{
  "best_metric": 1.9184670448303223,
  "best_model_checkpoint": "/home/datta0/models/lora_final/Qwen2-7B_pct_ortho_r16/checkpoint-32",
  "epoch": 0.9996779388083736,
  "eval_steps": 8,
  "global_step": 388,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0025764895330112722,
      "grad_norm": 2.5038321018218994,
      "learning_rate": 1.25e-05,
      "loss": 2.0463,
      "step": 1
    },
    {
      "epoch": 0.010305958132045089,
      "grad_norm": 2.0884063243865967,
      "learning_rate": 5e-05,
      "loss": 2.1379,
      "step": 4
    },
    {
      "epoch": 0.020611916264090178,
      "grad_norm": 2.1778032779693604,
      "learning_rate": 0.0001,
      "loss": 2.0486,
      "step": 8
    },
    {
      "epoch": 0.020611916264090178,
      "eval_loss": 1.997605562210083,
      "eval_runtime": 130.7432,
      "eval_samples_per_second": 1.874,
      "eval_steps_per_second": 0.941,
      "step": 8
    },
    {
      "epoch": 0.030917874396135265,
      "grad_norm": 1.3578734397888184,
      "learning_rate": 9.997266286704631e-05,
      "loss": 2.0017,
      "step": 12
    },
    {
      "epoch": 0.041223832528180356,
      "grad_norm": 1.0907150506973267,
      "learning_rate": 9.989068136093873e-05,
      "loss": 1.9839,
      "step": 16
    },
    {
      "epoch": 0.041223832528180356,
      "eval_loss": 1.9347867965698242,
      "eval_runtime": 133.851,
      "eval_samples_per_second": 1.83,
      "eval_steps_per_second": 0.919,
      "step": 16
    },
    {
      "epoch": 0.05152979066022544,
      "grad_norm": 1.1747814416885376,
      "learning_rate": 9.975414512725057e-05,
      "loss": 1.9421,
      "step": 20
    },
    {
      "epoch": 0.06183574879227053,
      "grad_norm": 1.2741020917892456,
      "learning_rate": 9.956320346634876e-05,
      "loss": 2.0083,
      "step": 24
    },
    {
      "epoch": 0.06183574879227053,
      "eval_loss": 1.9231261014938354,
      "eval_runtime": 123.8213,
      "eval_samples_per_second": 1.979,
      "eval_steps_per_second": 0.993,
      "step": 24
    },
    {
      "epoch": 0.07214170692431562,
      "grad_norm": 1.1227424144744873,
      "learning_rate": 9.931806517013612e-05,
      "loss": 1.9946,
      "step": 28
    },
    {
      "epoch": 0.08244766505636071,
      "grad_norm": 1.14728581905365,
      "learning_rate": 9.901899829374047e-05,
      "loss": 1.923,
      "step": 32
    },
    {
      "epoch": 0.08244766505636071,
      "eval_loss": 1.9184670448303223,
      "eval_runtime": 236.0583,
      "eval_samples_per_second": 1.038,
      "eval_steps_per_second": 0.521,
      "step": 32
    },
    {
      "epoch": 0.0927536231884058,
      "grad_norm": 1.0226891040802002,
      "learning_rate": 9.86663298624003e-05,
      "loss": 1.9968,
      "step": 36
    },
    {
      "epoch": 0.10305958132045089,
      "grad_norm": 1.1554687023162842,
      "learning_rate": 9.826044551386744e-05,
      "loss": 1.9734,
      "step": 40
    },
    {
      "epoch": 0.10305958132045089,
      "eval_loss": 1.9200371503829956,
      "eval_runtime": 131.8583,
      "eval_samples_per_second": 1.858,
      "eval_steps_per_second": 0.933,
      "step": 40
    },
    {
      "epoch": 0.11336553945249597,
      "grad_norm": 1.1349327564239502,
      "learning_rate": 9.780178907671789e-05,
      "loss": 1.8989,
      "step": 44
    },
    {
      "epoch": 0.12367149758454106,
      "grad_norm": 0.9330633282661438,
      "learning_rate": 9.729086208503174e-05,
      "loss": 1.9761,
      "step": 48
    },
    {
      "epoch": 0.12367149758454106,
      "eval_loss": 1.9230433702468872,
      "eval_runtime": 117.734,
      "eval_samples_per_second": 2.081,
      "eval_steps_per_second": 1.045,
      "step": 48
    },
    {
      "epoch": 0.13397745571658615,
      "grad_norm": 1.2882357835769653,
      "learning_rate": 9.672822322997305e-05,
      "loss": 1.9139,
      "step": 52
    },
    {
      "epoch": 0.14428341384863125,
      "grad_norm": 1.2212823629379272,
      "learning_rate": 9.611448774886924e-05,
      "loss": 1.9869,
      "step": 56
    },
    {
      "epoch": 0.14428341384863125,
      "eval_loss": 1.922610878944397,
      "eval_runtime": 133.6918,
      "eval_samples_per_second": 1.833,
      "eval_steps_per_second": 0.92,
      "step": 56
    },
    {
      "epoch": 0.15458937198067632,
      "grad_norm": 1.14687979221344,
      "learning_rate": 9.545032675245813e-05,
      "loss": 1.9617,
      "step": 60
    },
    {
      "epoch": 0.16489533011272142,
      "grad_norm": 1.2522599697113037,
      "learning_rate": 9.473646649103818e-05,
      "loss": 1.9196,
      "step": 64
    },
    {
      "epoch": 0.16489533011272142,
      "eval_loss": 1.923781156539917,
      "eval_runtime": 122.9351,
      "eval_samples_per_second": 1.993,
      "eval_steps_per_second": 1.001,
      "step": 64
    },
    {
      "epoch": 0.1752012882447665,
      "grad_norm": 1.1364014148712158,
      "learning_rate": 9.397368756032445e-05,
      "loss": 2.0391,
      "step": 68
    },
    {
      "epoch": 0.1855072463768116,
      "grad_norm": 1.2489475011825562,
      "learning_rate": 9.316282404787871e-05,
      "loss": 1.9871,
      "step": 72
    },
    {
      "epoch": 0.1855072463768116,
      "eval_loss": 1.9276381731033325,
      "eval_runtime": 228.2046,
      "eval_samples_per_second": 1.074,
      "eval_steps_per_second": 0.539,
      "step": 72
    },
    {
      "epoch": 0.19581320450885667,
      "grad_norm": 1.3833423852920532,
      "learning_rate": 9.230476262104677e-05,
      "loss": 2.0093,
      "step": 76
    },
    {
      "epoch": 0.20611916264090177,
      "grad_norm": 1.1054407358169556,
      "learning_rate": 9.140044155740101e-05,
      "loss": 2.0064,
      "step": 80
    },
    {
      "epoch": 0.20611916264090177,
      "eval_loss": 1.9251201152801514,
      "eval_runtime": 129.5681,
      "eval_samples_per_second": 1.891,
      "eval_steps_per_second": 0.949,
      "step": 80
    },
    {
      "epoch": 0.21642512077294687,
      "grad_norm": 1.2429131269454956,
      "learning_rate": 9.045084971874738e-05,
      "loss": 2.064,
      "step": 84
    },
    {
      "epoch": 0.22673107890499195,
      "grad_norm": 1.23136568069458,
      "learning_rate": 8.945702546981969e-05,
      "loss": 1.9864,
      "step": 88
    },
    {
      "epoch": 0.22673107890499195,
      "eval_loss": 1.9282381534576416,
      "eval_runtime": 127.4266,
      "eval_samples_per_second": 1.923,
      "eval_steps_per_second": 0.965,
      "step": 88
    },
    {
      "epoch": 0.23703703703703705,
      "grad_norm": 1.0550172328948975,
      "learning_rate": 8.842005554284296e-05,
      "loss": 1.903,
      "step": 92
    },
    {
      "epoch": 0.24734299516908212,
      "grad_norm": 1.214440941810608,
      "learning_rate": 8.73410738492077e-05,
      "loss": 1.9204,
      "step": 96
    },
    {
      "epoch": 0.24734299516908212,
      "eval_loss": 1.9318650960922241,
      "eval_runtime": 136.7115,
      "eval_samples_per_second": 1.792,
      "eval_steps_per_second": 0.9,
      "step": 96
    },
    {
      "epoch": 0.2576489533011272,
      "grad_norm": 1.0632654428482056,
      "learning_rate": 8.622126023955446e-05,
      "loss": 2.0197,
      "step": 100
    },
    {
      "epoch": 0.2679549114331723,
      "grad_norm": 1.2511614561080933,
      "learning_rate": 8.506183921362443e-05,
      "loss": 2.0003,
      "step": 104
    },
    {
      "epoch": 0.2679549114331723,
      "eval_loss": 1.9295417070388794,
      "eval_runtime": 133.2173,
      "eval_samples_per_second": 1.839,
      "eval_steps_per_second": 0.923,
      "step": 104
    },
    {
      "epoch": 0.2782608695652174,
      "grad_norm": 1.1069486141204834,
      "learning_rate": 8.386407858128706e-05,
      "loss": 2.0178,
      "step": 108
    },
    {
      "epoch": 0.2885668276972625,
      "grad_norm": 1.060439944267273,
      "learning_rate": 8.262928807620843e-05,
      "loss": 1.8821,
      "step": 112
    },
    {
      "epoch": 0.2885668276972625,
      "eval_loss": 1.9357030391693115,
      "eval_runtime": 233.1725,
      "eval_samples_per_second": 1.051,
      "eval_steps_per_second": 0.528,
      "step": 112
    },
    {
      "epoch": 0.29887278582930754,
      "grad_norm": 1.0485658645629883,
      "learning_rate": 8.135881792367686e-05,
      "loss": 2.0509,
      "step": 116
    },
    {
      "epoch": 0.30917874396135264,
      "grad_norm": 1.1843186616897583,
      "learning_rate": 8.005405736415126e-05,
      "loss": 1.9353,
      "step": 120
    },
    {
      "epoch": 0.30917874396135264,
      "eval_loss": 1.9354380369186401,
      "eval_runtime": 117.1403,
      "eval_samples_per_second": 2.092,
      "eval_steps_per_second": 1.05,
      "step": 120
    },
    {
      "epoch": 0.31948470209339774,
      "grad_norm": 1.2164537906646729,
      "learning_rate": 7.871643313414718e-05,
      "loss": 2.0319,
      "step": 124
    },
    {
      "epoch": 0.32979066022544284,
      "grad_norm": 1.0794636011123657,
      "learning_rate": 7.734740790612136e-05,
      "loss": 1.9737,
      "step": 128
    },
    {
      "epoch": 0.32979066022544284,
      "eval_loss": 1.939180850982666,
      "eval_runtime": 132.7839,
      "eval_samples_per_second": 1.845,
      "eval_steps_per_second": 0.926,
      "step": 128
    },
    {
      "epoch": 0.34009661835748795,
      "grad_norm": 1.2831894159317017,
      "learning_rate": 7.594847868906076e-05,
      "loss": 1.9799,
      "step": 132
    },
    {
      "epoch": 0.350402576489533,
      "grad_norm": 1.066001057624817,
      "learning_rate": 7.452117519152542e-05,
      "loss": 1.9608,
      "step": 136
    },
    {
      "epoch": 0.350402576489533,
      "eval_loss": 1.933682918548584,
      "eval_runtime": 118.9246,
      "eval_samples_per_second": 2.06,
      "eval_steps_per_second": 1.034,
      "step": 136
    },
    {
      "epoch": 0.3607085346215781,
      "grad_norm": 1.1267732381820679,
      "learning_rate": 7.30670581489344e-05,
      "loss": 1.9253,
      "step": 140
    },
    {
      "epoch": 0.3710144927536232,
      "grad_norm": 1.232575535774231,
      "learning_rate": 7.158771761692464e-05,
      "loss": 1.928,
      "step": 144
    },
    {
      "epoch": 0.3710144927536232,
      "eval_loss": 1.9364768266677856,
      "eval_runtime": 232.9722,
      "eval_samples_per_second": 1.052,
      "eval_steps_per_second": 0.528,
      "step": 144
    },
    {
      "epoch": 0.3813204508856683,
      "grad_norm": 1.1609725952148438,
      "learning_rate": 7.008477123264848e-05,
      "loss": 1.9995,
      "step": 148
    },
    {
      "epoch": 0.39162640901771334,
      "grad_norm": 1.2346417903900146,
      "learning_rate": 6.855986244591104e-05,
      "loss": 2.0019,
      "step": 152
    },
    {
      "epoch": 0.39162640901771334,
      "eval_loss": 1.9326261281967163,
      "eval_runtime": 248.862,
      "eval_samples_per_second": 0.984,
      "eval_steps_per_second": 0.494,
      "step": 152
    },
    {
      "epoch": 0.40193236714975844,
      "grad_norm": 1.368483304977417,
      "learning_rate": 6.701465872208216e-05,
      "loss": 1.9931,
      "step": 156
    },
    {
      "epoch": 0.41223832528180354,
      "grad_norm": 1.1064777374267578,
      "learning_rate": 6.545084971874738e-05,
      "loss": 2.0525,
      "step": 160
    },
    {
      "epoch": 0.41223832528180354,
      "eval_loss": 1.940269112586975,
      "eval_runtime": 124.8033,
      "eval_samples_per_second": 1.963,
      "eval_steps_per_second": 0.986,
      "step": 160
    },
    {
      "epoch": 0.42254428341384864,
      "grad_norm": 1.1312705278396606,
      "learning_rate": 6.387014543809223e-05,
      "loss": 2.0432,
      "step": 164
    },
    {
      "epoch": 0.43285024154589374,
      "grad_norm": 1.2631207704544067,
      "learning_rate": 6.227427435703997e-05,
      "loss": 2.053,
      "step": 168
    },
    {
      "epoch": 0.43285024154589374,
      "eval_loss": 1.9401638507843018,
      "eval_runtime": 127.3455,
      "eval_samples_per_second": 1.924,
      "eval_steps_per_second": 0.966,
      "step": 168
    },
    {
      "epoch": 0.4431561996779388,
      "grad_norm": 1.2143282890319824,
      "learning_rate": 6.066498153718735e-05,
      "loss": 1.9412,
      "step": 172
    },
    {
      "epoch": 0.4534621578099839,
      "grad_norm": 1.3119618892669678,
      "learning_rate": 5.90440267166055e-05,
      "loss": 1.9342,
      "step": 176
    },
    {
      "epoch": 0.4534621578099839,
      "eval_loss": 1.9374455213546753,
      "eval_runtime": 128.0971,
      "eval_samples_per_second": 1.913,
      "eval_steps_per_second": 0.96,
      "step": 176
    },
    {
      "epoch": 0.463768115942029,
      "grad_norm": 1.404941439628601,
      "learning_rate": 5.74131823855921e-05,
      "loss": 2.0474,
      "step": 180
    },
    {
      "epoch": 0.4740740740740741,
      "grad_norm": 1.1769001483917236,
      "learning_rate": 5.577423184847932e-05,
      "loss": 1.9931,
      "step": 184
    },
    {
      "epoch": 0.4740740740740741,
      "eval_loss": 1.9400098323822021,
      "eval_runtime": 255.4556,
      "eval_samples_per_second": 0.959,
      "eval_steps_per_second": 0.481,
      "step": 184
    },
    {
      "epoch": 0.48438003220611914,
      "grad_norm": 1.0599452257156372,
      "learning_rate": 5.4128967273616625e-05,
      "loss": 1.9876,
      "step": 188
    },
    {
      "epoch": 0.49468599033816424,
      "grad_norm": 1.1136999130249023,
      "learning_rate": 5.247918773366112e-05,
      "loss": 2.0008,
      "step": 192
    },
    {
      "epoch": 0.49468599033816424,
      "eval_loss": 1.9412612915039062,
      "eval_runtime": 120.4032,
      "eval_samples_per_second": 2.035,
      "eval_steps_per_second": 1.022,
      "step": 192
    },
    {
      "epoch": 0.5049919484702093,
      "grad_norm": 1.4237738847732544,
      "learning_rate": 5.0826697238317935e-05,
      "loss": 1.9186,
      "step": 196
    },
    {
      "epoch": 0.5152979066022544,
      "grad_norm": 1.5648188591003418,
      "learning_rate": 4.917330276168208e-05,
      "loss": 1.9426,
      "step": 200
    },
    {
      "epoch": 0.5152979066022544,
      "eval_loss": 1.9405626058578491,
      "eval_runtime": 128.5673,
      "eval_samples_per_second": 1.906,
      "eval_steps_per_second": 0.957,
      "step": 200
    },
    {
      "epoch": 0.5256038647342995,
      "grad_norm": 1.2770507335662842,
      "learning_rate": 4.7520812266338885e-05,
      "loss": 2.062,
      "step": 204
    },
    {
      "epoch": 0.5359098228663446,
      "grad_norm": 1.2542366981506348,
      "learning_rate": 4.5871032726383386e-05,
      "loss": 1.9732,
      "step": 208
    },
    {
      "epoch": 0.5359098228663446,
      "eval_loss": 1.9409449100494385,
      "eval_runtime": 121.087,
      "eval_samples_per_second": 2.023,
      "eval_steps_per_second": 1.016,
      "step": 208
    },
    {
      "epoch": 0.5462157809983897,
      "grad_norm": 1.4167399406433105,
      "learning_rate": 4.4225768151520694e-05,
      "loss": 2.0026,
      "step": 212
    },
    {
      "epoch": 0.5565217391304348,
      "grad_norm": 1.3207523822784424,
      "learning_rate": 4.2586817614407895e-05,
      "loss": 2.0263,
      "step": 216
    },
    {
      "epoch": 0.5565217391304348,
      "eval_loss": 1.9430506229400635,
      "eval_runtime": 119.0525,
      "eval_samples_per_second": 2.058,
      "eval_steps_per_second": 1.033,
      "step": 216
    },
    {
      "epoch": 0.5668276972624798,
      "grad_norm": 1.2683689594268799,
      "learning_rate": 4.095597328339452e-05,
      "loss": 2.0507,
      "step": 220
    },
    {
      "epoch": 0.577133655394525,
      "grad_norm": 1.2979661226272583,
      "learning_rate": 3.933501846281267e-05,
      "loss": 1.9589,
      "step": 224
    },
    {
      "epoch": 0.577133655394525,
      "eval_loss": 1.944420337677002,
      "eval_runtime": 241.5642,
      "eval_samples_per_second": 1.014,
      "eval_steps_per_second": 0.509,
      "step": 224
    },
    {
      "epoch": 0.58743961352657,
      "grad_norm": 1.3029229640960693,
      "learning_rate": 3.772572564296005e-05,
      "loss": 2.0413,
      "step": 228
    },
    {
      "epoch": 0.5977455716586151,
      "grad_norm": 1.288469910621643,
      "learning_rate": 3.612985456190778e-05,
      "loss": 1.9824,
      "step": 232
    },
    {
      "epoch": 0.5977455716586151,
      "eval_loss": 1.9459607601165771,
      "eval_runtime": 135.6582,
      "eval_samples_per_second": 1.806,
      "eval_steps_per_second": 0.907,
      "step": 232
    },
    {
      "epoch": 0.6080515297906602,
      "grad_norm": 1.2693860530853271,
      "learning_rate": 3.4549150281252636e-05,
      "loss": 1.9893,
      "step": 236
    },
    {
      "epoch": 0.6183574879227053,
      "grad_norm": 1.4493002891540527,
      "learning_rate": 3.298534127791785e-05,
      "loss": 1.9252,
      "step": 240
    },
    {
      "epoch": 0.6183574879227053,
      "eval_loss": 1.9398853778839111,
      "eval_runtime": 123.6178,
      "eval_samples_per_second": 1.982,
      "eval_steps_per_second": 0.995,
      "step": 240
    },
    {
      "epoch": 0.6286634460547504,
      "grad_norm": 1.1343238353729248,
      "learning_rate": 3.144013755408895e-05,
      "loss": 1.9731,
      "step": 244
    },
    {
      "epoch": 0.6389694041867955,
      "grad_norm": 1.2282459735870361,
      "learning_rate": 2.991522876735154e-05,
      "loss": 1.9563,
      "step": 248
    },
    {
      "epoch": 0.6389694041867955,
      "eval_loss": 1.9399926662445068,
      "eval_runtime": 114.543,
      "eval_samples_per_second": 2.139,
      "eval_steps_per_second": 1.074,
      "step": 248
    },
    {
      "epoch": 0.6492753623188405,
      "grad_norm": 1.312157154083252,
      "learning_rate": 2.8412282383075363e-05,
      "loss": 1.9806,
      "step": 252
    },
    {
      "epoch": 0.6595813204508857,
      "grad_norm": 1.2436367273330688,
      "learning_rate": 2.693294185106562e-05,
      "loss": 2.0096,
      "step": 256
    },
    {
      "epoch": 0.6595813204508857,
      "eval_loss": 1.9414441585540771,
      "eval_runtime": 132.1293,
      "eval_samples_per_second": 1.854,
      "eval_steps_per_second": 0.931,
      "step": 256
    },
    {
      "epoch": 0.6698872785829307,
      "grad_norm": 1.241859793663025,
      "learning_rate": 2.547882480847461e-05,
      "loss": 1.9994,
      "step": 260
    },
    {
      "epoch": 0.6801932367149759,
      "grad_norm": 1.2852636575698853,
      "learning_rate": 2.405152131093926e-05,
      "loss": 1.9355,
      "step": 264
    },
    {
      "epoch": 0.6801932367149759,
      "eval_loss": 1.9419505596160889,
      "eval_runtime": 240.4599,
      "eval_samples_per_second": 1.019,
      "eval_steps_per_second": 0.512,
      "step": 264
    },
    {
      "epoch": 0.6904991948470209,
      "grad_norm": 1.2122435569763184,
      "learning_rate": 2.2652592093878666e-05,
      "loss": 2.0342,
      "step": 268
    },
    {
      "epoch": 0.700805152979066,
      "grad_norm": 1.203086495399475,
      "learning_rate": 2.128356686585282e-05,
      "loss": 2.003,
      "step": 272
    },
    {
      "epoch": 0.700805152979066,
      "eval_loss": 1.941483497619629,
      "eval_runtime": 127.3272,
      "eval_samples_per_second": 1.924,
      "eval_steps_per_second": 0.966,
      "step": 272
    },
    {
      "epoch": 0.7111111111111111,
      "grad_norm": 1.333521842956543,
      "learning_rate": 1.9945942635848748e-05,
      "loss": 1.9418,
      "step": 276
    },
    {
      "epoch": 0.7214170692431562,
      "grad_norm": 1.4001682996749878,
      "learning_rate": 1.8641182076323148e-05,
      "loss": 1.877,
      "step": 280
    },
    {
      "epoch": 0.7214170692431562,
      "eval_loss": 1.9395906925201416,
      "eval_runtime": 138.1205,
      "eval_samples_per_second": 1.774,
      "eval_steps_per_second": 0.891,
      "step": 280
    },
    {
      "epoch": 0.7317230273752013,
      "grad_norm": 1.3213990926742554,
      "learning_rate": 1.7370711923791567e-05,
      "loss": 2.0164,
      "step": 284
    },
    {
      "epoch": 0.7420289855072464,
      "grad_norm": 1.401374340057373,
      "learning_rate": 1.6135921418712956e-05,
      "loss": 2.0395,
      "step": 288
    },
    {
      "epoch": 0.7420289855072464,
      "eval_loss": 1.937800407409668,
      "eval_runtime": 131.2826,
      "eval_samples_per_second": 1.866,
      "eval_steps_per_second": 0.937,
      "step": 288
    },
    {
      "epoch": 0.7523349436392914,
      "grad_norm": 1.3368889093399048,
      "learning_rate": 1.4938160786375572e-05,
      "loss": 2.0054,
      "step": 292
    },
    {
      "epoch": 0.7626409017713366,
      "grad_norm": 1.3905627727508545,
      "learning_rate": 1.3778739760445552e-05,
      "loss": 1.9447,
      "step": 296
    },
    {
      "epoch": 0.7626409017713366,
      "eval_loss": 1.9381885528564453,
      "eval_runtime": 123.5018,
      "eval_samples_per_second": 1.984,
      "eval_steps_per_second": 0.996,
      "step": 296
    },
    {
      "epoch": 0.7729468599033816,
      "grad_norm": 1.365517020225525,
      "learning_rate": 1.2658926150792322e-05,
      "loss": 2.0237,
      "step": 300
    },
    {
      "epoch": 0.7832528180354267,
      "grad_norm": 1.2530505657196045,
      "learning_rate": 1.157994445715706e-05,
      "loss": 1.965,
      "step": 304
    },
    {
      "epoch": 0.7832528180354267,
      "eval_loss": 1.9391155242919922,
      "eval_runtime": 238.0568,
      "eval_samples_per_second": 1.029,
      "eval_steps_per_second": 0.517,
      "step": 304
    },
    {
      "epoch": 0.7935587761674718,
      "grad_norm": 1.2440646886825562,
      "learning_rate": 1.0542974530180327e-05,
      "loss": 1.9169,
      "step": 308
    },
    {
      "epoch": 0.8038647342995169,
      "grad_norm": 1.2581100463867188,
      "learning_rate": 9.549150281252633e-06,
      "loss": 1.9656,
      "step": 312
    },
    {
      "epoch": 0.8038647342995169,
      "eval_loss": 1.935294508934021,
      "eval_runtime": 113.5421,
      "eval_samples_per_second": 2.158,
      "eval_steps_per_second": 1.083,
      "step": 312
    },
    {
      "epoch": 0.814170692431562,
      "grad_norm": 1.2223581075668335,
      "learning_rate": 8.599558442598998e-06,
      "loss": 1.9945,
      "step": 316
    },
    {
      "epoch": 0.8244766505636071,
      "grad_norm": 1.2913447618484497,
      "learning_rate": 7.695237378953223e-06,
      "loss": 1.9928,
      "step": 320
    },
    {
      "epoch": 0.8244766505636071,
      "eval_loss": 1.9398421049118042,
      "eval_runtime": 127.4313,
      "eval_samples_per_second": 1.923,
      "eval_steps_per_second": 0.965,
      "step": 320
    },
    {
      "epoch": 0.8347826086956521,
      "grad_norm": 1.181248664855957,
      "learning_rate": 6.837175952121306e-06,
      "loss": 2.0377,
      "step": 324
    },
    {
      "epoch": 0.8450885668276973,
      "grad_norm": 1.2499854564666748,
      "learning_rate": 6.026312439675552e-06,
      "loss": 2.0004,
      "step": 328
    },
    {
      "epoch": 0.8450885668276973,
      "eval_loss": 1.9392446279525757,
      "eval_runtime": 122.338,
      "eval_samples_per_second": 2.003,
      "eval_steps_per_second": 1.005,
      "step": 328
    },
    {
      "epoch": 0.8553945249597423,
      "grad_norm": 1.2896592617034912,
      "learning_rate": 5.263533508961827e-06,
      "loss": 1.9349,
      "step": 332
    },
    {
      "epoch": 0.8657004830917875,
      "grad_norm": 1.1416115760803223,
      "learning_rate": 4.549673247541875e-06,
      "loss": 1.9883,
      "step": 336
    },
    {
      "epoch": 0.8657004830917875,
      "eval_loss": 1.9389346837997437,
      "eval_runtime": 229.9634,
      "eval_samples_per_second": 1.065,
      "eval_steps_per_second": 0.535,
      "step": 336
    },
    {
      "epoch": 0.8760064412238325,
      "grad_norm": 1.350136399269104,
      "learning_rate": 3.885512251130763e-06,
      "loss": 1.9733,
      "step": 340
    },
    {
      "epoch": 0.8863123993558776,
      "grad_norm": 1.3249493837356567,
      "learning_rate": 3.271776770026963e-06,
      "loss": 1.9764,
      "step": 344
    },
    {
      "epoch": 0.8863123993558776,
      "eval_loss": 1.9394721984863281,
      "eval_runtime": 123.8835,
      "eval_samples_per_second": 1.978,
      "eval_steps_per_second": 0.993,
      "step": 344
    },
    {
      "epoch": 0.8966183574879227,
      "grad_norm": 1.1916362047195435,
      "learning_rate": 2.7091379149682685e-06,
      "loss": 1.918,
      "step": 348
    },
    {
      "epoch": 0.9069243156199678,
      "grad_norm": 1.3328787088394165,
      "learning_rate": 2.1982109232821178e-06,
      "loss": 1.9474,
      "step": 352
    },
    {
      "epoch": 0.9069243156199678,
      "eval_loss": 1.9390062093734741,
      "eval_runtime": 132.5632,
      "eval_samples_per_second": 1.848,
      "eval_steps_per_second": 0.928,
      "step": 352
    },
    {
      "epoch": 0.9172302737520129,
      "grad_norm": 1.3123880624771118,
      "learning_rate": 1.7395544861325718e-06,
      "loss": 2.0644,
      "step": 356
    },
    {
      "epoch": 0.927536231884058,
      "grad_norm": 1.2262526750564575,
      "learning_rate": 1.333670137599713e-06,
      "loss": 2.0375,
      "step": 360
    },
    {
      "epoch": 0.927536231884058,
      "eval_loss": 1.9382147789001465,
      "eval_runtime": 125.7134,
      "eval_samples_per_second": 1.949,
      "eval_steps_per_second": 0.978,
      "step": 360
    },
    {
      "epoch": 0.937842190016103,
      "grad_norm": 1.253958821296692,
      "learning_rate": 9.810017062595322e-07,
      "loss": 2.0065,
      "step": 364
    },
    {
      "epoch": 0.9481481481481482,
      "grad_norm": 1.2366716861724854,
      "learning_rate": 6.819348298638839e-07,
      "loss": 1.9424,
      "step": 368
    },
    {
      "epoch": 0.9481481481481482,
      "eval_loss": 1.9385864734649658,
      "eval_runtime": 48.1426,
      "eval_samples_per_second": 5.089,
      "eval_steps_per_second": 2.555,
      "step": 368
    },
    {
      "epoch": 0.9584541062801932,
      "grad_norm": 1.3085380792617798,
      "learning_rate": 4.367965336512403e-07,
      "loss": 1.9222,
      "step": 372
    },
    {
      "epoch": 0.9687600644122383,
      "grad_norm": 1.3303335905075073,
      "learning_rate": 2.458548727494292e-07,
      "loss": 2.0088,
      "step": 376
    },
    {
      "epoch": 0.9687600644122383,
      "eval_loss": 1.9384690523147583,
      "eval_runtime": 83.7915,
      "eval_samples_per_second": 2.924,
      "eval_steps_per_second": 1.468,
      "step": 376
    },
    {
      "epoch": 0.9790660225442834,
      "grad_norm": 1.3368890285491943,
      "learning_rate": 1.0931863906127327e-07,
      "loss": 1.8973,
      "step": 380
    },
    {
      "epoch": 0.9893719806763285,
      "grad_norm": 1.4014090299606323,
      "learning_rate": 2.7337132953697554e-08,
      "loss": 1.9043,
      "step": 384
    },
    {
      "epoch": 0.9893719806763285,
      "eval_loss": 1.9385173320770264,
      "eval_runtime": 46.0839,
      "eval_samples_per_second": 5.316,
      "eval_steps_per_second": 2.669,
      "step": 384
    },
    {
      "epoch": 0.9996779388083736,
      "grad_norm": 1.3734824657440186,
      "learning_rate": 0.0,
      "loss": 2.0065,
      "step": 388
    }
  ],
  "logging_steps": 4,
  "max_steps": 388,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 8,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 5.427475676604334e+17,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}