{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.596169193934557,
  "eval_steps": 10,
  "global_step": 1000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.007980845969672785,
      "grad_norm": 0.7833889722824097,
      "learning_rate": 0.00016,
      "loss": 1.3081,
      "step": 5
    },
    {
      "epoch": 0.01596169193934557,
      "grad_norm": 0.6466716527938843,
      "learning_rate": 0.00012,
      "loss": 1.2368,
      "step": 10
    },
    {
      "epoch": 0.01596169193934557,
      "eval_loss": 1.2068681716918945,
      "eval_runtime": 20.8667,
      "eval_samples_per_second": 1.486,
      "eval_steps_per_second": 0.767,
      "step": 10
    },
    {
      "epoch": 0.023942537909018357,
      "grad_norm": 0.6432256698608398,
      "learning_rate": 8e-05,
      "loss": 1.1477,
      "step": 15
    },
    {
      "epoch": 0.03192338387869114,
      "grad_norm": 0.6714484095573425,
      "learning_rate": 4e-05,
      "loss": 1.0776,
      "step": 20
    },
    {
      "epoch": 0.03192338387869114,
      "eval_loss": 1.16266667842865,
      "eval_runtime": 21.285,
      "eval_samples_per_second": 1.456,
      "eval_steps_per_second": 0.752,
      "step": 20
    },
    {
      "epoch": 0.03990422984836393,
      "grad_norm": 0.6418822407722473,
      "learning_rate": 0.0,
      "loss": 1.1614,
      "step": 25
    },
    {
      "epoch": 0.047885075818036714,
      "grad_norm": 0.8091150522232056,
      "learning_rate": 0.000194,
      "loss": 1.1676,
      "step": 30
    },
    {
      "epoch": 0.047885075818036714,
      "eval_loss": 1.151872992515564,
      "eval_runtime": 21.3531,
      "eval_samples_per_second": 1.452,
      "eval_steps_per_second": 0.749,
      "step": 30
    },
    {
      "epoch": 0.055865921787709494,
      "grad_norm": 0.7295385003089905,
      "learning_rate": 0.000193,
      "loss": 1.1319,
      "step": 35
    },
    {
      "epoch": 0.06384676775738228,
      "grad_norm": 1.056991696357727,
      "learning_rate": 0.000192,
      "loss": 1.0188,
      "step": 40
    },
    {
      "epoch": 0.06384676775738228,
      "eval_loss": 1.1241333484649658,
      "eval_runtime": 21.3665,
      "eval_samples_per_second": 1.451,
      "eval_steps_per_second": 0.749,
      "step": 40
    },
    {
      "epoch": 0.07182761372705507,
      "grad_norm": 0.7033870816230774,
      "learning_rate": 0.000191,
      "loss": 1.0331,
      "step": 45
    },
    {
      "epoch": 0.07980845969672785,
      "grad_norm": 0.6337556838989258,
      "learning_rate": 0.00019,
      "loss": 1.1616,
      "step": 50
    },
    {
      "epoch": 0.07980845969672785,
      "eval_loss": 1.1174383163452148,
      "eval_runtime": 20.7909,
      "eval_samples_per_second": 1.491,
      "eval_steps_per_second": 0.77,
      "step": 50
    },
    {
      "epoch": 0.08778930566640064,
      "grad_norm": 0.7127214074134827,
      "learning_rate": 0.00018899999999999999,
      "loss": 1.0874,
      "step": 55
    },
    {
      "epoch": 0.09577015163607343,
      "grad_norm": 0.7299643158912659,
      "learning_rate": 0.000188,
      "loss": 1.0098,
      "step": 60
    },
    {
      "epoch": 0.09577015163607343,
      "eval_loss": 1.106464147567749,
      "eval_runtime": 21.2918,
      "eval_samples_per_second": 1.456,
      "eval_steps_per_second": 0.751,
      "step": 60
    },
    {
      "epoch": 0.10375099760574621,
      "grad_norm": 0.7458897829055786,
      "learning_rate": 0.00018700000000000002,
      "loss": 1.0114,
      "step": 65
    },
    {
      "epoch": 0.11173184357541899,
      "grad_norm": 0.7154062986373901,
      "learning_rate": 0.00018600000000000002,
      "loss": 1.0804,
      "step": 70
    },
    {
      "epoch": 0.11173184357541899,
      "eval_loss": 1.087841510772705,
      "eval_runtime": 21.2696,
      "eval_samples_per_second": 1.457,
      "eval_steps_per_second": 0.752,
      "step": 70
    },
    {
      "epoch": 0.11971268954509177,
      "grad_norm": 0.630651593208313,
      "learning_rate": 0.00018500000000000002,
      "loss": 0.988,
      "step": 75
    },
    {
      "epoch": 0.12769353551476456,
      "grad_norm": 0.6520997285842896,
      "learning_rate": 0.00018400000000000003,
      "loss": 0.9957,
      "step": 80
    },
    {
      "epoch": 0.12769353551476456,
      "eval_loss": 1.0693910121917725,
      "eval_runtime": 21.2801,
      "eval_samples_per_second": 1.457,
      "eval_steps_per_second": 0.752,
      "step": 80
    },
    {
      "epoch": 0.13567438148443736,
      "grad_norm": 0.612789511680603,
      "learning_rate": 0.000183,
      "loss": 1.0659,
      "step": 85
    },
    {
      "epoch": 0.14365522745411013,
      "grad_norm": 0.7308149337768555,
      "learning_rate": 0.000182,
      "loss": 1.0044,
      "step": 90
    },
    {
      "epoch": 0.14365522745411013,
      "eval_loss": 1.0618408918380737,
      "eval_runtime": 21.4836,
      "eval_samples_per_second": 1.443,
      "eval_steps_per_second": 0.745,
      "step": 90
    },
    {
      "epoch": 0.1516360734237829,
      "grad_norm": 0.6600955724716187,
      "learning_rate": 0.000181,
      "loss": 0.936,
      "step": 95
    },
    {
      "epoch": 0.1596169193934557,
      "grad_norm": 0.5939749479293823,
      "learning_rate": 0.00018,
      "loss": 1.0213,
      "step": 100
    },
    {
      "epoch": 0.1596169193934557,
      "eval_loss": 1.06107497215271,
      "eval_runtime": 20.9063,
      "eval_samples_per_second": 1.483,
      "eval_steps_per_second": 0.765,
      "step": 100
    },
    {
      "epoch": 0.16759776536312848,
      "grad_norm": 0.6475443840026855,
      "learning_rate": 0.00017900000000000001,
      "loss": 1.0099,
      "step": 105
    },
    {
      "epoch": 0.17557861133280128,
      "grad_norm": 0.6048471927642822,
      "learning_rate": 0.00017800000000000002,
      "loss": 0.9587,
      "step": 110
    },
    {
      "epoch": 0.17557861133280128,
      "eval_loss": 1.0483193397521973,
      "eval_runtime": 21.4083,
      "eval_samples_per_second": 1.448,
      "eval_steps_per_second": 0.747,
      "step": 110
    },
    {
      "epoch": 0.18355945730247406,
      "grad_norm": 0.5856403112411499,
      "learning_rate": 0.00017700000000000002,
      "loss": 0.9908,
      "step": 115
    },
    {
      "epoch": 0.19154030327214686,
      "grad_norm": 0.6811533570289612,
      "learning_rate": 0.00017600000000000002,
      "loss": 0.9662,
      "step": 120
    },
    {
      "epoch": 0.19154030327214686,
      "eval_loss": 1.055928349494934,
      "eval_runtime": 21.2973,
      "eval_samples_per_second": 1.456,
      "eval_steps_per_second": 0.751,
      "step": 120
    },
    {
      "epoch": 0.19952114924181963,
      "grad_norm": 0.5635804533958435,
      "learning_rate": 0.000175,
      "loss": 0.9637,
      "step": 125
    },
    {
      "epoch": 0.20750199521149243,
      "grad_norm": 0.6101672053337097,
      "learning_rate": 0.000174,
      "loss": 0.9846,
      "step": 130
    },
    {
      "epoch": 0.20750199521149243,
      "eval_loss": 1.0435600280761719,
      "eval_runtime": 21.4526,
      "eval_samples_per_second": 1.445,
      "eval_steps_per_second": 0.746,
      "step": 130
    },
    {
      "epoch": 0.2154828411811652,
      "grad_norm": 0.649726152420044,
      "learning_rate": 0.000173,
      "loss": 0.9995,
      "step": 135
    },
    {
      "epoch": 0.22346368715083798,
      "grad_norm": 0.6642662286758423,
      "learning_rate": 0.000172,
      "loss": 0.9537,
      "step": 140
    },
    {
      "epoch": 0.22346368715083798,
      "eval_loss": 1.0289674997329712,
      "eval_runtime": 21.0765,
      "eval_samples_per_second": 1.471,
      "eval_steps_per_second": 0.759,
      "step": 140
    },
    {
      "epoch": 0.23144453312051078,
      "grad_norm": 0.5957498550415039,
      "learning_rate": 0.000171,
      "loss": 1.0444,
      "step": 145
    },
    {
      "epoch": 0.23942537909018355,
      "grad_norm": 0.7650970816612244,
      "learning_rate": 0.00017,
      "loss": 1.0177,
      "step": 150
    },
    {
      "epoch": 0.23942537909018355,
      "eval_loss": 1.0302073955535889,
      "eval_runtime": 21.0268,
      "eval_samples_per_second": 1.474,
      "eval_steps_per_second": 0.761,
      "step": 150
    },
    {
      "epoch": 0.24740622505985635,
      "grad_norm": 0.6636956334114075,
      "learning_rate": 0.00016900000000000002,
      "loss": 0.9786,
      "step": 155
    },
    {
      "epoch": 0.2553870710295291,
      "grad_norm": 0.7419523596763611,
      "learning_rate": 0.000168,
      "loss": 0.9837,
      "step": 160
    },
    {
      "epoch": 0.2553870710295291,
      "eval_loss": 1.0296545028686523,
      "eval_runtime": 21.4194,
      "eval_samples_per_second": 1.447,
      "eval_steps_per_second": 0.747,
      "step": 160
    },
    {
      "epoch": 0.2633679169992019,
      "grad_norm": 0.5570073127746582,
      "learning_rate": 0.000167,
      "loss": 0.954,
      "step": 165
    },
    {
      "epoch": 0.2713487629688747,
      "grad_norm": 0.5050713419914246,
      "learning_rate": 0.000166,
      "loss": 1.0796,
      "step": 170
    },
    {
      "epoch": 0.2713487629688747,
      "eval_loss": 1.0314409732818604,
      "eval_runtime": 21.4116,
      "eval_samples_per_second": 1.448,
      "eval_steps_per_second": 0.747,
      "step": 170
    },
    {
      "epoch": 0.27932960893854747,
      "grad_norm": 0.48484355211257935,
      "learning_rate": 0.000165,
      "loss": 0.9461,
      "step": 175
    },
    {
      "epoch": 0.28731045490822027,
      "grad_norm": 0.6661683320999146,
      "learning_rate": 0.000164,
      "loss": 1.0311,
      "step": 180
    },
    {
      "epoch": 0.28731045490822027,
      "eval_loss": 1.0315138101577759,
      "eval_runtime": 21.2472,
      "eval_samples_per_second": 1.459,
      "eval_steps_per_second": 0.753,
      "step": 180
    },
    {
      "epoch": 0.29529130087789307,
      "grad_norm": 0.6185592412948608,
      "learning_rate": 0.000163,
      "loss": 0.9745,
      "step": 185
    },
    {
      "epoch": 0.3032721468475658,
      "grad_norm": 0.5738425850868225,
      "learning_rate": 0.000162,
      "loss": 0.9619,
      "step": 190
    },
    {
      "epoch": 0.3032721468475658,
      "eval_loss": 1.0191501379013062,
      "eval_runtime": 21.0898,
      "eval_samples_per_second": 1.47,
      "eval_steps_per_second": 0.759,
      "step": 190
    },
    {
      "epoch": 0.3112529928172386,
      "grad_norm": 0.6744123101234436,
      "learning_rate": 0.000161,
      "loss": 1.0097,
      "step": 195
    },
    {
      "epoch": 0.3192338387869114,
      "grad_norm": 0.7204273343086243,
      "learning_rate": 0.00016,
      "loss": 0.9916,
      "step": 200
    },
    {
      "epoch": 0.3192338387869114,
      "eval_loss": 1.0153534412384033,
      "eval_runtime": 21.3333,
      "eval_samples_per_second": 1.453,
      "eval_steps_per_second": 0.75,
      "step": 200
    },
    {
      "epoch": 0.3272146847565842,
      "grad_norm": 0.6031942367553711,
      "learning_rate": 0.00015900000000000002,
      "loss": 0.9673,
      "step": 205
    },
    {
      "epoch": 0.33519553072625696,
      "grad_norm": 0.5967186689376831,
      "learning_rate": 0.00015800000000000002,
      "loss": 0.9655,
      "step": 210
    },
    {
      "epoch": 0.33519553072625696,
      "eval_loss": 1.0138649940490723,
      "eval_runtime": 21.1862,
      "eval_samples_per_second": 1.463,
      "eval_steps_per_second": 0.755,
      "step": 210
    },
    {
      "epoch": 0.34317637669592976,
      "grad_norm": 0.6408796906471252,
      "learning_rate": 0.00015700000000000002,
      "loss": 0.9453,
      "step": 215
    },
    {
      "epoch": 0.35115722266560256,
      "grad_norm": 0.6557004451751709,
      "learning_rate": 0.00015600000000000002,
      "loss": 0.9536,
      "step": 220
    },
    {
      "epoch": 0.35115722266560256,
      "eval_loss": 1.0107518434524536,
      "eval_runtime": 21.3902,
      "eval_samples_per_second": 1.449,
      "eval_steps_per_second": 0.748,
      "step": 220
    },
    {
      "epoch": 0.35913806863527536,
      "grad_norm": 0.6112158894538879,
      "learning_rate": 0.000155,
      "loss": 0.9645,
      "step": 225
    },
    {
      "epoch": 0.3671189146049481,
      "grad_norm": 0.6465075016021729,
      "learning_rate": 0.000154,
      "loss": 1.0773,
      "step": 230
    },
    {
      "epoch": 0.3671189146049481,
      "eval_loss": 1.004480242729187,
      "eval_runtime": 21.1771,
      "eval_samples_per_second": 1.464,
      "eval_steps_per_second": 0.756,
      "step": 230
    },
    {
      "epoch": 0.3750997605746209,
      "grad_norm": 0.7187998294830322,
      "learning_rate": 0.000153,
      "loss": 0.9467,
      "step": 235
    },
    {
      "epoch": 0.3830806065442937,
      "grad_norm": 0.8026754260063171,
      "learning_rate": 0.000152,
      "loss": 0.9329,
      "step": 240
    },
    {
      "epoch": 0.3830806065442937,
      "eval_loss": 0.9965100288391113,
      "eval_runtime": 21.1786,
      "eval_samples_per_second": 1.464,
      "eval_steps_per_second": 0.755,
      "step": 240
    },
    {
      "epoch": 0.39106145251396646,
      "grad_norm": 0.6788911819458008,
      "learning_rate": 0.000151,
      "loss": 0.8683,
      "step": 245
    },
    {
      "epoch": 0.39904229848363926,
      "grad_norm": 0.6607213020324707,
      "learning_rate": 0.00015000000000000001,
      "loss": 0.9826,
      "step": 250
    },
    {
      "epoch": 0.39904229848363926,
      "eval_loss": 1.0054864883422852,
      "eval_runtime": 21.1094,
      "eval_samples_per_second": 1.469,
      "eval_steps_per_second": 0.758,
      "step": 250
    },
    {
      "epoch": 0.40702314445331206,
      "grad_norm": 0.6742439866065979,
      "learning_rate": 0.00014900000000000002,
      "loss": 0.9507,
      "step": 255
    },
    {
      "epoch": 0.41500399042298486,
      "grad_norm": 0.7669168710708618,
      "learning_rate": 0.000148,
      "loss": 0.9609,
      "step": 260
    },
    {
      "epoch": 0.41500399042298486,
      "eval_loss": 1.0046597719192505,
      "eval_runtime": 21.2545,
      "eval_samples_per_second": 1.459,
      "eval_steps_per_second": 0.753,
      "step": 260
    },
    {
      "epoch": 0.4229848363926576,
      "grad_norm": 0.6186738014221191,
      "learning_rate": 0.000147,
      "loss": 0.9875,
      "step": 265
    },
    {
      "epoch": 0.4309656823623304,
      "grad_norm": 0.633393406867981,
      "learning_rate": 0.000146,
      "loss": 0.9949,
      "step": 270
    },
    {
      "epoch": 0.4309656823623304,
      "eval_loss": 0.9924024343490601,
      "eval_runtime": 20.9746,
      "eval_samples_per_second": 1.478,
      "eval_steps_per_second": 0.763,
      "step": 270
    },
    {
      "epoch": 0.4389465283320032,
      "grad_norm": 0.5856119394302368,
      "learning_rate": 0.000145,
      "loss": 0.9611,
      "step": 275
    },
    {
      "epoch": 0.44692737430167595,
      "grad_norm": 0.5712947249412537,
      "learning_rate": 0.000144,
      "loss": 1.0001,
      "step": 280
    },
    {
      "epoch": 0.44692737430167595,
      "eval_loss": 0.9949964284896851,
      "eval_runtime": 21.5928,
      "eval_samples_per_second": 1.436,
      "eval_steps_per_second": 0.741,
      "step": 280
    },
    {
      "epoch": 0.45490822027134875,
      "grad_norm": 0.6601425409317017,
      "learning_rate": 0.000143,
      "loss": 0.9198,
      "step": 285
    },
    {
      "epoch": 0.46288906624102155,
      "grad_norm": 0.6215651631355286,
      "learning_rate": 0.000142,
      "loss": 1.0944,
      "step": 290
    },
    {
      "epoch": 0.46288906624102155,
      "eval_loss": 0.9959576725959778,
      "eval_runtime": 21.3626,
      "eval_samples_per_second": 1.451,
      "eval_steps_per_second": 0.749,
      "step": 290
    },
    {
      "epoch": 0.47086991221069435,
      "grad_norm": 0.6404448747634888,
      "learning_rate": 0.000141,
      "loss": 0.9556,
      "step": 295
    },
    {
      "epoch": 0.4788507581803671,
      "grad_norm": 0.6889925003051758,
      "learning_rate": 0.00014,
      "loss": 1.011,
      "step": 300
    },
    {
      "epoch": 0.4788507581803671,
      "eval_loss": 0.9947748780250549,
      "eval_runtime": 21.4252,
      "eval_samples_per_second": 1.447,
      "eval_steps_per_second": 0.747,
      "step": 300
    },
    {
      "epoch": 0.4868316041500399,
      "grad_norm": 0.6349315047264099,
      "learning_rate": 0.000139,
      "loss": 0.9343,
      "step": 305
    },
    {
      "epoch": 0.4948124501197127,
      "grad_norm": 0.753274142742157,
      "learning_rate": 0.000138,
      "loss": 0.9356,
      "step": 310
    },
    {
      "epoch": 0.4948124501197127,
      "eval_loss": 0.9897598624229431,
      "eval_runtime": 21.7041,
      "eval_samples_per_second": 1.428,
      "eval_steps_per_second": 0.737,
      "step": 310
    },
    {
      "epoch": 0.5027932960893855,
      "grad_norm": 0.6351450085639954,
      "learning_rate": 0.00013700000000000002,
      "loss": 0.9691,
      "step": 315
    },
    {
      "epoch": 0.5107741420590582,
      "grad_norm": 0.6660261154174805,
      "learning_rate": 0.00013600000000000003,
      "loss": 0.9213,
      "step": 320
    },
    {
      "epoch": 0.5107741420590582,
      "eval_loss": 0.9789547920227051,
      "eval_runtime": 20.997,
      "eval_samples_per_second": 1.476,
      "eval_steps_per_second": 0.762,
      "step": 320
    },
    {
      "epoch": 0.518754988028731,
      "grad_norm": 0.7038460969924927,
      "learning_rate": 0.00013500000000000003,
      "loss": 0.8654,
      "step": 325
    },
    {
      "epoch": 0.5267358339984038,
      "grad_norm": 0.6426422595977783,
      "learning_rate": 0.000134,
      "loss": 0.8611,
      "step": 330
    },
    {
      "epoch": 0.5267358339984038,
      "eval_loss": 0.9792807698249817,
      "eval_runtime": 20.7975,
      "eval_samples_per_second": 1.491,
      "eval_steps_per_second": 0.769,
      "step": 330
    },
    {
      "epoch": 0.5347166799680766,
      "grad_norm": 0.5722155570983887,
      "learning_rate": 0.000133,
      "loss": 0.961,
      "step": 335
    },
    {
      "epoch": 0.5426975259377494,
      "grad_norm": 0.7792693376541138,
      "learning_rate": 0.000132,
      "loss": 0.9647,
      "step": 340
    },
    {
      "epoch": 0.5426975259377494,
      "eval_loss": 0.9812614321708679,
      "eval_runtime": 21.6024,
      "eval_samples_per_second": 1.435,
      "eval_steps_per_second": 0.741,
      "step": 340
    },
    {
      "epoch": 0.5506783719074222,
      "grad_norm": 0.7510095834732056,
      "learning_rate": 0.000131,
      "loss": 0.817,
      "step": 345
    },
    {
      "epoch": 0.5586592178770949,
      "grad_norm": 0.6685438752174377,
      "learning_rate": 0.00013000000000000002,
      "loss": 0.8392,
      "step": 350
    },
    {
      "epoch": 0.5586592178770949,
      "eval_loss": 0.9788187742233276,
      "eval_runtime": 21.2885,
      "eval_samples_per_second": 1.456,
      "eval_steps_per_second": 0.752,
      "step": 350
    },
    {
      "epoch": 0.5666400638467678,
      "grad_norm": 0.6660378575325012,
      "learning_rate": 0.00012900000000000002,
      "loss": 0.9494,
      "step": 355
    },
    {
      "epoch": 0.5746209098164405,
      "grad_norm": 0.6914508938789368,
      "learning_rate": 0.00012800000000000002,
      "loss": 1.0105,
      "step": 360
    },
    {
      "epoch": 0.5746209098164405,
      "eval_loss": 0.9770519733428955,
      "eval_runtime": 21.0568,
      "eval_samples_per_second": 1.472,
      "eval_steps_per_second": 0.76,
      "step": 360
    },
    {
      "epoch": 0.5826017557861133,
      "grad_norm": 0.665715217590332,
      "learning_rate": 0.000127,
      "loss": 0.9159,
      "step": 365
    },
    {
      "epoch": 0.5905826017557861,
      "grad_norm": 0.686400830745697,
      "learning_rate": 0.000126,
      "loss": 1.0525,
      "step": 370
    },
    {
      "epoch": 0.5905826017557861,
      "eval_loss": 0.9769095778465271,
      "eval_runtime": 20.9592,
      "eval_samples_per_second": 1.479,
      "eval_steps_per_second": 0.763,
      "step": 370
    },
    {
      "epoch": 0.5985634477254589,
      "grad_norm": 0.6291290521621704,
      "learning_rate": 0.000125,
      "loss": 0.801,
      "step": 375
    },
    {
      "epoch": 0.6065442936951316,
      "grad_norm": 0.7124271988868713,
      "learning_rate": 0.000124,
      "loss": 0.8913,
      "step": 380
    },
    {
      "epoch": 0.6065442936951316,
      "eval_loss": 0.9799799919128418,
      "eval_runtime": 21.397,
      "eval_samples_per_second": 1.449,
      "eval_steps_per_second": 0.748,
      "step": 380
    },
    {
      "epoch": 0.6145251396648045,
      "grad_norm": 0.8505049347877502,
      "learning_rate": 0.000123,
      "loss": 0.9319,
      "step": 385
    },
    {
      "epoch": 0.6225059856344772,
      "grad_norm": 0.7308720350265503,
      "learning_rate": 0.000122,
      "loss": 0.8906,
      "step": 390
    },
    {
      "epoch": 0.6225059856344772,
      "eval_loss": 0.9712496995925903,
      "eval_runtime": 21.2172,
      "eval_samples_per_second": 1.461,
      "eval_steps_per_second": 0.754,
      "step": 390
    },
    {
      "epoch": 0.6304868316041501,
      "grad_norm": 0.6782912611961365,
      "learning_rate": 0.000121,
      "loss": 0.8928,
      "step": 395
    },
    {
      "epoch": 0.6384676775738228,
      "grad_norm": 0.5951765179634094,
      "learning_rate": 0.00012,
      "loss": 0.9059,
      "step": 400
    },
    {
      "epoch": 0.6384676775738228,
      "eval_loss": 0.9641667008399963,
      "eval_runtime": 20.89,
      "eval_samples_per_second": 1.484,
      "eval_steps_per_second": 0.766,
      "step": 400
    },
    {
      "epoch": 0.6464485235434956,
      "grad_norm": 0.7105025053024292,
      "learning_rate": 0.000119,
      "loss": 0.8986,
      "step": 405
    },
    {
      "epoch": 0.6544293695131684,
      "grad_norm": 0.835155725479126,
      "learning_rate": 0.000118,
      "loss": 0.9309,
      "step": 410
    },
    {
      "epoch": 0.6544293695131684,
      "eval_loss": 0.9708338975906372,
      "eval_runtime": 21.5141,
      "eval_samples_per_second": 1.441,
      "eval_steps_per_second": 0.744,
      "step": 410
    },
    {
      "epoch": 0.6624102154828412,
      "grad_norm": 0.7432790994644165,
      "learning_rate": 0.000117,
      "loss": 0.9653,
      "step": 415
    },
    {
      "epoch": 0.6703910614525139,
      "grad_norm": 0.590387761592865,
      "learning_rate": 0.000116,
      "loss": 0.8429,
      "step": 420
    },
    {
      "epoch": 0.6703910614525139,
      "eval_loss": 0.9680631160736084,
      "eval_runtime": 21.0468,
      "eval_samples_per_second": 1.473,
      "eval_steps_per_second": 0.76,
      "step": 420
    },
    {
      "epoch": 0.6783719074221868,
      "grad_norm": 0.6864650249481201,
      "learning_rate": 0.00011499999999999999,
      "loss": 0.9585,
      "step": 425
    },
    {
      "epoch": 0.6863527533918595,
      "grad_norm": 0.6845282912254333,
      "learning_rate": 0.00011399999999999999,
      "loss": 0.9032,
      "step": 430
    },
    {
      "epoch": 0.6863527533918595,
      "eval_loss": 0.96528559923172,
      "eval_runtime": 21.6521,
      "eval_samples_per_second": 1.432,
      "eval_steps_per_second": 0.739,
      "step": 430
    },
    {
      "epoch": 0.6943335993615323,
      "grad_norm": 0.717858612537384,
      "learning_rate": 0.000113,
      "loss": 0.9401,
      "step": 435
    },
    {
      "epoch": 0.7023144453312051,
      "grad_norm": 0.6566243171691895,
      "learning_rate": 0.00011200000000000001,
      "loss": 0.85,
      "step": 440
    },
    {
      "epoch": 0.7023144453312051,
      "eval_loss": 0.9651073217391968,
      "eval_runtime": 21.2202,
      "eval_samples_per_second": 1.461,
      "eval_steps_per_second": 0.754,
      "step": 440
    },
    {
      "epoch": 0.7102952913008779,
      "grad_norm": 0.6869224905967712,
      "learning_rate": 0.00011100000000000001,
      "loss": 0.8701,
      "step": 445
    },
    {
      "epoch": 0.7182761372705507,
      "grad_norm": 0.5897542834281921,
      "learning_rate": 0.00011000000000000002,
      "loss": 0.9531,
      "step": 450
    },
    {
      "epoch": 0.7182761372705507,
      "eval_loss": 0.9623759984970093,
      "eval_runtime": 21.1274,
      "eval_samples_per_second": 1.467,
      "eval_steps_per_second": 0.757,
      "step": 450
    },
    {
      "epoch": 0.7262569832402235,
      "grad_norm": 0.7018331289291382,
      "learning_rate": 0.000109,
      "loss": 0.8888,
      "step": 455
    },
    {
      "epoch": 0.7342378292098962,
      "grad_norm": 0.6487181186676025,
      "learning_rate": 0.00010800000000000001,
      "loss": 0.8862,
      "step": 460
    },
    {
      "epoch": 0.7342378292098962,
      "eval_loss": 0.9587427973747253,
      "eval_runtime": 21.6398,
      "eval_samples_per_second": 1.433,
      "eval_steps_per_second": 0.739,
      "step": 460
    },
    {
      "epoch": 0.7422186751795691,
      "grad_norm": 0.8121400475502014,
      "learning_rate": 0.00010700000000000001,
      "loss": 0.8856,
      "step": 465
    },
    {
      "epoch": 0.7501995211492418,
      "grad_norm": 0.6554559469223022,
      "learning_rate": 0.00010600000000000002,
      "loss": 0.8736,
      "step": 470
    },
    {
      "epoch": 0.7501995211492418,
      "eval_loss": 0.9547150731086731,
      "eval_runtime": 21.1236,
      "eval_samples_per_second": 1.468,
      "eval_steps_per_second": 0.757,
      "step": 470
    },
    {
      "epoch": 0.7581803671189146,
      "grad_norm": 0.7241584062576294,
      "learning_rate": 0.000105,
      "loss": 0.8002,
      "step": 475
    },
    {
      "epoch": 0.7661612130885874,
      "grad_norm": 0.7639961242675781,
      "learning_rate": 0.00010400000000000001,
      "loss": 0.93,
      "step": 480
    },
    {
      "epoch": 0.7661612130885874,
      "eval_loss": 0.9609348773956299,
      "eval_runtime": 21.2046,
      "eval_samples_per_second": 1.462,
      "eval_steps_per_second": 0.755,
      "step": 480
    },
    {
      "epoch": 0.7741420590582602,
      "grad_norm": 0.6694309115409851,
      "learning_rate": 0.00010300000000000001,
      "loss": 0.8696,
      "step": 485
    },
    {
      "epoch": 0.7821229050279329,
      "grad_norm": 0.6586005687713623,
      "learning_rate": 0.00010200000000000001,
      "loss": 0.9035,
      "step": 490
    },
    {
      "epoch": 0.7821229050279329,
      "eval_loss": 0.9565220475196838,
      "eval_runtime": 21.5471,
      "eval_samples_per_second": 1.439,
      "eval_steps_per_second": 0.743,
      "step": 490
    },
    {
      "epoch": 0.7901037509976058,
      "grad_norm": 0.6772391200065613,
      "learning_rate": 0.000101,
      "loss": 0.9016,
      "step": 495
    },
    {
      "epoch": 0.7980845969672785,
      "grad_norm": 0.7700143456459045,
      "learning_rate": 0.0001,
      "loss": 0.8871,
      "step": 500
    },
    {
      "epoch": 0.7980845969672785,
      "eval_loss": 0.9532312750816345,
      "eval_runtime": 21.3527,
      "eval_samples_per_second": 1.452,
      "eval_steps_per_second": 0.749,
      "step": 500
    },
    {
      "epoch": 0.8060654429369514,
      "grad_norm": 0.7088077068328857,
      "learning_rate": 9.900000000000001e-05,
      "loss": 0.9628,
      "step": 505
    },
    {
      "epoch": 0.8140462889066241,
      "grad_norm": 0.7058229446411133,
      "learning_rate": 9.8e-05,
      "loss": 0.8395,
      "step": 510
    },
    {
      "epoch": 0.8140462889066241,
      "eval_loss": 0.9453433752059937,
      "eval_runtime": 20.6711,
      "eval_samples_per_second": 1.5,
      "eval_steps_per_second": 0.774,
      "step": 510
    },
    {
      "epoch": 0.8220271348762969,
      "grad_norm": 0.6609935164451599,
      "learning_rate": 9.7e-05,
      "loss": 0.8693,
      "step": 515
    },
    {
      "epoch": 0.8300079808459697,
      "grad_norm": 0.8539516925811768,
      "learning_rate": 9.6e-05,
      "loss": 0.8661,
      "step": 520
    },
    {
      "epoch": 0.8300079808459697,
      "eval_loss": 0.9413149356842041,
      "eval_runtime": 21.4056,
      "eval_samples_per_second": 1.448,
      "eval_steps_per_second": 0.747,
      "step": 520
    },
    {
      "epoch": 0.8379888268156425,
      "grad_norm": 0.6991361975669861,
      "learning_rate": 9.5e-05,
      "loss": 1.0091,
      "step": 525
    },
    {
      "epoch": 0.8459696727853152,
      "grad_norm": 0.6568119525909424,
      "learning_rate": 9.4e-05,
      "loss": 0.8034,
      "step": 530
    },
    {
      "epoch": 0.8459696727853152,
      "eval_loss": 0.942388117313385,
      "eval_runtime": 21.1726,
      "eval_samples_per_second": 1.464,
      "eval_steps_per_second": 0.756,
      "step": 530
    },
    {
      "epoch": 0.8539505187549881,
      "grad_norm": 0.6941240429878235,
      "learning_rate": 9.300000000000001e-05,
      "loss": 0.8947,
      "step": 535
    },
    {
      "epoch": 0.8619313647246608,
      "grad_norm": 0.717548131942749,
      "learning_rate": 9.200000000000001e-05,
      "loss": 0.9049,
      "step": 540
    },
    {
      "epoch": 0.8619313647246608,
      "eval_loss": 0.9332796335220337,
      "eval_runtime": 21.1935,
      "eval_samples_per_second": 1.463,
      "eval_steps_per_second": 0.755,
      "step": 540
    },
    {
      "epoch": 0.8699122106943336,
      "grad_norm": 0.6851376295089722,
      "learning_rate": 9.1e-05,
      "loss": 0.9446,
      "step": 545
    },
    {
      "epoch": 0.8778930566640064,
      "grad_norm": 0.7654580473899841,
      "learning_rate": 9e-05,
      "loss": 0.8878,
      "step": 550
    },
    {
      "epoch": 0.8778930566640064,
      "eval_loss": 0.9288386106491089,
      "eval_runtime": 21.0235,
      "eval_samples_per_second": 1.475,
      "eval_steps_per_second": 0.761,
      "step": 550
    },
    {
      "epoch": 0.8858739026336792,
      "grad_norm": 0.6864151954650879,
      "learning_rate": 8.900000000000001e-05,
      "loss": 0.8758,
      "step": 555
    },
    {
      "epoch": 0.8938547486033519,
      "grad_norm": 0.6886463761329651,
      "learning_rate": 8.800000000000001e-05,
      "loss": 0.895,
      "step": 560
    },
    {
      "epoch": 0.8938547486033519,
      "eval_loss": 0.9302241802215576,
      "eval_runtime": 21.0245,
      "eval_samples_per_second": 1.474,
      "eval_steps_per_second": 0.761,
      "step": 560
    },
    {
      "epoch": 0.9018355945730248,
      "grad_norm": 0.7045288681983948,
      "learning_rate": 8.7e-05,
      "loss": 0.7896,
      "step": 565
    },
    {
      "epoch": 0.9098164405426975,
      "grad_norm": 0.8066936135292053,
      "learning_rate": 8.6e-05,
      "loss": 0.9071,
      "step": 570
    },
    {
      "epoch": 0.9098164405426975,
      "eval_loss": 0.9250284433364868,
      "eval_runtime": 21.0108,
      "eval_samples_per_second": 1.475,
      "eval_steps_per_second": 0.762,
      "step": 570
    },
    {
      "epoch": 0.9177972865123704,
      "grad_norm": 0.6377101540565491,
      "learning_rate": 8.5e-05,
      "loss": 0.8045,
      "step": 575
    },
    {
      "epoch": 0.9257781324820431,
      "grad_norm": 0.7454017996788025,
      "learning_rate": 8.4e-05,
      "loss": 0.9337,
      "step": 580
    },
    {
      "epoch": 0.9257781324820431,
      "eval_loss": 0.9239058494567871,
      "eval_runtime": 20.9525,
      "eval_samples_per_second": 1.48,
      "eval_steps_per_second": 0.764,
      "step": 580
    },
    {
      "epoch": 0.9337589784517158,
      "grad_norm": 0.6952985525131226,
      "learning_rate": 8.3e-05,
      "loss": 0.8703,
      "step": 585
    },
    {
      "epoch": 0.9417398244213887,
      "grad_norm": 0.7358236312866211,
      "learning_rate": 8.2e-05,
      "loss": 0.8793,
      "step": 590
    },
    {
      "epoch": 0.9417398244213887,
      "eval_loss": 0.9293895363807678,
      "eval_runtime": 20.9523,
      "eval_samples_per_second": 1.48,
      "eval_steps_per_second": 0.764,
      "step": 590
    },
    {
      "epoch": 0.9497206703910615,
      "grad_norm": 0.5817161202430725,
      "learning_rate": 8.1e-05,
      "loss": 0.8298,
      "step": 595
    },
    {
      "epoch": 0.9577015163607342,
      "grad_norm": 0.7251338362693787,
      "learning_rate": 8e-05,
      "loss": 0.8766,
      "step": 600
    },
    {
      "epoch": 0.9577015163607342,
      "eval_loss": 0.9293678998947144,
      "eval_runtime": 20.9457,
      "eval_samples_per_second": 1.48,
      "eval_steps_per_second": 0.764,
      "step": 600
    },
    {
      "epoch": 0.965682362330407,
      "grad_norm": 0.8278694748878479,
      "learning_rate": 7.900000000000001e-05,
      "loss": 0.7507,
      "step": 605
    },
    {
      "epoch": 0.9736632083000798,
      "grad_norm": 0.7793962955474854,
      "learning_rate": 7.800000000000001e-05,
      "loss": 0.8682,
      "step": 610
    },
    {
      "epoch": 0.9736632083000798,
      "eval_loss": 0.9281108975410461,
      "eval_runtime": 21.2355,
      "eval_samples_per_second": 1.46,
      "eval_steps_per_second": 0.753,
      "step": 610
    },
    {
      "epoch": 0.9816440542697525,
      "grad_norm": 0.8250753283500671,
      "learning_rate": 7.7e-05,
      "loss": 0.8196,
      "step": 615
    },
    {
      "epoch": 0.9896249002394254,
      "grad_norm": 0.7124294638633728,
      "learning_rate": 7.6e-05,
      "loss": 0.7849,
      "step": 620
    },
    {
      "epoch": 0.9896249002394254,
      "eval_loss": 0.9263650178909302,
      "eval_runtime": 21.1863,
      "eval_samples_per_second": 1.463,
      "eval_steps_per_second": 0.755,
      "step": 620
    },
    {
      "epoch": 0.9976057462090981,
      "grad_norm": 0.768418550491333,
      "learning_rate": 7.500000000000001e-05,
      "loss": 0.8589,
      "step": 625
    },
    {
      "epoch": 1.005586592178771,
      "grad_norm": 0.6237154603004456,
      "learning_rate": 7.4e-05,
      "loss": 0.8746,
      "step": 630
    },
    {
      "epoch": 1.005586592178771,
      "eval_loss": 0.9216476082801819,
      "eval_runtime": 21.1504,
      "eval_samples_per_second": 1.466,
      "eval_steps_per_second": 0.756,
      "step": 630
    },
    {
      "epoch": 1.0135674381484436,
      "grad_norm": 0.691764235496521,
      "learning_rate": 7.3e-05,
      "loss": 0.8455,
      "step": 635
    },
    {
      "epoch": 1.0215482841181165,
      "grad_norm": 0.6523061990737915,
      "learning_rate": 7.2e-05,
      "loss": 0.8149,
      "step": 640
    },
    {
      "epoch": 1.0215482841181165,
      "eval_loss": 0.9180006384849548,
      "eval_runtime": 21.3153,
      "eval_samples_per_second": 1.454,
      "eval_steps_per_second": 0.751,
      "step": 640
    },
    {
      "epoch": 1.0295291300877893,
      "grad_norm": 0.9022932648658752,
      "learning_rate": 7.1e-05,
      "loss": 0.8321,
      "step": 645
    },
    {
      "epoch": 1.037509976057462,
      "grad_norm": 0.6842184066772461,
      "learning_rate": 7e-05,
      "loss": 0.806,
      "step": 650
    },
    {
      "epoch": 1.037509976057462,
      "eval_loss": 0.9218840599060059,
      "eval_runtime": 21.1263,
      "eval_samples_per_second": 1.467,
      "eval_steps_per_second": 0.757,
      "step": 650
    },
    {
      "epoch": 1.0454908220271348,
      "grad_norm": 0.9547672271728516,
      "learning_rate": 6.9e-05,
      "loss": 0.8023,
      "step": 655
    },
    {
      "epoch": 1.0534716679968077,
      "grad_norm": 0.6439852714538574,
      "learning_rate": 6.800000000000001e-05,
      "loss": 0.7533,
      "step": 660
    },
    {
      "epoch": 1.0534716679968077,
      "eval_loss": 0.9165079593658447,
      "eval_runtime": 21.1104,
      "eval_samples_per_second": 1.468,
      "eval_steps_per_second": 0.758,
      "step": 660
    },
    {
      "epoch": 1.0614525139664805,
      "grad_norm": 0.7850003242492676,
      "learning_rate": 6.7e-05,
      "loss": 0.858,
      "step": 665
    },
    {
      "epoch": 1.0694333599361532,
      "grad_norm": 0.6285898685455322,
      "learning_rate": 6.6e-05,
      "loss": 0.7146,
      "step": 670
    },
    {
      "epoch": 1.0694333599361532,
      "eval_loss": 0.9204993844032288,
      "eval_runtime": 21.1527,
      "eval_samples_per_second": 1.466,
      "eval_steps_per_second": 0.756,
      "step": 670
    },
    {
      "epoch": 1.077414205905826,
      "grad_norm": 0.7350932955741882,
      "learning_rate": 6.500000000000001e-05,
      "loss": 0.8328,
      "step": 675
    },
    {
      "epoch": 1.085395051875499,
      "grad_norm": 0.7773166298866272,
      "learning_rate": 6.400000000000001e-05,
      "loss": 0.8075,
      "step": 680
    },
    {
      "epoch": 1.085395051875499,
      "eval_loss": 0.9216338992118835,
      "eval_runtime": 21.0323,
      "eval_samples_per_second": 1.474,
      "eval_steps_per_second": 0.761,
      "step": 680
    },
    {
      "epoch": 1.0933758978451715,
      "grad_norm": 0.9075634479522705,
      "learning_rate": 6.3e-05,
      "loss": 0.8809,
      "step": 685
    },
    {
      "epoch": 1.1013567438148444,
      "grad_norm": 0.7483291029930115,
      "learning_rate": 6.2e-05,
      "loss": 0.8302,
      "step": 690
    },
    {
      "epoch": 1.1013567438148444,
      "eval_loss": 0.9141597747802734,
      "eval_runtime": 20.8831,
      "eval_samples_per_second": 1.484,
      "eval_steps_per_second": 0.766,
      "step": 690
    },
    {
      "epoch": 1.1093375897845172,
      "grad_norm": 0.8029285073280334,
      "learning_rate": 6.1e-05,
      "loss": 0.7757,
      "step": 695
    },
    {
      "epoch": 1.1173184357541899,
      "grad_norm": 0.7638797760009766,
      "learning_rate": 6e-05,
      "loss": 0.8258,
      "step": 700
    },
    {
      "epoch": 1.1173184357541899,
      "eval_loss": 0.9134471416473389,
      "eval_runtime": 21.3635,
      "eval_samples_per_second": 1.451,
      "eval_steps_per_second": 0.749,
      "step": 700
    },
    {
      "epoch": 1.1252992817238627,
      "grad_norm": 0.8713815212249756,
      "learning_rate": 5.9e-05,
      "loss": 0.8017,
      "step": 705
    },
    {
      "epoch": 1.1332801276935356,
      "grad_norm": 0.8253017663955688,
      "learning_rate": 5.8e-05,
      "loss": 0.8873,
      "step": 710
    },
    {
      "epoch": 1.1332801276935356,
      "eval_loss": 0.91542649269104,
      "eval_runtime": 21.1885,
      "eval_samples_per_second": 1.463,
      "eval_steps_per_second": 0.755,
      "step": 710
    },
    {
      "epoch": 1.1412609736632082,
      "grad_norm": 0.8720806241035461,
      "learning_rate": 5.6999999999999996e-05,
      "loss": 0.8808,
      "step": 715
    },
    {
      "epoch": 1.149241819632881,
      "grad_norm": 0.855076253414154,
      "learning_rate": 5.6000000000000006e-05,
      "loss": 0.8285,
      "step": 720
    },
    {
      "epoch": 1.149241819632881,
      "eval_loss": 0.9051924347877502,
      "eval_runtime": 21.0529,
      "eval_samples_per_second": 1.472,
      "eval_steps_per_second": 0.76,
      "step": 720
    },
    {
      "epoch": 1.157222665602554,
      "grad_norm": 0.8113986849784851,
      "learning_rate": 5.500000000000001e-05,
      "loss": 0.8119,
      "step": 725
    },
    {
      "epoch": 1.1652035115722266,
      "grad_norm": 0.754554271697998,
      "learning_rate": 5.4000000000000005e-05,
      "loss": 0.7881,
      "step": 730
    },
    {
      "epoch": 1.1652035115722266,
      "eval_loss": 0.9082400798797607,
      "eval_runtime": 20.8583,
      "eval_samples_per_second": 1.486,
      "eval_steps_per_second": 0.767,
      "step": 730
    },
    {
      "epoch": 1.1731843575418994,
      "grad_norm": 0.9242857694625854,
      "learning_rate": 5.300000000000001e-05,
      "loss": 0.7508,
      "step": 735
    },
    {
      "epoch": 1.1811652035115723,
      "grad_norm": 0.7203059792518616,
      "learning_rate": 5.2000000000000004e-05,
      "loss": 0.8788,
      "step": 740
    },
    {
      "epoch": 1.1811652035115723,
      "eval_loss": 0.9105547666549683,
      "eval_runtime": 20.8797,
      "eval_samples_per_second": 1.485,
      "eval_steps_per_second": 0.766,
      "step": 740
    },
    {
      "epoch": 1.189146049481245,
      "grad_norm": 0.821240246295929,
      "learning_rate": 5.1000000000000006e-05,
      "loss": 0.767,
      "step": 745
    },
    {
      "epoch": 1.1971268954509178,
      "grad_norm": 1.0884546041488647,
      "learning_rate": 5e-05,
      "loss": 0.8585,
      "step": 750
    },
    {
      "epoch": 1.1971268954509178,
      "eval_loss": 0.9092571139335632,
      "eval_runtime": 21.2995,
      "eval_samples_per_second": 1.455,
      "eval_steps_per_second": 0.751,
      "step": 750
    },
    {
      "epoch": 1.2051077414205906,
      "grad_norm": 0.7183963656425476,
      "learning_rate": 4.9e-05,
      "loss": 0.8462,
      "step": 755
    },
    {
      "epoch": 1.2130885873902635,
      "grad_norm": 0.7556672692298889,
      "learning_rate": 4.8e-05,
      "loss": 0.7203,
      "step": 760
    },
    {
      "epoch": 1.2130885873902635,
      "eval_loss": 0.9125378727912903,
      "eval_runtime": 20.9445,
      "eval_samples_per_second": 1.48,
      "eval_steps_per_second": 0.764,
      "step": 760
    },
    {
      "epoch": 1.2210694333599361,
      "grad_norm": 0.8119292259216309,
      "learning_rate": 4.7e-05,
      "loss": 0.8294,
      "step": 765
    },
    {
      "epoch": 1.229050279329609,
      "grad_norm": 0.7207946181297302,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.8084,
      "step": 770
    },
    {
      "epoch": 1.229050279329609,
      "eval_loss": 0.9099369645118713,
      "eval_runtime": 20.5873,
      "eval_samples_per_second": 1.506,
      "eval_steps_per_second": 0.777,
      "step": 770
    },
    {
      "epoch": 1.2370311252992816,
      "grad_norm": 0.9262428283691406,
      "learning_rate": 4.5e-05,
      "loss": 0.7861,
      "step": 775
    },
    {
      "epoch": 1.2450119712689545,
      "grad_norm": 0.9154334664344788,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.7298,
      "step": 780
    },
    {
      "epoch": 1.2450119712689545,
      "eval_loss": 0.9098423719406128,
      "eval_runtime": 21.1045,
      "eval_samples_per_second": 1.469,
      "eval_steps_per_second": 0.758,
      "step": 780
    },
    {
      "epoch": 1.2529928172386273,
      "grad_norm": 0.8377973437309265,
      "learning_rate": 4.3e-05,
      "loss": 0.7593,
      "step": 785
    },
    {
      "epoch": 1.2609736632083002,
      "grad_norm": 0.8660017848014832,
      "learning_rate": 4.2e-05,
      "loss": 0.7811,
      "step": 790
    },
    {
      "epoch": 1.2609736632083002,
      "eval_loss": 0.9046799540519714,
      "eval_runtime": 20.9842,
      "eval_samples_per_second": 1.477,
      "eval_steps_per_second": 0.762,
      "step": 790
    },
    {
      "epoch": 1.2689545091779728,
      "grad_norm": 0.7700662612915039,
      "learning_rate": 4.1e-05,
      "loss": 0.8493,
      "step": 795
    },
    {
      "epoch": 1.2769353551476457,
      "grad_norm": 0.8167340159416199,
      "learning_rate": 4e-05,
      "loss": 0.7774,
      "step": 800
    },
    {
      "epoch": 1.2769353551476457,
      "eval_loss": 0.8998985886573792,
      "eval_runtime": 21.1467,
      "eval_samples_per_second": 1.466,
      "eval_steps_per_second": 0.757,
      "step": 800
    },
    {
      "epoch": 1.2849162011173183,
      "grad_norm": 0.8162097334861755,
      "learning_rate": 3.9000000000000006e-05,
      "loss": 0.7892,
      "step": 805
    },
    {
      "epoch": 1.2928970470869912,
      "grad_norm": 0.8468261957168579,
      "learning_rate": 3.8e-05,
      "loss": 0.8324,
      "step": 810
    },
    {
      "epoch": 1.2928970470869912,
      "eval_loss": 0.8993770480155945,
      "eval_runtime": 20.9539,
      "eval_samples_per_second": 1.479,
      "eval_steps_per_second": 0.764,
      "step": 810
    },
    {
      "epoch": 1.300877893056664,
      "grad_norm": 0.746923565864563,
      "learning_rate": 3.7e-05,
      "loss": 0.7961,
      "step": 815
    },
    {
      "epoch": 1.3088587390263369,
      "grad_norm": 0.7284272313117981,
      "learning_rate": 3.6e-05,
      "loss": 0.6872,
      "step": 820
    },
    {
      "epoch": 1.3088587390263369,
      "eval_loss": 0.8972273468971252,
      "eval_runtime": 20.6969,
      "eval_samples_per_second": 1.498,
      "eval_steps_per_second": 0.773,
      "step": 820
    },
    {
      "epoch": 1.3168395849960095,
      "grad_norm": 0.8922063708305359,
      "learning_rate": 3.5e-05,
      "loss": 0.8041,
      "step": 825
    },
    {
      "epoch": 1.3248204309656824,
      "grad_norm": 0.7598963975906372,
      "learning_rate": 3.4000000000000007e-05,
      "loss": 0.7028,
      "step": 830
    },
    {
      "epoch": 1.3248204309656824,
      "eval_loss": 0.8973636031150818,
      "eval_runtime": 20.8738,
      "eval_samples_per_second": 1.485,
      "eval_steps_per_second": 0.767,
      "step": 830
    },
    {
      "epoch": 1.3328012769353552,
      "grad_norm": 0.7866420745849609,
      "learning_rate": 3.3e-05,
      "loss": 0.8392,
      "step": 835
    },
    {
      "epoch": 1.3407821229050279,
      "grad_norm": 0.8417834639549255,
      "learning_rate": 3.2000000000000005e-05,
      "loss": 0.7329,
      "step": 840
    },
    {
      "epoch": 1.3407821229050279,
      "eval_loss": 0.9000219106674194,
      "eval_runtime": 21.0159,
      "eval_samples_per_second": 1.475,
      "eval_steps_per_second": 0.761,
      "step": 840
    },
    {
      "epoch": 1.3487629688747007,
      "grad_norm": 0.8085408806800842,
      "learning_rate": 3.1e-05,
      "loss": 0.7452,
      "step": 845
    },
    {
      "epoch": 1.3567438148443736,
      "grad_norm": 0.9095738530158997,
      "learning_rate": 3e-05,
      "loss": 0.8114,
      "step": 850
    },
    {
      "epoch": 1.3567438148443736,
      "eval_loss": 0.899355411529541,
      "eval_runtime": 21.1951,
      "eval_samples_per_second": 1.463,
      "eval_steps_per_second": 0.755,
      "step": 850
    },
    {
      "epoch": 1.3647246608140464,
      "grad_norm": 0.8077012896537781,
      "learning_rate": 2.9e-05,
      "loss": 0.7842,
      "step": 855
    },
    {
      "epoch": 1.372705506783719,
      "grad_norm": 0.9387809038162231,
      "learning_rate": 2.8000000000000003e-05,
      "loss": 0.7739,
      "step": 860
    },
    {
      "epoch": 1.372705506783719,
      "eval_loss": 0.9006592631340027,
      "eval_runtime": 20.7294,
      "eval_samples_per_second": 1.495,
      "eval_steps_per_second": 0.772,
      "step": 860
    },
    {
      "epoch": 1.380686352753392,
      "grad_norm": 0.8230510950088501,
      "learning_rate": 2.7000000000000002e-05,
      "loss": 0.7765,
      "step": 865
    },
    {
      "epoch": 1.3886671987230645,
      "grad_norm": 0.9421499371528625,
      "learning_rate": 2.6000000000000002e-05,
      "loss": 0.7469,
      "step": 870
    },
    {
      "epoch": 1.3886671987230645,
      "eval_loss": 0.8988057374954224,
      "eval_runtime": 20.7795,
      "eval_samples_per_second": 1.492,
      "eval_steps_per_second": 0.77,
      "step": 870
    },
    {
      "epoch": 1.3966480446927374,
      "grad_norm": 0.8720149993896484,
      "learning_rate": 2.5e-05,
      "loss": 0.7674,
      "step": 875
    },
    {
      "epoch": 1.4046288906624103,
      "grad_norm": 0.9915176033973694,
      "learning_rate": 2.4e-05,
      "loss": 0.7925,
      "step": 880
    },
    {
      "epoch": 1.4046288906624103,
      "eval_loss": 0.8960800170898438,
      "eval_runtime": 21.1416,
      "eval_samples_per_second": 1.466,
      "eval_steps_per_second": 0.757,
      "step": 880
    },
    {
      "epoch": 1.4126097366320831,
      "grad_norm": 0.8962202072143555,
      "learning_rate": 2.3000000000000003e-05,
      "loss": 0.7848,
      "step": 885
    },
    {
      "epoch": 1.4205905826017557,
      "grad_norm": 0.9091677069664001,
      "learning_rate": 2.2000000000000003e-05,
      "loss": 0.8638,
      "step": 890
    },
    {
      "epoch": 1.4205905826017557,
      "eval_loss": 0.893966555595398,
      "eval_runtime": 21.0044,
      "eval_samples_per_second": 1.476,
      "eval_steps_per_second": 0.762,
      "step": 890
    },
    {
      "epoch": 1.4285714285714286,
      "grad_norm": 0.7839522361755371,
      "learning_rate": 2.1e-05,
      "loss": 0.8426,
      "step": 895
    },
    {
      "epoch": 1.4365522745411012,
      "grad_norm": 0.8735964894294739,
      "learning_rate": 2e-05,
      "loss": 0.852,
      "step": 900
    },
    {
      "epoch": 1.4365522745411012,
      "eval_loss": 0.8946609497070312,
      "eval_runtime": 20.7777,
      "eval_samples_per_second": 1.492,
      "eval_steps_per_second": 0.77,
      "step": 900
    },
    {
      "epoch": 1.444533120510774,
      "grad_norm": 0.8208412528038025,
      "learning_rate": 1.9e-05,
      "loss": 0.7983,
      "step": 905
    },
    {
      "epoch": 1.452513966480447,
      "grad_norm": 0.9176042675971985,
      "learning_rate": 1.8e-05,
      "loss": 0.8296,
      "step": 910
    },
    {
      "epoch": 1.452513966480447,
      "eval_loss": 0.8942002654075623,
      "eval_runtime": 21.0465,
      "eval_samples_per_second": 1.473,
      "eval_steps_per_second": 0.76,
      "step": 910
    },
    {
      "epoch": 1.4604948124501198,
      "grad_norm": 0.7592383027076721,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 0.809,
      "step": 915
    },
    {
      "epoch": 1.4684756584197924,
      "grad_norm": 0.8816229701042175,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 0.7502,
      "step": 920
    },
    {
      "epoch": 1.4684756584197924,
      "eval_loss": 0.8975324034690857,
      "eval_runtime": 21.1221,
      "eval_samples_per_second": 1.468,
      "eval_steps_per_second": 0.758,
      "step": 920
    },
    {
      "epoch": 1.4764565043894653,
      "grad_norm": 0.8397924900054932,
      "learning_rate": 1.5e-05,
      "loss": 0.791,
      "step": 925
    },
    {
      "epoch": 1.484437350359138,
      "grad_norm": 0.9116021394729614,
      "learning_rate": 1.4000000000000001e-05,
      "loss": 0.77,
      "step": 930
    },
    {
      "epoch": 1.484437350359138,
      "eval_loss": 0.8986864686012268,
      "eval_runtime": 21.0542,
      "eval_samples_per_second": 1.472,
      "eval_steps_per_second": 0.76,
      "step": 930
    },
    {
      "epoch": 1.4924181963288108,
      "grad_norm": 0.8789558410644531,
      "learning_rate": 1.3000000000000001e-05,
      "loss": 0.8362,
      "step": 935
    },
    {
      "epoch": 1.5003990422984836,
      "grad_norm": 1.1174862384796143,
      "learning_rate": 1.2e-05,
      "loss": 0.7821,
      "step": 940
    },
    {
      "epoch": 1.5003990422984836,
      "eval_loss": 0.8970634341239929,
      "eval_runtime": 20.7967,
      "eval_samples_per_second": 1.491,
      "eval_steps_per_second": 0.769,
      "step": 940
    },
    {
      "epoch": 1.5083798882681565,
      "grad_norm": 0.8152345418930054,
      "learning_rate": 1.1000000000000001e-05,
      "loss": 0.7953,
      "step": 945
    },
    {
      "epoch": 1.5163607342378294,
      "grad_norm": 0.8047603964805603,
      "learning_rate": 1e-05,
      "loss": 0.7759,
      "step": 950
    },
    {
      "epoch": 1.5163607342378294,
      "eval_loss": 0.8968480229377747,
      "eval_runtime": 21.2746,
      "eval_samples_per_second": 1.457,
      "eval_steps_per_second": 0.752,
      "step": 950
    },
    {
      "epoch": 1.524341580207502,
      "grad_norm": 0.8262729048728943,
      "learning_rate": 9e-06,
      "loss": 0.8128,
      "step": 955
    },
    {
      "epoch": 1.5323224261771746,
      "grad_norm": 0.8479042649269104,
      "learning_rate": 8.000000000000001e-06,
      "loss": 0.7778,
      "step": 960
    },
    {
      "epoch": 1.5323224261771746,
      "eval_loss": 0.8949146270751953,
      "eval_runtime": 21.0684,
      "eval_samples_per_second": 1.471,
      "eval_steps_per_second": 0.759,
      "step": 960
    },
    {
      "epoch": 1.5403032721468475,
      "grad_norm": 0.7638890147209167,
      "learning_rate": 7.000000000000001e-06,
      "loss": 0.8371,
      "step": 965
    },
    {
      "epoch": 1.5482841181165203,
      "grad_norm": 1.009861707687378,
      "learning_rate": 6e-06,
      "loss": 0.6906,
      "step": 970
    },
    {
      "epoch": 1.5482841181165203,
      "eval_loss": 0.8940587043762207,
      "eval_runtime": 21.4138,
      "eval_samples_per_second": 1.448,
      "eval_steps_per_second": 0.747,
      "step": 970
    },
    {
      "epoch": 1.5562649640861932,
      "grad_norm": 0.8926388025283813,
      "learning_rate": 5e-06,
      "loss": 0.7935,
      "step": 975
    },
    {
      "epoch": 1.564245810055866,
      "grad_norm": 0.7741861939430237,
      "learning_rate": 4.000000000000001e-06,
      "loss": 0.8137,
      "step": 980
    },
    {
      "epoch": 1.564245810055866,
      "eval_loss": 0.895005464553833,
      "eval_runtime": 20.8613,
      "eval_samples_per_second": 1.486,
      "eval_steps_per_second": 0.767,
      "step": 980
    },
    {
      "epoch": 1.5722266560255387,
      "grad_norm": 0.950386106967926,
      "learning_rate": 3e-06,
      "loss": 0.6965,
      "step": 985
    },
    {
      "epoch": 1.5802075019952115,
      "grad_norm": 0.846916675567627,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 0.799,
      "step": 990
    },
    {
      "epoch": 1.5802075019952115,
      "eval_loss": 0.8942360281944275,
      "eval_runtime": 21.1875,
      "eval_samples_per_second": 1.463,
      "eval_steps_per_second": 0.755,
      "step": 990
    },
    {
      "epoch": 1.5881883479648842,
      "grad_norm": 0.8533262610435486,
      "learning_rate": 1.0000000000000002e-06,
      "loss": 0.753,
      "step": 995
    },
    {
      "epoch": 1.596169193934557,
      "grad_norm": 0.7492548227310181,
      "learning_rate": 0.0,
      "loss": 0.7454,
      "step": 1000
    },
    {
      "epoch": 1.596169193934557,
      "eval_loss": 0.8934003114700317,
      "eval_runtime": 21.313,
      "eval_samples_per_second": 1.455,
      "eval_steps_per_second": 0.751,
      "step": 1000
    }
  ],
  "logging_steps": 5,
  "max_steps": 1000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 25,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 5.292030717299424e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}