{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 8.162230671736376,
  "eval_steps": 100,
  "global_step": 800,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.050697084917617236,
      "grad_norm": 3.785501003265381,
      "learning_rate": 4.999948617395915e-05,
      "loss": 0.8619,
      "num_input_tokens_seen": 162496,
      "step": 5
    },
    {
      "epoch": 0.10139416983523447,
      "grad_norm": 1.068943738937378,
      "learning_rate": 4.9997944716957985e-05,
      "loss": 0.2854,
      "num_input_tokens_seen": 326848,
      "step": 10
    },
    {
      "epoch": 0.1520912547528517,
      "grad_norm": 0.6261367201805115,
      "learning_rate": 4.9995375692359755e-05,
      "loss": 0.2279,
      "num_input_tokens_seen": 491648,
      "step": 15
    },
    {
      "epoch": 0.20278833967046894,
      "grad_norm": 0.27561241388320923,
      "learning_rate": 4.9991779205767e-05,
      "loss": 0.1964,
      "num_input_tokens_seen": 656640,
      "step": 20
    },
    {
      "epoch": 0.2534854245880862,
      "grad_norm": 0.2867205739021301,
      "learning_rate": 4.99871554050172e-05,
      "loss": 0.2034,
      "num_input_tokens_seen": 820416,
      "step": 25
    },
    {
      "epoch": 0.3041825095057034,
      "grad_norm": 0.2154628187417984,
      "learning_rate": 4.9981504480176696e-05,
      "loss": 0.1831,
      "num_input_tokens_seen": 984000,
      "step": 30
    },
    {
      "epoch": 0.3548795944233207,
      "grad_norm": 0.39752644300460815,
      "learning_rate": 4.997482666353287e-05,
      "loss": 0.1736,
      "num_input_tokens_seen": 1150080,
      "step": 35
    },
    {
      "epoch": 0.4055766793409379,
      "grad_norm": 0.40215054154396057,
      "learning_rate": 4.996712222958461e-05,
      "loss": 0.167,
      "num_input_tokens_seen": 1313152,
      "step": 40
    },
    {
      "epoch": 0.45627376425855515,
      "grad_norm": 0.24769514799118042,
      "learning_rate": 4.9958391495031026e-05,
      "loss": 0.176,
      "num_input_tokens_seen": 1479104,
      "step": 45
    },
    {
      "epoch": 0.5069708491761724,
      "grad_norm": 0.5237451195716858,
      "learning_rate": 4.994863481875841e-05,
      "loss": 0.1769,
      "num_input_tokens_seen": 1645120,
      "step": 50
    },
    {
      "epoch": 0.5576679340937896,
      "grad_norm": 0.4815216362476349,
      "learning_rate": 4.993785260182552e-05,
      "loss": 0.1544,
      "num_input_tokens_seen": 1809024,
      "step": 55
    },
    {
      "epoch": 0.6083650190114068,
      "grad_norm": 0.24125471711158752,
      "learning_rate": 4.992604528744705e-05,
      "loss": 0.1466,
      "num_input_tokens_seen": 1973696,
      "step": 60
    },
    {
      "epoch": 0.6590621039290241,
      "grad_norm": 0.43303415179252625,
      "learning_rate": 4.991321336097546e-05,
      "loss": 0.1475,
      "num_input_tokens_seen": 2138560,
      "step": 65
    },
    {
      "epoch": 0.7097591888466414,
      "grad_norm": 0.41517049074172974,
      "learning_rate": 4.989935734988098e-05,
      "loss": 0.1248,
      "num_input_tokens_seen": 2302016,
      "step": 70
    },
    {
      "epoch": 0.7604562737642585,
      "grad_norm": 0.3203924000263214,
      "learning_rate": 4.9884477823729956e-05,
      "loss": 0.116,
      "num_input_tokens_seen": 2465664,
      "step": 75
    },
    {
      "epoch": 0.8111533586818758,
      "grad_norm": 0.3540217876434326,
      "learning_rate": 4.986857539416144e-05,
      "loss": 0.1304,
      "num_input_tokens_seen": 2628544,
      "step": 80
    },
    {
      "epoch": 0.861850443599493,
      "grad_norm": 0.6859760880470276,
      "learning_rate": 4.9851650714862006e-05,
      "loss": 0.1265,
      "num_input_tokens_seen": 2792064,
      "step": 85
    },
    {
      "epoch": 0.9125475285171103,
      "grad_norm": 0.43165433406829834,
      "learning_rate": 4.983370448153896e-05,
      "loss": 0.1324,
      "num_input_tokens_seen": 2955968,
      "step": 90
    },
    {
      "epoch": 0.9632446134347274,
      "grad_norm": 0.37339577078819275,
      "learning_rate": 4.981473743189163e-05,
      "loss": 0.1093,
      "num_input_tokens_seen": 3118976,
      "step": 95
    },
    {
      "epoch": 1.020278833967047,
      "grad_norm": 0.3954649269580841,
      "learning_rate": 4.979475034558115e-05,
      "loss": 0.1196,
      "num_input_tokens_seen": 3291584,
      "step": 100
    },
    {
      "epoch": 1.020278833967047,
      "eval_loss": 0.09924378991127014,
      "eval_runtime": 104.8348,
      "eval_samples_per_second": 6.687,
      "eval_steps_per_second": 0.839,
      "num_input_tokens_seen": 3291584,
      "step": 100
    },
    {
      "epoch": 1.0709759188846641,
      "grad_norm": 0.405510812997818,
      "learning_rate": 4.977374404419837e-05,
      "loss": 0.0985,
      "num_input_tokens_seen": 3453568,
      "step": 105
    },
    {
      "epoch": 1.1216730038022813,
      "grad_norm": 0.3010367453098297,
      "learning_rate": 4.975171939123005e-05,
      "loss": 0.0914,
      "num_input_tokens_seen": 3618304,
      "step": 110
    },
    {
      "epoch": 1.1723700887198987,
      "grad_norm": 0.32824525237083435,
      "learning_rate": 4.9728677292023405e-05,
      "loss": 0.0872,
      "num_input_tokens_seen": 3783424,
      "step": 115
    },
    {
      "epoch": 1.2230671736375158,
      "grad_norm": 0.6524387001991272,
      "learning_rate": 4.970461869374889e-05,
      "loss": 0.0986,
      "num_input_tokens_seen": 3947136,
      "step": 120
    },
    {
      "epoch": 1.2737642585551332,
      "grad_norm": 0.5022702813148499,
      "learning_rate": 4.967954458536126e-05,
      "loss": 0.0929,
      "num_input_tokens_seen": 4113344,
      "step": 125
    },
    {
      "epoch": 1.3244613434727504,
      "grad_norm": 0.6579980850219727,
      "learning_rate": 4.965345599755887e-05,
      "loss": 0.1045,
      "num_input_tokens_seen": 4277312,
      "step": 130
    },
    {
      "epoch": 1.3751584283903675,
      "grad_norm": 0.3814283013343811,
      "learning_rate": 4.962635400274142e-05,
      "loss": 0.0759,
      "num_input_tokens_seen": 4441344,
      "step": 135
    },
    {
      "epoch": 1.4258555133079849,
      "grad_norm": 0.4254150986671448,
      "learning_rate": 4.959823971496574e-05,
      "loss": 0.1045,
      "num_input_tokens_seen": 4605056,
      "step": 140
    },
    {
      "epoch": 1.476552598225602,
      "grad_norm": 0.7336615324020386,
      "learning_rate": 4.95691142899001e-05,
      "loss": 0.1082,
      "num_input_tokens_seen": 4767872,
      "step": 145
    },
    {
      "epoch": 1.5272496831432192,
      "grad_norm": 0.42282769083976746,
      "learning_rate": 4.9538978924776634e-05,
      "loss": 0.0754,
      "num_input_tokens_seen": 4931392,
      "step": 150
    },
    {
      "epoch": 1.5779467680608366,
      "grad_norm": 0.3356935381889343,
      "learning_rate": 4.9507834858342186e-05,
      "loss": 0.0889,
      "num_input_tokens_seen": 5095552,
      "step": 155
    },
    {
      "epoch": 1.6286438529784537,
      "grad_norm": 0.2882402539253235,
      "learning_rate": 4.9475683370807326e-05,
      "loss": 0.0809,
      "num_input_tokens_seen": 5258624,
      "step": 160
    },
    {
      "epoch": 1.6793409378960709,
      "grad_norm": 0.6393563151359558,
      "learning_rate": 4.9442525783793794e-05,
      "loss": 0.0765,
      "num_input_tokens_seen": 5422464,
      "step": 165
    },
    {
      "epoch": 1.7300380228136882,
      "grad_norm": 0.46365517377853394,
      "learning_rate": 4.940836346028011e-05,
      "loss": 0.0777,
      "num_input_tokens_seen": 5586944,
      "step": 170
    },
    {
      "epoch": 1.7807351077313056,
      "grad_norm": 0.7445991039276123,
      "learning_rate": 4.937319780454559e-05,
      "loss": 0.0776,
      "num_input_tokens_seen": 5751104,
      "step": 175
    },
    {
      "epoch": 1.8314321926489225,
      "grad_norm": 0.4604029357433319,
      "learning_rate": 4.933703026211262e-05,
      "loss": 0.0753,
      "num_input_tokens_seen": 5916608,
      "step": 180
    },
    {
      "epoch": 1.88212927756654,
      "grad_norm": 0.5083752274513245,
      "learning_rate": 4.9299862319687204e-05,
      "loss": 0.0787,
      "num_input_tokens_seen": 6080768,
      "step": 185
    },
    {
      "epoch": 1.9328263624841573,
      "grad_norm": 0.3158109188079834,
      "learning_rate": 4.926169550509787e-05,
      "loss": 0.0712,
      "num_input_tokens_seen": 6244992,
      "step": 190
    },
    {
      "epoch": 1.9835234474017744,
      "grad_norm": 0.2848895788192749,
      "learning_rate": 4.9222531387232885e-05,
      "loss": 0.0657,
      "num_input_tokens_seen": 6409536,
      "step": 195
    },
    {
      "epoch": 2.040557667934094,
      "grad_norm": 0.6200438737869263,
      "learning_rate": 4.9182371575975736e-05,
      "loss": 0.0632,
      "num_input_tokens_seen": 6580928,
      "step": 200
    },
    {
      "epoch": 2.040557667934094,
      "eval_loss": 0.07071041315793991,
      "eval_runtime": 104.6768,
      "eval_samples_per_second": 6.697,
      "eval_steps_per_second": 0.841,
      "num_input_tokens_seen": 6580928,
      "step": 200
    },
    {
      "epoch": 2.091254752851711,
      "grad_norm": 0.3254186511039734,
      "learning_rate": 4.914121772213898e-05,
      "loss": 0.0709,
      "num_input_tokens_seen": 6746432,
      "step": 205
    },
    {
      "epoch": 2.1419518377693283,
      "grad_norm": 0.42892560362815857,
      "learning_rate": 4.909907151739633e-05,
      "loss": 0.0523,
      "num_input_tokens_seen": 6909760,
      "step": 210
    },
    {
      "epoch": 2.1926489226869457,
      "grad_norm": 0.349162757396698,
      "learning_rate": 4.905593469421323e-05,
      "loss": 0.0665,
      "num_input_tokens_seen": 7073728,
      "step": 215
    },
    {
      "epoch": 2.2433460076045626,
      "grad_norm": 0.4294612407684326,
      "learning_rate": 4.9011809025775486e-05,
      "loss": 0.0591,
      "num_input_tokens_seen": 7238912,
      "step": 220
    },
    {
      "epoch": 2.29404309252218,
      "grad_norm": 0.6912519335746765,
      "learning_rate": 4.8966696325916515e-05,
      "loss": 0.0599,
      "num_input_tokens_seen": 7404480,
      "step": 225
    },
    {
      "epoch": 2.3447401774397973,
      "grad_norm": 0.8554105758666992,
      "learning_rate": 4.892059844904272e-05,
      "loss": 0.0732,
      "num_input_tokens_seen": 7568384,
      "step": 230
    },
    {
      "epoch": 2.3954372623574143,
      "grad_norm": 0.4321288764476776,
      "learning_rate": 4.887351729005726e-05,
      "loss": 0.0561,
      "num_input_tokens_seen": 7732544,
      "step": 235
    },
    {
      "epoch": 2.4461343472750317,
      "grad_norm": 0.43583258986473083,
      "learning_rate": 4.882545478428218e-05,
      "loss": 0.064,
      "num_input_tokens_seen": 7894656,
      "step": 240
    },
    {
      "epoch": 2.496831432192649,
      "grad_norm": 0.4895538091659546,
      "learning_rate": 4.877641290737884e-05,
      "loss": 0.054,
      "num_input_tokens_seen": 8057216,
      "step": 245
    },
    {
      "epoch": 2.5475285171102664,
      "grad_norm": 0.4905809462070465,
      "learning_rate": 4.8726393675266716e-05,
      "loss": 0.0635,
      "num_input_tokens_seen": 8221248,
      "step": 250
    },
    {
      "epoch": 2.5982256020278833,
      "grad_norm": 0.8135025501251221,
      "learning_rate": 4.8675399144040537e-05,
      "loss": 0.0519,
      "num_input_tokens_seen": 8384512,
      "step": 255
    },
    {
      "epoch": 2.6489226869455007,
      "grad_norm": 0.4855857193470001,
      "learning_rate": 4.862343140988573e-05,
      "loss": 0.0554,
      "num_input_tokens_seen": 8549440,
      "step": 260
    },
    {
      "epoch": 2.6996197718631176,
      "grad_norm": 0.39417335391044617,
      "learning_rate": 4.8570492608992325e-05,
      "loss": 0.0547,
      "num_input_tokens_seen": 8713856,
      "step": 265
    },
    {
      "epoch": 2.750316856780735,
      "grad_norm": 0.48521384596824646,
      "learning_rate": 4.851658491746707e-05,
      "loss": 0.0814,
      "num_input_tokens_seen": 8877888,
      "step": 270
    },
    {
      "epoch": 2.8010139416983524,
      "grad_norm": 0.31612926721572876,
      "learning_rate": 4.846171055124401e-05,
      "loss": 0.0619,
      "num_input_tokens_seen": 9041920,
      "step": 275
    },
    {
      "epoch": 2.8517110266159698,
      "grad_norm": 0.5412792563438416,
      "learning_rate": 4.8405871765993433e-05,
      "loss": 0.061,
      "num_input_tokens_seen": 9205824,
      "step": 280
    },
    {
      "epoch": 2.9024081115335867,
      "grad_norm": 0.3346579074859619,
      "learning_rate": 4.834907085702908e-05,
      "loss": 0.052,
      "num_input_tokens_seen": 9369280,
      "step": 285
    },
    {
      "epoch": 2.953105196451204,
      "grad_norm": 0.5361500382423401,
      "learning_rate": 4.829131015921385e-05,
      "loss": 0.0547,
      "num_input_tokens_seen": 9534464,
      "step": 290
    },
    {
      "epoch": 3.0101394169835234,
      "grad_norm": 0.6115812063217163,
      "learning_rate": 4.82325920468638e-05,
      "loss": 0.0796,
      "num_input_tokens_seen": 9708032,
      "step": 295
    },
    {
      "epoch": 3.0608365019011408,
      "grad_norm": 0.3585353195667267,
      "learning_rate": 4.817291893365055e-05,
      "loss": 0.0481,
      "num_input_tokens_seen": 9871936,
      "step": 300
    },
    {
      "epoch": 3.0608365019011408,
      "eval_loss": 0.061122000217437744,
      "eval_runtime": 104.817,
      "eval_samples_per_second": 6.688,
      "eval_steps_per_second": 0.84,
      "num_input_tokens_seen": 9871936,
      "step": 300
    },
    {
      "epoch": 3.111533586818758,
      "grad_norm": 0.5046743154525757,
      "learning_rate": 4.8112293272502043e-05,
      "loss": 0.0568,
      "num_input_tokens_seen": 10036352,
      "step": 305
    },
    {
      "epoch": 3.162230671736375,
      "grad_norm": 0.4078225791454315,
      "learning_rate": 4.805071755550177e-05,
      "loss": 0.0614,
      "num_input_tokens_seen": 10201088,
      "step": 310
    },
    {
      "epoch": 3.2129277566539924,
      "grad_norm": 0.337298184633255,
      "learning_rate": 4.7988194313786275e-05,
      "loss": 0.0287,
      "num_input_tokens_seen": 10366016,
      "step": 315
    },
    {
      "epoch": 3.26362484157161,
      "grad_norm": 0.6969597339630127,
      "learning_rate": 4.7924726117441135e-05,
      "loss": 0.0452,
      "num_input_tokens_seen": 10528704,
      "step": 320
    },
    {
      "epoch": 3.3143219264892267,
      "grad_norm": 0.6235790252685547,
      "learning_rate": 4.7860315575395316e-05,
      "loss": 0.0497,
      "num_input_tokens_seen": 10693760,
      "step": 325
    },
    {
      "epoch": 3.365019011406844,
      "grad_norm": 0.3542743921279907,
      "learning_rate": 4.7794965335313926e-05,
      "loss": 0.0523,
      "num_input_tokens_seen": 10859200,
      "step": 330
    },
    {
      "epoch": 3.4157160963244615,
      "grad_norm": 0.45905178785324097,
      "learning_rate": 4.772867808348938e-05,
      "loss": 0.0618,
      "num_input_tokens_seen": 11022400,
      "step": 335
    },
    {
      "epoch": 3.4664131812420784,
      "grad_norm": 0.7552258372306824,
      "learning_rate": 4.766145654473095e-05,
      "loss": 0.0516,
      "num_input_tokens_seen": 11187328,
      "step": 340
    },
    {
      "epoch": 3.517110266159696,
      "grad_norm": 0.4128399193286896,
      "learning_rate": 4.759330348225284e-05,
      "loss": 0.0428,
      "num_input_tokens_seen": 11352832,
      "step": 345
    },
    {
      "epoch": 3.567807351077313,
      "grad_norm": 0.67793869972229,
      "learning_rate": 4.752422169756048e-05,
      "loss": 0.0582,
      "num_input_tokens_seen": 11518016,
      "step": 350
    },
    {
      "epoch": 3.6185044359949305,
      "grad_norm": 0.4128822684288025,
      "learning_rate": 4.745421403033548e-05,
      "loss": 0.0393,
      "num_input_tokens_seen": 11680640,
      "step": 355
    },
    {
      "epoch": 3.6692015209125475,
      "grad_norm": 0.475558340549469,
      "learning_rate": 4.738328335831883e-05,
      "loss": 0.0533,
      "num_input_tokens_seen": 11842880,
      "step": 360
    },
    {
      "epoch": 3.719898605830165,
      "grad_norm": 0.3762592077255249,
      "learning_rate": 4.731143259719265e-05,
      "loss": 0.0362,
      "num_input_tokens_seen": 12009344,
      "step": 365
    },
    {
      "epoch": 3.770595690747782,
      "grad_norm": 0.5744361281394958,
      "learning_rate": 4.72386647004603e-05,
      "loss": 0.0553,
      "num_input_tokens_seen": 12173056,
      "step": 370
    },
    {
      "epoch": 3.821292775665399,
      "grad_norm": 0.5609379410743713,
      "learning_rate": 4.716498265932501e-05,
      "loss": 0.0501,
      "num_input_tokens_seen": 12337408,
      "step": 375
    },
    {
      "epoch": 3.8719898605830165,
      "grad_norm": 0.42430299520492554,
      "learning_rate": 4.709038950256688e-05,
      "loss": 0.0384,
      "num_input_tokens_seen": 12501312,
      "step": 380
    },
    {
      "epoch": 3.922686945500634,
      "grad_norm": 0.5434412956237793,
      "learning_rate": 4.701488829641845e-05,
      "loss": 0.0354,
      "num_input_tokens_seen": 12665152,
      "step": 385
    },
    {
      "epoch": 3.973384030418251,
      "grad_norm": 0.5311803221702576,
      "learning_rate": 4.693848214443858e-05,
      "loss": 0.0348,
      "num_input_tokens_seen": 12828608,
      "step": 390
    },
    {
      "epoch": 4.030418250950571,
      "grad_norm": 0.5105487704277039,
      "learning_rate": 4.686117418738489e-05,
      "loss": 0.036,
      "num_input_tokens_seen": 13001024,
      "step": 395
    },
    {
      "epoch": 4.081115335868188,
      "grad_norm": 0.7211757898330688,
      "learning_rate": 4.678296760308474e-05,
      "loss": 0.0352,
      "num_input_tokens_seen": 13164032,
      "step": 400
    },
    {
      "epoch": 4.081115335868188,
      "eval_loss": 0.06444713473320007,
      "eval_runtime": 104.6657,
      "eval_samples_per_second": 6.698,
      "eval_steps_per_second": 0.841,
      "num_input_tokens_seen": 13164032,
      "step": 400
    },
    {
      "epoch": 4.1318124207858045,
      "grad_norm": 0.6069203615188599,
      "learning_rate": 4.6703865606304465e-05,
      "loss": 0.0295,
      "num_input_tokens_seen": 13329344,
      "step": 405
    },
    {
      "epoch": 4.182509505703422,
      "grad_norm": 0.6209607720375061,
      "learning_rate": 4.662387144861734e-05,
      "loss": 0.0323,
      "num_input_tokens_seen": 13492800,
      "step": 410
    },
    {
      "epoch": 4.233206590621039,
      "grad_norm": 0.772402822971344,
      "learning_rate": 4.6542988418269876e-05,
      "loss": 0.0381,
      "num_input_tokens_seen": 13656704,
      "step": 415
    },
    {
      "epoch": 4.283903675538657,
      "grad_norm": 1.5642151832580566,
      "learning_rate": 4.6461219840046654e-05,
      "loss": 0.0397,
      "num_input_tokens_seen": 13821568,
      "step": 420
    },
    {
      "epoch": 4.334600760456274,
      "grad_norm": 0.7930517792701721,
      "learning_rate": 4.637856907513366e-05,
      "loss": 0.0291,
      "num_input_tokens_seen": 13985792,
      "step": 425
    },
    {
      "epoch": 4.385297845373891,
      "grad_norm": 1.0651878118515015,
      "learning_rate": 4.629503952098011e-05,
      "loss": 0.037,
      "num_input_tokens_seen": 14150080,
      "step": 430
    },
    {
      "epoch": 4.435994930291509,
      "grad_norm": 0.7544033527374268,
      "learning_rate": 4.6210634611158816e-05,
      "loss": 0.0309,
      "num_input_tokens_seen": 14314432,
      "step": 435
    },
    {
      "epoch": 4.486692015209125,
      "grad_norm": 0.6338365077972412,
      "learning_rate": 4.612535781522504e-05,
      "loss": 0.0256,
      "num_input_tokens_seen": 14480128,
      "step": 440
    },
    {
      "epoch": 4.537389100126743,
      "grad_norm": 0.4086742103099823,
      "learning_rate": 4.6039212638573833e-05,
      "loss": 0.0213,
      "num_input_tokens_seen": 14643712,
      "step": 445
    },
    {
      "epoch": 4.58808618504436,
      "grad_norm": 0.750950038433075,
      "learning_rate": 4.595220262229601e-05,
      "loss": 0.027,
      "num_input_tokens_seen": 14809600,
      "step": 450
    },
    {
      "epoch": 4.638783269961977,
      "grad_norm": 0.5362918376922607,
      "learning_rate": 4.586433134303257e-05,
      "loss": 0.0206,
      "num_input_tokens_seen": 14973248,
      "step": 455
    },
    {
      "epoch": 4.689480354879595,
      "grad_norm": 1.1060535907745361,
      "learning_rate": 4.5775602412827604e-05,
      "loss": 0.0326,
      "num_input_tokens_seen": 15136128,
      "step": 460
    },
    {
      "epoch": 4.740177439797211,
      "grad_norm": 0.9209577441215515,
      "learning_rate": 4.5686019478979915e-05,
      "loss": 0.0368,
      "num_input_tokens_seen": 15299008,
      "step": 465
    },
    {
      "epoch": 4.7908745247148286,
      "grad_norm": 0.5697187781333923,
      "learning_rate": 4.559558622389304e-05,
      "loss": 0.0318,
      "num_input_tokens_seen": 15462528,
      "step": 470
    },
    {
      "epoch": 4.841571609632446,
      "grad_norm": 0.382828950881958,
      "learning_rate": 4.55043063649239e-05,
      "loss": 0.0285,
      "num_input_tokens_seen": 15626240,
      "step": 475
    },
    {
      "epoch": 4.892268694550063,
      "grad_norm": 0.3667276203632355,
      "learning_rate": 4.5412183654229965e-05,
      "loss": 0.0254,
      "num_input_tokens_seen": 15791168,
      "step": 480
    },
    {
      "epoch": 4.942965779467681,
      "grad_norm": 0.4793355166912079,
      "learning_rate": 4.531922187861507e-05,
      "loss": 0.0407,
      "num_input_tokens_seen": 15955712,
      "step": 485
    },
    {
      "epoch": 4.993662864385298,
      "grad_norm": 0.6076450347900391,
      "learning_rate": 4.522542485937369e-05,
      "loss": 0.0307,
      "num_input_tokens_seen": 16121216,
      "step": 490
    },
    {
      "epoch": 5.050697084917617,
      "grad_norm": 0.33613860607147217,
      "learning_rate": 4.51307964521339e-05,
      "loss": 0.0189,
      "num_input_tokens_seen": 16292672,
      "step": 495
    },
    {
      "epoch": 5.101394169835235,
      "grad_norm": 0.6635271310806274,
      "learning_rate": 4.503534054669892e-05,
      "loss": 0.0188,
      "num_input_tokens_seen": 16456000,
      "step": 500
    },
    {
      "epoch": 5.101394169835235,
      "eval_loss": 0.05884933844208717,
      "eval_runtime": 104.9474,
      "eval_samples_per_second": 6.68,
      "eval_steps_per_second": 0.839,
      "num_input_tokens_seen": 16456000,
      "step": 500
    },
    {
      "epoch": 5.152091254752852,
      "grad_norm": 0.3716290593147278,
      "learning_rate": 4.493906106688712e-05,
      "loss": 0.0092,
      "num_input_tokens_seen": 16620608,
      "step": 505
    },
    {
      "epoch": 5.202788339670469,
      "grad_norm": 0.7663172483444214,
      "learning_rate": 4.484196197037082e-05,
      "loss": 0.0174,
      "num_input_tokens_seen": 16783936,
      "step": 510
    },
    {
      "epoch": 5.253485424588086,
      "grad_norm": 0.9232292771339417,
      "learning_rate": 4.474404724851356e-05,
      "loss": 0.0093,
      "num_input_tokens_seen": 16948544,
      "step": 515
    },
    {
      "epoch": 5.304182509505703,
      "grad_norm": 0.5671281218528748,
      "learning_rate": 4.4645320926206064e-05,
      "loss": 0.0141,
      "num_input_tokens_seen": 17112640,
      "step": 520
    },
    {
      "epoch": 5.354879594423321,
      "grad_norm": 0.7603790163993835,
      "learning_rate": 4.454578706170075e-05,
      "loss": 0.0197,
      "num_input_tokens_seen": 17275968,
      "step": 525
    },
    {
      "epoch": 5.405576679340938,
      "grad_norm": 0.7700641751289368,
      "learning_rate": 4.444544974644493e-05,
      "loss": 0.0125,
      "num_input_tokens_seen": 17439680,
      "step": 530
    },
    {
      "epoch": 5.4562737642585555,
      "grad_norm": 0.876714825630188,
      "learning_rate": 4.434431310491267e-05,
      "loss": 0.0218,
      "num_input_tokens_seen": 17603136,
      "step": 535
    },
    {
      "epoch": 5.506970849176172,
      "grad_norm": 0.6164574027061462,
      "learning_rate": 4.4242381294435154e-05,
      "loss": 0.0235,
      "num_input_tokens_seen": 17768192,
      "step": 540
    },
    {
      "epoch": 5.557667934093789,
      "grad_norm": 0.500507652759552,
      "learning_rate": 4.413965850502987e-05,
      "loss": 0.0181,
      "num_input_tokens_seen": 17932928,
      "step": 545
    },
    {
      "epoch": 5.608365019011407,
      "grad_norm": 0.6840673089027405,
      "learning_rate": 4.4036148959228365e-05,
      "loss": 0.0214,
      "num_input_tokens_seen": 18096768,
      "step": 550
    },
    {
      "epoch": 5.659062103929024,
      "grad_norm": 0.41719505190849304,
      "learning_rate": 4.393185691190264e-05,
      "loss": 0.0196,
      "num_input_tokens_seen": 18261184,
      "step": 555
    },
    {
      "epoch": 5.7097591888466415,
      "grad_norm": 0.30571097135543823,
      "learning_rate": 4.382678665009028e-05,
      "loss": 0.0177,
      "num_input_tokens_seen": 18426880,
      "step": 560
    },
    {
      "epoch": 5.760456273764259,
      "grad_norm": 0.7241319417953491,
      "learning_rate": 4.372094249281821e-05,
      "loss": 0.0181,
      "num_input_tokens_seen": 18591488,
      "step": 565
    },
    {
      "epoch": 5.811153358681876,
      "grad_norm": 1.0090749263763428,
      "learning_rate": 4.3614328790925177e-05,
      "loss": 0.017,
      "num_input_tokens_seen": 18755584,
      "step": 570
    },
    {
      "epoch": 5.861850443599493,
      "grad_norm": 1.0040756464004517,
      "learning_rate": 4.350694992688289e-05,
      "loss": 0.02,
      "num_input_tokens_seen": 18919360,
      "step": 575
    },
    {
      "epoch": 5.91254752851711,
      "grad_norm": 1.0349421501159668,
      "learning_rate": 4.3398810314615876e-05,
      "loss": 0.0238,
      "num_input_tokens_seen": 19082816,
      "step": 580
    },
    {
      "epoch": 5.9632446134347274,
      "grad_norm": 1.2083125114440918,
      "learning_rate": 4.3289914399320034e-05,
      "loss": 0.0148,
      "num_input_tokens_seen": 19247040,
      "step": 585
    },
    {
      "epoch": 6.020278833967047,
      "grad_norm": 0.5085653066635132,
      "learning_rate": 4.318026665727993e-05,
      "loss": 0.0116,
      "num_input_tokens_seen": 19420480,
      "step": 590
    },
    {
      "epoch": 6.070975918884664,
      "grad_norm": 0.5758472681045532,
      "learning_rate": 4.306987159568479e-05,
      "loss": 0.0126,
      "num_input_tokens_seen": 19583680,
      "step": 595
    },
    {
      "epoch": 6.1216730038022815,
      "grad_norm": 0.11992669850587845,
      "learning_rate": 4.2958733752443195e-05,
      "loss": 0.0041,
      "num_input_tokens_seen": 19746816,
      "step": 600
    },
    {
      "epoch": 6.1216730038022815,
      "eval_loss": 0.07074201852083206,
      "eval_runtime": 106.0122,
      "eval_samples_per_second": 6.612,
      "eval_steps_per_second": 0.83,
      "num_input_tokens_seen": 19746816,
      "step": 600
    },
    {
      "epoch": 6.172370088719899,
      "grad_norm": 0.5216241478919983,
      "learning_rate": 4.284685769599658e-05,
      "loss": 0.008,
      "num_input_tokens_seen": 19911552,
      "step": 605
    },
    {
      "epoch": 6.223067173637516,
      "grad_norm": 1.355669617652893,
      "learning_rate": 4.273424802513145e-05,
      "loss": 0.0069,
      "num_input_tokens_seen": 20075200,
      "step": 610
    },
    {
      "epoch": 6.273764258555133,
      "grad_norm": 0.1907280832529068,
      "learning_rate": 4.262090936879029e-05,
      "loss": 0.0016,
      "num_input_tokens_seen": 20241472,
      "step": 615
    },
    {
      "epoch": 6.32446134347275,
      "grad_norm": 0.0559777207672596,
      "learning_rate": 4.250684638588138e-05,
      "loss": 0.0091,
      "num_input_tokens_seen": 20403904,
      "step": 620
    },
    {
      "epoch": 6.3751584283903675,
      "grad_norm": 1.0960038900375366,
      "learning_rate": 4.239206376508717e-05,
      "loss": 0.0139,
      "num_input_tokens_seen": 20569536,
      "step": 625
    },
    {
      "epoch": 6.425855513307985,
      "grad_norm": 0.3808915317058563,
      "learning_rate": 4.227656622467162e-05,
      "loss": 0.0122,
      "num_input_tokens_seen": 20732032,
      "step": 630
    },
    {
      "epoch": 6.476552598225602,
      "grad_norm": 1.5081939697265625,
      "learning_rate": 4.216035851228626e-05,
      "loss": 0.0105,
      "num_input_tokens_seen": 20895552,
      "step": 635
    },
    {
      "epoch": 6.52724968314322,
      "grad_norm": 0.4840177893638611,
      "learning_rate": 4.204344540477499e-05,
      "loss": 0.0049,
      "num_input_tokens_seen": 21060736,
      "step": 640
    },
    {
      "epoch": 6.577946768060836,
      "grad_norm": 1.1747405529022217,
      "learning_rate": 4.192583170797774e-05,
      "loss": 0.0093,
      "num_input_tokens_seen": 21224320,
      "step": 645
    },
    {
      "epoch": 6.6286438529784535,
      "grad_norm": 0.9056284427642822,
      "learning_rate": 4.180752225653292e-05,
      "loss": 0.0137,
      "num_input_tokens_seen": 21387328,
      "step": 650
    },
    {
      "epoch": 6.679340937896071,
      "grad_norm": 1.1171244382858276,
      "learning_rate": 4.16885219136787e-05,
      "loss": 0.0054,
      "num_input_tokens_seen": 21551616,
      "step": 655
    },
    {
      "epoch": 6.730038022813688,
      "grad_norm": 0.4659402370452881,
      "learning_rate": 4.1568835571053075e-05,
      "loss": 0.0068,
      "num_input_tokens_seen": 21716864,
      "step": 660
    },
    {
      "epoch": 6.780735107731306,
      "grad_norm": 0.12265567481517792,
      "learning_rate": 4.144846814849282e-05,
      "loss": 0.0106,
      "num_input_tokens_seen": 21882304,
      "step": 665
    },
    {
      "epoch": 6.831432192648923,
      "grad_norm": 0.3785063326358795,
      "learning_rate": 4.132742459383122e-05,
      "loss": 0.0095,
      "num_input_tokens_seen": 22045504,
      "step": 670
    },
    {
      "epoch": 6.8821292775665395,
      "grad_norm": 0.9941157102584839,
      "learning_rate": 4.120570988269472e-05,
      "loss": 0.0079,
      "num_input_tokens_seen": 22210368,
      "step": 675
    },
    {
      "epoch": 6.932826362484157,
      "grad_norm": 2.0760669708251953,
      "learning_rate": 4.108332901829836e-05,
      "loss": 0.0113,
      "num_input_tokens_seen": 22374976,
      "step": 680
    },
    {
      "epoch": 6.983523447401774,
      "grad_norm": 1.1653289794921875,
      "learning_rate": 4.096028703124014e-05,
      "loss": 0.0101,
      "num_input_tokens_seen": 22540160,
      "step": 685
    },
    {
      "epoch": 7.0405576679340935,
      "grad_norm": 0.6426116824150085,
      "learning_rate": 4.083658897929426e-05,
      "loss": 0.0094,
      "num_input_tokens_seen": 22713344,
      "step": 690
    },
    {
      "epoch": 7.091254752851711,
      "grad_norm": 0.1562691628932953,
      "learning_rate": 4.071223994720309e-05,
      "loss": 0.0083,
      "num_input_tokens_seen": 22876672,
      "step": 695
    },
    {
      "epoch": 7.141951837769328,
      "grad_norm": 0.24841922521591187,
      "learning_rate": 4.058724504646834e-05,
      "loss": 0.0018,
      "num_input_tokens_seen": 23041536,
      "step": 700
    },
    {
      "epoch": 7.141951837769328,
      "eval_loss": 0.08085346221923828,
      "eval_runtime": 104.8865,
      "eval_samples_per_second": 6.683,
      "eval_steps_per_second": 0.839,
      "num_input_tokens_seen": 23041536,
      "step": 700
    },
    {
      "epoch": 7.192648922686946,
      "grad_norm": 0.3186471462249756,
      "learning_rate": 4.046160941514079e-05,
      "loss": 0.0131,
      "num_input_tokens_seen": 23206528,
      "step": 705
    },
    {
      "epoch": 7.243346007604563,
      "grad_norm": 0.1616538017988205,
      "learning_rate": 4.033533821760917e-05,
      "loss": 0.0045,
      "num_input_tokens_seen": 23370880,
      "step": 710
    },
    {
      "epoch": 7.29404309252218,
      "grad_norm": 0.5200349688529968,
      "learning_rate": 4.0208436644387834e-05,
      "loss": 0.006,
      "num_input_tokens_seen": 23533184,
      "step": 715
    },
    {
      "epoch": 7.344740177439797,
      "grad_norm": 0.3917842507362366,
      "learning_rate": 4.008090991190341e-05,
      "loss": 0.0049,
      "num_input_tokens_seen": 23696000,
      "step": 720
    },
    {
      "epoch": 7.395437262357414,
      "grad_norm": 0.20676745474338531,
      "learning_rate": 3.9952763262280405e-05,
      "loss": 0.0014,
      "num_input_tokens_seen": 23861120,
      "step": 725
    },
    {
      "epoch": 7.446134347275032,
      "grad_norm": 0.4431841969490051,
      "learning_rate": 3.982400196312564e-05,
      "loss": 0.0037,
      "num_input_tokens_seen": 24025280,
      "step": 730
    },
    {
      "epoch": 7.496831432192649,
      "grad_norm": 0.7824487686157227,
      "learning_rate": 3.969463130731183e-05,
      "loss": 0.0083,
      "num_input_tokens_seen": 24190144,
      "step": 735
    },
    {
      "epoch": 7.547528517110266,
      "grad_norm": 1.250172734260559,
      "learning_rate": 3.95646566127599e-05,
      "loss": 0.002,
      "num_input_tokens_seen": 24356352,
      "step": 740
    },
    {
      "epoch": 7.598225602027884,
      "grad_norm": 0.06873621046543121,
      "learning_rate": 3.943408322222049e-05,
      "loss": 0.0014,
      "num_input_tokens_seen": 24521024,
      "step": 745
    },
    {
      "epoch": 7.6489226869455,
      "grad_norm": 0.9064635038375854,
      "learning_rate": 3.9302916503054246e-05,
      "loss": 0.0062,
      "num_input_tokens_seen": 24683072,
      "step": 750
    },
    {
      "epoch": 7.699619771863118,
      "grad_norm": 0.5123156905174255,
      "learning_rate": 3.917116184701125e-05,
      "loss": 0.0031,
      "num_input_tokens_seen": 24848384,
      "step": 755
    },
    {
      "epoch": 7.750316856780735,
      "grad_norm": 1.4605395793914795,
      "learning_rate": 3.903882467000937e-05,
      "loss": 0.0048,
      "num_input_tokens_seen": 25011648,
      "step": 760
    },
    {
      "epoch": 7.801013941698352,
      "grad_norm": 0.05572497099637985,
      "learning_rate": 3.8905910411911625e-05,
      "loss": 0.0066,
      "num_input_tokens_seen": 25175360,
      "step": 765
    },
    {
      "epoch": 7.85171102661597,
      "grad_norm": 0.5212889909744263,
      "learning_rate": 3.8772424536302564e-05,
      "loss": 0.0033,
      "num_input_tokens_seen": 25339072,
      "step": 770
    },
    {
      "epoch": 7.902408111533587,
      "grad_norm": 0.07258739322423935,
      "learning_rate": 3.8638372530263715e-05,
      "loss": 0.0024,
      "num_input_tokens_seen": 25502656,
      "step": 775
    },
    {
      "epoch": 7.9531051964512045,
      "grad_norm": 0.0964067205786705,
      "learning_rate": 3.850375990414801e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 25666560,
      "step": 780
    },
    {
      "epoch": 8.010139416983524,
      "grad_norm": 0.596034824848175,
      "learning_rate": 3.836859219135324e-05,
      "loss": 0.0052,
      "num_input_tokens_seen": 25837632,
      "step": 785
    },
    {
      "epoch": 8.060836501901141,
      "grad_norm": 0.19794920086860657,
      "learning_rate": 3.823287494809469e-05,
      "loss": 0.0018,
      "num_input_tokens_seen": 26002176,
      "step": 790
    },
    {
      "epoch": 8.111533586818759,
      "grad_norm": 0.016648059710860252,
      "learning_rate": 3.8096613753176634e-05,
      "loss": 0.0032,
      "num_input_tokens_seen": 26164736,
      "step": 795
    },
    {
      "epoch": 8.162230671736376,
      "grad_norm": 0.25073590874671936,
      "learning_rate": 3.7959814207763135e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 26329664,
      "step": 800
    },
    {
      "epoch": 8.162230671736376,
      "eval_loss": 0.1101459190249443,
      "eval_runtime": 104.9186,
      "eval_samples_per_second": 6.681,
      "eval_steps_per_second": 0.839,
      "num_input_tokens_seen": 26329664,
      "step": 800
    }
  ],
  "logging_steps": 5,
  "max_steps": 2450,
  "num_input_tokens_seen": 26329664,
  "num_train_epochs": 25,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.227070984070234e+18,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}