "step": 120
    },
    {
      "epoch": 1.1904761904761905,
      "grad_norm": 0.20323781517968,
      "learning_rate": 4.4855604309897496e-05,
      "loss": 0.2533,
      "mean_token_accuracy": 0.9193067252635956,
      "step": 125
    },
    {
      "epoch": 1.2380952380952381,
      "grad_norm": 0.2089213813676996,
      "learning_rate": 4.4245760888404665e-05,
      "loss": 0.2451,
      "mean_token_accuracy": 0.921397477388382,
      "step": 130
    },
    {
      "epoch": 1.2857142857142856,
      "grad_norm": 0.20871576941556397,
      "learning_rate": 4.360700410834367e-05,
      "loss": 0.2487,
      "mean_token_accuracy": 0.9207526803016662,
      "step": 135
    },
    {
      "epoch": 1.3333333333333333,
      "grad_norm": 0.20056126493311283,
      "learning_rate": 4.294043685204651e-05,
      "loss": 0.2361,
      "mean_token_accuracy": 0.924414473772049,
      "step": 140
    },
    {
      "epoch": 1.380952380952381,
      "grad_norm": 0.2115275835238231,
      "learning_rate": 4.224721001962573e-05,
      "loss": 0.2409,
      "mean_token_accuracy": 0.9230122208595276,
      "step": 145
    },
    {
      "epoch": 1.4285714285714286,
      "grad_norm": 0.20541500240177765,
      "learning_rate": 4.1528520541821506e-05,
      "loss": 0.2398,
      "mean_token_accuracy": 0.9232278406620026,
      "step": 150
    },
    {
      "epoch": 1.4761904761904763,
      "grad_norm": 0.20798560704125774,
      "learning_rate": 4.078560931337187e-05,
      "loss": 0.2471,
      "mean_token_accuracy": 0.9215823113918304,
      "step": 155
    },
    {
      "epoch": 1.5238095238095237,
      "grad_norm": 0.1939929705798728,
      "learning_rate": 4.001975905047442e-05,
      "loss": 0.2401,
      "mean_token_accuracy": 0.923470401763916,
      "step": 160
    },
    {
      "epoch": 1.5714285714285714,
      "grad_norm": 0.21344717934526472,
      "learning_rate": 3.923229207603871e-05,
      "loss": 0.2396,
      "mean_token_accuracy": 0.9238272488117218,
      "step": 165
    },
    {
      "epoch": 1.619047619047619,
      "grad_norm": 0.2191275580698787,
      "learning_rate": 3.842456803655342e-05,
      "loss": 0.2406,
      "mean_token_accuracy": 0.923265916109085,
      "step": 170
    },
    {
      "epoch": 1.6666666666666665,
      "grad_norm": 0.19421469398044988,
      "learning_rate": 3.75979815545104e-05,
      "loss": 0.2359,
      "mean_token_accuracy": 0.924458909034729,
      "step": 175
    },
    {
      "epoch": 1.7142857142857144,
      "grad_norm": 0.19470949781451338,
      "learning_rate": 3.6753959820438764e-05,
      "loss": 0.2321,
      "mean_token_accuracy": 0.9256431758403778,
      "step": 180
    },
    {
      "epoch": 1.7619047619047619,
      "grad_norm": 0.1876708972166408,
      "learning_rate": 3.589396012870687e-05,
      "loss": 0.2291,
      "mean_token_accuracy": 0.9266167640686035,
      "step": 185
    },
    {
      "epoch": 1.8095238095238095,
      "grad_norm": 0.2147218447176946,
      "learning_rate": 3.5019467361346724e-05,
      "loss": 0.2329,
      "mean_token_accuracy": 0.925212299823761,
      "step": 190
    },
    {
      "epoch": 1.8571428571428572,
      "grad_norm": 0.19962238819866412,
      "learning_rate": 3.413199142424535e-05,
      "loss": 0.2296,
      "mean_token_accuracy": 0.9259015142917633,
      "step": 195
    },
    {
      "epoch": 1.9047619047619047,
      "grad_norm": 0.19475591470139603,
      "learning_rate": 3.3233064640129734e-05,
      "loss": 0.2277,
      "mean_token_accuracy": 0.92720667719841,
      "step": 200
    },
    {
      "epoch": 1.9523809523809523,
      "grad_norm": 0.20861501443310124,
      "learning_rate": 3.232423910284672e-05,
      "loss": 0.2284,
      "mean_token_accuracy": 0.9266561150550843,
      "step": 205
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.19761114102987562,
      "learning_rate": 3.140708399750594e-05,
      "loss": 0.2291,
      "mean_token_accuracy": 0.9258988976478577,
      "step": 210
    },
    {
      "epoch": 2.0476190476190474,
      "grad_norm": 0.24523566793415144,
      "learning_rate": 3.048318289111279e-05,
      "loss": 0.1563,
      "mean_token_accuracy": 0.9481804072856903,
      "step": 215
    },
    {
      "epoch": 2.0952380952380953,
      "grad_norm": 0.204067402376235,
      "learning_rate": 2.955413099836959e-05,
      "loss": 0.1542,
      "mean_token_accuracy": 0.9486050844192505,
      "step": 220
    },
    {
      "epoch": 2.142857142857143,
      "grad_norm": 0.19593572116438304,
      "learning_rate": 2.8621532427365687e-05,
      "loss": 0.1553,
      "mean_token_accuracy": 0.9480521976947784,
      "step": 225
    },
    {
      "epoch": 2.1904761904761907,
      "grad_norm": 0.19800587405132705,
      "learning_rate": 2.7686997409912192e-05,
      "loss": 0.1566,
      "mean_token_accuracy": 0.9476682603359222,
      "step": 230
    },
    {
      "epoch": 2.238095238095238,
      "grad_norm": 0.19599017566764934,
      "learning_rate": 2.6752139521303403e-05,
      "loss": 0.1525,
      "mean_token_accuracy": 0.9489890992641449,
      "step": 235
    },
    {
      "epoch": 2.2857142857142856,
      "grad_norm": 0.1957327582988075,
      "learning_rate": 2.5818572894305453e-05,
      "loss": 0.1515,
      "mean_token_accuracy": 0.9491177141666413,
      "step": 240
    },
    {
      "epoch": 2.3333333333333335,
      "grad_norm": 0.19017281050653162,
      "learning_rate": 2.4887909432182316e-05,
      "loss": 0.1512,
      "mean_token_accuracy": 0.9494254469871521,
      "step": 245
    },
    {
      "epoch": 2.380952380952381,
      "grad_norm": 0.19356138909333898,
      "learning_rate": 2.3961756025571336e-05,
      "loss": 0.1535,
      "mean_token_accuracy": 0.9488085567951202,
      "step": 250
    },
    {
      "epoch": 2.4285714285714284,
      "grad_norm": 0.21519917588614698,
      "learning_rate": 2.304171177801356e-05,
      "loss": 0.1521,
      "mean_token_accuracy": 0.9488667666912078,
      "step": 255
    },
    {
      "epoch": 2.4761904761904763,
      "grad_norm": 0.18356462457323924,
      "learning_rate": 2.2129365244929402e-05,
      "loss": 0.1503,
      "mean_token_accuracy": 0.9497200846672058,
      "step": 260
    },
    {
      "epoch": 2.5238095238095237,
      "grad_norm": 0.19727737940368512,
      "learning_rate": 2.1226291690806715e-05,
      "loss": 0.1502,
      "mean_token_accuracy": 0.9498027801513672,
      "step": 265
    },
    {
      "epoch": 2.571428571428571,
      "grad_norm": 0.18491161543383636,
      "learning_rate": 2.0334050369337104e-05,
      "loss": 0.1493,
      "mean_token_accuracy": 0.9500907003879547,
      "step": 270
    },
    {
      "epoch": 2.619047619047619,
      "grad_norm": 0.180689936763116,
      "learning_rate": 1.945418183119656e-05,
      "loss": 0.1461,
      "mean_token_accuracy": 0.9510073781013488,
      "step": 275
    },
    {
      "epoch": 2.6666666666666665,
      "grad_norm": 0.18357579016167094,
      "learning_rate": 1.8588205264118974e-05,
      "loss": 0.1522,
      "mean_token_accuracy": 0.9489753544330597,
      "step": 280
    },
    {
      "epoch": 2.7142857142857144,
      "grad_norm": 0.17455870908089519,
      "learning_rate": 1.7737615869854944e-05,
      "loss": 0.1476,
      "mean_token_accuracy": 0.9504713833332061,
      "step": 285
    },
    {
      "epoch": 2.761904761904762,
      "grad_norm": 0.1895734172134673,
      "learning_rate": 1.6903882282545055e-05,
      "loss": 0.1471,
      "mean_token_accuracy": 0.9508952736854553,
      "step": 290
    },
    {
      "epoch": 2.8095238095238093,
      "grad_norm": 0.1814138039625087,
      "learning_rate": 1.6088444032964923e-05,
      "loss": 0.1471,
      "mean_token_accuracy": 0.9506540060043335,
      "step": 295
    },
    {
      "epoch": 2.857142857142857,
      "grad_norm": 0.18505343239566618,
      "learning_rate": 1.5292709063020415e-05,
      "loss": 0.1453,
      "mean_token_accuracy": 0.9513262331485748,
      "step": 300
    },
    {
      "epoch": 2.9047619047619047,
      "grad_norm": 0.1818729025239862,
      "learning_rate": 1.4518051294784384e-05,
      "loss": 0.1427,
      "mean_token_accuracy": 0.9522208392620086,
      "step": 305
    },
    {
      "epoch": 2.9523809523809526,
      "grad_norm": 0.18282477277115136,
      "learning_rate": 1.3765808258272334e-05,
      "loss": 0.1457,
      "mean_token_accuracy": 0.9513007164001465,
      "step": 310
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.17250893276801252,
      "learning_rate": 1.3037278782052863e-05,
      "loss": 0.1448,
      "mean_token_accuracy": 0.9508862257003784,
      "step": 315
    },
    {
      "epoch": 3.0476190476190474,
      "grad_norm": 0.21843020721408246,
      "learning_rate": 1.2333720750680403e-05,
      "loss": 0.0972,
      "mean_token_accuracy": 0.9678010582923889,
      "step": 320
    },
    {
      "epoch": 3.0952380952380953,
      "grad_norm": 0.18713956774840773,
      "learning_rate": 1.1656348932822086e-05,
      "loss": 0.0951,
      "mean_token_accuracy": 0.9680294811725616,
      "step": 325
    },
    {
      "epoch": 3.142857142857143,
      "grad_norm": 0.17099224072656777,
      "learning_rate": 1.1006332883828913e-05,
      "loss": 0.0928,
      "mean_token_accuracy": 0.9688285231590271,
      "step": 330
    },
    {
      "epoch": 3.1904761904761907,
      "grad_norm": 0.1889152084893897,
      "learning_rate": 1.0384794926372563e-05,
      "loss": 0.0945,
      "mean_token_accuracy": 0.9682107090950012,
      "step": 335
    },
    {
      "epoch": 3.238095238095238,
      "grad_norm": 0.16909157911032757,
      "learning_rate": 9.792808212634502e-06,
      "loss": 0.0911,
      "mean_token_accuracy": 0.9692863464355469,
      "step": 340
    },
    {
      "epoch": 3.2857142857142856,
      "grad_norm": 0.16462195486055214,
      "learning_rate": 9.231394871393228e-06,
      "loss": 0.0926,
      "mean_token_accuracy": 0.9688502609729767,
      "step": 345
    },
    {
      "epoch": 3.3333333333333335,
      "grad_norm": 0.16694354195202218,
      "learning_rate": 8.701524243208935e-06,
      "loss": 0.0941,
      "mean_token_accuracy": 0.968300586938858,
      "step": 350
    },
    {
      "epoch": 3.380952380952381,
      "grad_norm": 0.1692222626696623,
      "learning_rate": 8.204111206752663e-06,
      "loss": 0.0918,
      "mean_token_accuracy": 0.9689784705638885,
      "step": 355
    },
    {
      "epoch": 3.4285714285714284,
      "grad_norm": 0.17221192973282243,
      "learning_rate": 7.740014599169857e-06,
      "loss": 0.0904,
      "mean_token_accuracy": 0.9694355607032776,
      "step": 360
    },
    {
      "epoch": 3.4761904761904763,
      "grad_norm": 0.1741810830817914,
      "learning_rate": 7.3100357332055624e-06,
      "loss": 0.0896,
      "mean_token_accuracy": 0.9696675717830658,
      "step": 365
    },
    {
      "epoch": 3.5238095238095237,
      "grad_norm": 0.17501221405018258,
      "learning_rate": 6.914917013651723e-06,
      "loss": 0.092,
      "mean_token_accuracy": 0.968971711397171,
      "step": 370
    },
    {
      "epoch": 3.571428571428571,
      "grad_norm": 0.16570291902290668,
      "learning_rate": 6.555340655505407e-06,
      "loss": 0.0917,
      "mean_token_accuracy": 0.9693442165851593,
      "step": 375
    },
    {
      "epoch": 3.619047619047619,
      "grad_norm": 0.15923829992676775,
      "learning_rate": 6.231927506051192e-06,
      "loss": 0.0915,
      "mean_token_accuracy": 0.969234949350357,
      "step": 380
    },
    {
      "epoch": 3.6666666666666665,
      "grad_norm": 0.17338149522120277,
      "learning_rate": 5.9452359729015004e-06,
      "loss": 0.0907,
      "mean_token_accuracy": 0.9692964613437652,
      "step": 385
    },
    {
      "epoch": 3.7142857142857144,
      "grad_norm": 0.17158993366626976,
      "learning_rate": 5.695761059845749e-06,
      "loss": 0.0902,
      "mean_token_accuracy": 0.9696565389633178,
      "step": 390
    },
    {
      "epoch": 3.761904761904762,
      "grad_norm": 0.17688545350709856,
      "learning_rate": 5.483933512173022e-06,
      "loss": 0.0918,
      "mean_token_accuracy": 0.9692005813121796,
      "step": 395
    },
    {
      "epoch": 3.8095238095238093,
      "grad_norm": 0.1621556558547238,
      "learning_rate": 5.310119072943991e-06,
      "loss": 0.0892,
      "mean_token_accuracy": 0.9700611054897308,
      "step": 400
    },
    {
      "epoch": 3.857142857142857,
      "grad_norm": 0.16839822104636895,
      "learning_rate": 5.174617851496128e-06,
      "loss": 0.0904,
      "mean_token_accuracy": 0.9694457828998566,
      "step": 405
    },
    {
      "epoch": 3.9047619047619047,
      "grad_norm": 0.16305557552247782,
      "learning_rate": 5.077663805272652e-06,
      "loss": 0.0902,
      "mean_token_accuracy": 0.9695923388004303,
      "step": 410
    },
    {
      "epoch": 3.9523809523809526,
      "grad_norm": 0.16067720695857812,
      "learning_rate": 5.019424335869808e-06,
      "loss": 0.0903,
      "mean_token_accuracy": 0.9695243299007416,
      "step": 415
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.1603132269708069,
      "learning_rate": 5e-06,
      "loss": 0.0903,
      "mean_token_accuracy": 0.9692174196243286,
      "step": 420
    },
    {
      "epoch": 4.0,
      "step": 420,
      "total_flos": 216761967837184.0,
      "train_loss": 0.20915592369579133,
      "train_runtime": 2440.0132,
      "train_samples_per_second": 2.749,
      "train_steps_per_second": 0.172
    }
  ],
  "logging_steps": 5,
  "max_steps": 420,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 216761967837184.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}