{
  "best_metric": 0.7318872809410095,
  "best_model_checkpoint": "saves/CADICA_qwenvl_stenosis_classily_scale4/lora/sft/checkpoint-700",
  "epoch": 0.3605459696111254,
  "eval_steps": 50,
  "global_step": 700,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0025753283543651817,
      "grad_norm": 21.379672872549193,
      "learning_rate": 2.564102564102564e-06,
      "loss": 3.0388,
      "num_input_tokens_seen": 58496,
      "step": 5
    },
    {
      "epoch": 0.0051506567087303634,
      "grad_norm": 20.76117223991023,
      "learning_rate": 5.128205128205128e-06,
      "loss": 2.9831,
      "num_input_tokens_seen": 116960,
      "step": 10
    },
    {
      "epoch": 0.007725985063095545,
      "grad_norm": 22.5213517141881,
      "learning_rate": 7.692307692307694e-06,
      "loss": 2.8696,
      "num_input_tokens_seen": 175448,
      "step": 15
    },
    {
      "epoch": 0.010301313417460727,
      "grad_norm": 20.673071198727328,
      "learning_rate": 1.0256410256410256e-05,
      "loss": 2.6316,
      "num_input_tokens_seen": 233944,
      "step": 20
    },
    {
      "epoch": 0.012876641771825908,
      "grad_norm": 18.902291974538457,
      "learning_rate": 1.282051282051282e-05,
      "loss": 1.9707,
      "num_input_tokens_seen": 292416,
      "step": 25
    },
    {
      "epoch": 0.01545197012619109,
      "grad_norm": 8.05718270484028,
      "learning_rate": 1.5384615384615387e-05,
      "loss": 1.3782,
      "num_input_tokens_seen": 350904,
      "step": 30
    },
    {
      "epoch": 0.018027298480556272,
      "grad_norm": 3.6465275188422344,
      "learning_rate": 1.794871794871795e-05,
      "loss": 1.0628,
      "num_input_tokens_seen": 409384,
      "step": 35
    },
    {
      "epoch": 0.020602626834921454,
      "grad_norm": 4.842154180410959,
      "learning_rate": 2.0512820512820512e-05,
      "loss": 0.9789,
      "num_input_tokens_seen": 467864,
      "step": 40
    },
    {
      "epoch": 0.023177955189286635,
      "grad_norm": 2.6799567517341396,
      "learning_rate": 2.307692307692308e-05,
      "loss": 0.9327,
      "num_input_tokens_seen": 526384,
      "step": 45
    },
    {
      "epoch": 0.025753283543651816,
      "grad_norm": 2.629272923472648,
      "learning_rate": 2.564102564102564e-05,
      "loss": 0.9233,
      "num_input_tokens_seen": 584856,
      "step": 50
    },
    {
      "epoch": 0.025753283543651816,
      "eval_loss": 0.9281821846961975,
      "eval_runtime": 48.2484,
      "eval_samples_per_second": 1.244,
      "eval_steps_per_second": 0.311,
      "num_input_tokens_seen": 584856,
      "step": 50
    },
    {
      "epoch": 0.028328611898016998,
      "grad_norm": 1.2858813899048422,
      "learning_rate": 2.8205128205128207e-05,
      "loss": 0.897,
      "num_input_tokens_seen": 643344,
      "step": 55
    },
    {
      "epoch": 0.03090394025238218,
      "grad_norm": 1.177678811476692,
      "learning_rate": 3.0769230769230774e-05,
      "loss": 0.9169,
      "num_input_tokens_seen": 701808,
      "step": 60
    },
    {
      "epoch": 0.03347926860674736,
      "grad_norm": 1.2077065633120996,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 0.9019,
      "num_input_tokens_seen": 760304,
      "step": 65
    },
    {
      "epoch": 0.036054596961112545,
      "grad_norm": 1.1560644429967823,
      "learning_rate": 3.58974358974359e-05,
      "loss": 0.8996,
      "num_input_tokens_seen": 818760,
      "step": 70
    },
    {
      "epoch": 0.03862992531547772,
      "grad_norm": 0.732907212566054,
      "learning_rate": 3.846153846153846e-05,
      "loss": 0.9073,
      "num_input_tokens_seen": 877256,
      "step": 75
    },
    {
      "epoch": 0.04120525366984291,
      "grad_norm": 0.9616993870089134,
      "learning_rate": 4.1025641025641023e-05,
      "loss": 0.9081,
      "num_input_tokens_seen": 935752,
      "step": 80
    },
    {
      "epoch": 0.043780582024208085,
      "grad_norm": 0.8384067209941525,
      "learning_rate": 4.358974358974359e-05,
      "loss": 0.906,
      "num_input_tokens_seen": 994216,
      "step": 85
    },
    {
      "epoch": 0.04635591037857327,
      "grad_norm": 0.4045876972188175,
      "learning_rate": 4.615384615384616e-05,
      "loss": 0.8952,
      "num_input_tokens_seen": 1052704,
      "step": 90
    },
    {
      "epoch": 0.04893123873293845,
      "grad_norm": 0.6062678622593307,
      "learning_rate": 4.871794871794872e-05,
      "loss": 0.8996,
      "num_input_tokens_seen": 1111176,
      "step": 95
    },
    {
      "epoch": 0.05150656708730363,
      "grad_norm": 0.5316642041721752,
      "learning_rate": 5.128205128205128e-05,
      "loss": 0.9024,
      "num_input_tokens_seen": 1169664,
      "step": 100
    },
    {
      "epoch": 0.05150656708730363,
      "eval_loss": 0.911374032497406,
      "eval_runtime": 19.566,
      "eval_samples_per_second": 3.067,
      "eval_steps_per_second": 0.767,
      "num_input_tokens_seen": 1169664,
      "step": 100
    },
    {
      "epoch": 0.05408189544166881,
      "grad_norm": 0.47189419512253006,
      "learning_rate": 5.384615384615385e-05,
      "loss": 0.9142,
      "num_input_tokens_seen": 1228112,
      "step": 105
    },
    {
      "epoch": 0.056657223796033995,
      "grad_norm": 0.4885000351277984,
      "learning_rate": 5.6410256410256414e-05,
      "loss": 0.9054,
      "num_input_tokens_seen": 1286608,
      "step": 110
    },
    {
      "epoch": 0.05923255215039917,
      "grad_norm": 1.0232694160031948,
      "learning_rate": 5.897435897435898e-05,
      "loss": 0.8997,
      "num_input_tokens_seen": 1345072,
      "step": 115
    },
    {
      "epoch": 0.06180788050476436,
      "grad_norm": 0.6656697152989639,
      "learning_rate": 6.153846153846155e-05,
      "loss": 0.8988,
      "num_input_tokens_seen": 1403544,
      "step": 120
    },
    {
      "epoch": 0.06438320885912954,
      "grad_norm": 0.6273175951192728,
      "learning_rate": 6.410256410256412e-05,
      "loss": 0.9087,
      "num_input_tokens_seen": 1462024,
      "step": 125
    },
    {
      "epoch": 0.06695853721349472,
      "grad_norm": 0.707089894516894,
      "learning_rate": 6.666666666666667e-05,
      "loss": 0.8961,
      "num_input_tokens_seen": 1520528,
      "step": 130
    },
    {
      "epoch": 0.0695338655678599,
      "grad_norm": 0.4633668497982238,
      "learning_rate": 6.923076923076924e-05,
      "loss": 0.903,
      "num_input_tokens_seen": 1579024,
      "step": 135
    },
    {
      "epoch": 0.07210919392222509,
      "grad_norm": 0.5052802522069755,
      "learning_rate": 7.17948717948718e-05,
      "loss": 0.899,
      "num_input_tokens_seen": 1637504,
      "step": 140
    },
    {
      "epoch": 0.07468452227659027,
      "grad_norm": 0.7577940010204668,
      "learning_rate": 7.435897435897436e-05,
      "loss": 0.9071,
      "num_input_tokens_seen": 1696024,
      "step": 145
    },
    {
      "epoch": 0.07725985063095545,
      "grad_norm": 0.5812587904219971,
      "learning_rate": 7.692307692307693e-05,
      "loss": 0.9045,
      "num_input_tokens_seen": 1754512,
      "step": 150
    },
    {
      "epoch": 0.07725985063095545,
      "eval_loss": 0.8934853076934814,
      "eval_runtime": 19.8765,
      "eval_samples_per_second": 3.019,
      "eval_steps_per_second": 0.755,
      "num_input_tokens_seen": 1754512,
      "step": 150
    },
    {
      "epoch": 0.07983517898532062,
      "grad_norm": 0.5167982536583405,
      "learning_rate": 7.948717948717948e-05,
      "loss": 0.8992,
      "num_input_tokens_seen": 1812976,
      "step": 155
    },
    {
      "epoch": 0.08241050733968582,
      "grad_norm": 0.4971816797735092,
      "learning_rate": 8.205128205128205e-05,
      "loss": 0.8965,
      "num_input_tokens_seen": 1871464,
      "step": 160
    },
    {
      "epoch": 0.08498583569405099,
      "grad_norm": 0.6561749633642688,
      "learning_rate": 8.461538461538461e-05,
      "loss": 0.9094,
      "num_input_tokens_seen": 1929928,
      "step": 165
    },
    {
      "epoch": 0.08756116404841617,
      "grad_norm": 0.5010857314708574,
      "learning_rate": 8.717948717948718e-05,
      "loss": 0.903,
      "num_input_tokens_seen": 1988432,
      "step": 170
    },
    {
      "epoch": 0.09013649240278135,
      "grad_norm": 0.48794512034251364,
      "learning_rate": 8.974358974358975e-05,
      "loss": 0.902,
      "num_input_tokens_seen": 2046920,
      "step": 175
    },
    {
      "epoch": 0.09271182075714654,
      "grad_norm": 0.4040014684262414,
      "learning_rate": 9.230769230769232e-05,
      "loss": 0.9006,
      "num_input_tokens_seen": 2105392,
      "step": 180
    },
    {
      "epoch": 0.09528714911151172,
      "grad_norm": 0.5312840597942438,
      "learning_rate": 9.487179487179487e-05,
      "loss": 0.9042,
      "num_input_tokens_seen": 2163872,
      "step": 185
    },
    {
      "epoch": 0.0978624774658769,
      "grad_norm": 0.3535119366494406,
      "learning_rate": 9.743589743589744e-05,
      "loss": 0.9096,
      "num_input_tokens_seen": 2222352,
      "step": 190
    },
    {
      "epoch": 0.10043780582024209,
      "grad_norm": 0.30590378285024006,
      "learning_rate": 0.0001,
      "loss": 0.9037,
      "num_input_tokens_seen": 2280800,
      "step": 195
    },
    {
      "epoch": 0.10301313417460727,
      "grad_norm": 0.3055264226667786,
      "learning_rate": 9.999954623308172e-05,
      "loss": 0.904,
      "num_input_tokens_seen": 2339304,
      "step": 200
    },
    {
      "epoch": 0.10301313417460727,
      "eval_loss": 0.8980139493942261,
      "eval_runtime": 19.316,
      "eval_samples_per_second": 3.106,
      "eval_steps_per_second": 0.777,
      "num_input_tokens_seen": 2339304,
      "step": 200
    },
    {
      "epoch": 0.10558846252897244,
      "grad_norm": 0.8828178200664915,
      "learning_rate": 9.999818494056303e-05,
      "loss": 0.9029,
      "num_input_tokens_seen": 2397808,
      "step": 205
    },
    {
      "epoch": 0.10816379088333762,
      "grad_norm": 0.4308314655260644,
      "learning_rate": 9.99959161471523e-05,
      "loss": 0.9005,
      "num_input_tokens_seen": 2456288,
      "step": 210
    },
    {
      "epoch": 0.11073911923770281,
      "grad_norm": 0.4482188659643584,
      "learning_rate": 9.99927398940297e-05,
      "loss": 0.9096,
      "num_input_tokens_seen": 2514784,
      "step": 215
    },
    {
      "epoch": 0.11331444759206799,
      "grad_norm": 0.49014741417238206,
      "learning_rate": 9.998865623884635e-05,
      "loss": 0.9036,
      "num_input_tokens_seen": 2573240,
      "step": 220
    },
    {
      "epoch": 0.11588977594643317,
      "grad_norm": 0.2774850522391394,
      "learning_rate": 9.998366525572336e-05,
      "loss": 0.901,
      "num_input_tokens_seen": 2631672,
      "step": 225
    },
    {
      "epoch": 0.11846510430079835,
      "grad_norm": 0.49390873315018263,
      "learning_rate": 9.997776703525046e-05,
      "loss": 0.9018,
      "num_input_tokens_seen": 2690112,
      "step": 230
    },
    {
      "epoch": 0.12104043265516354,
      "grad_norm": 0.3284306399258997,
      "learning_rate": 9.997096168448432e-05,
      "loss": 0.8934,
      "num_input_tokens_seen": 2748608,
      "step": 235
    },
    {
      "epoch": 0.12361576100952872,
      "grad_norm": 0.7182680023403506,
      "learning_rate": 9.996324932694668e-05,
      "loss": 0.8876,
      "num_input_tokens_seen": 2807080,
      "step": 240
    },
    {
      "epoch": 0.1261910893638939,
      "grad_norm": 0.7305499346526235,
      "learning_rate": 9.995463010262206e-05,
      "loss": 0.9084,
      "num_input_tokens_seen": 2865520,
      "step": 245
    },
    {
      "epoch": 0.12876641771825909,
      "grad_norm": 0.5773211522908436,
      "learning_rate": 9.994510416795519e-05,
      "loss": 0.9106,
      "num_input_tokens_seen": 2924016,
      "step": 250
    },
    {
      "epoch": 0.12876641771825909,
      "eval_loss": 0.8958488702774048,
      "eval_runtime": 19.507,
      "eval_samples_per_second": 3.076,
      "eval_steps_per_second": 0.769,
      "num_input_tokens_seen": 2924016,
      "step": 250
    },
    {
      "epoch": 0.13134174607262425,
      "grad_norm": 0.44306061088962184,
      "learning_rate": 9.993467169584824e-05,
      "loss": 0.9012,
      "num_input_tokens_seen": 2982520,
      "step": 255
    },
    {
      "epoch": 0.13391707442698944,
      "grad_norm": 0.7851687259125024,
      "learning_rate": 9.992333287565765e-05,
      "loss": 0.9069,
      "num_input_tokens_seen": 3041008,
      "step": 260
    },
    {
      "epoch": 0.13649240278135463,
      "grad_norm": 0.5705235716557865,
      "learning_rate": 9.991108791319066e-05,
      "loss": 0.8918,
      "num_input_tokens_seen": 3099464,
      "step": 265
    },
    {
      "epoch": 0.1390677311357198,
      "grad_norm": 0.6202972137914602,
      "learning_rate": 9.989793703070163e-05,
      "loss": 0.8996,
      "num_input_tokens_seen": 3157944,
      "step": 270
    },
    {
      "epoch": 0.141643059490085,
      "grad_norm": 0.7583768377175583,
      "learning_rate": 9.988388046688799e-05,
      "loss": 0.9009,
      "num_input_tokens_seen": 3216448,
      "step": 275
    },
    {
      "epoch": 0.14421838784445018,
      "grad_norm": 0.7180540444266581,
      "learning_rate": 9.986891847688587e-05,
      "loss": 0.9059,
      "num_input_tokens_seen": 3274928,
      "step": 280
    },
    {
      "epoch": 0.14679371619881534,
      "grad_norm": 0.4173225854654158,
      "learning_rate": 9.985305133226553e-05,
      "loss": 0.8939,
      "num_input_tokens_seen": 3333408,
      "step": 285
    },
    {
      "epoch": 0.14936904455318054,
      "grad_norm": 0.7825855108807762,
      "learning_rate": 9.983627932102638e-05,
      "loss": 0.8899,
      "num_input_tokens_seen": 3391896,
      "step": 290
    },
    {
      "epoch": 0.1519443729075457,
      "grad_norm": 0.4850249272160501,
      "learning_rate": 9.981860274759173e-05,
      "loss": 0.9092,
      "num_input_tokens_seen": 3450392,
      "step": 295
    },
    {
      "epoch": 0.1545197012619109,
      "grad_norm": 0.3325682106309916,
      "learning_rate": 9.980002193280342e-05,
      "loss": 0.8901,
      "num_input_tokens_seen": 3508888,
      "step": 300
    },
    {
      "epoch": 0.1545197012619109,
      "eval_loss": 0.8932263255119324,
      "eval_runtime": 19.7633,
      "eval_samples_per_second": 3.036,
      "eval_steps_per_second": 0.759,
      "num_input_tokens_seen": 3508888,
      "step": 300
    },
    {
      "epoch": 0.15709502961627608,
      "grad_norm": 0.36562722920113416,
      "learning_rate": 9.978053721391578e-05,
      "loss": 0.9042,
      "num_input_tokens_seen": 3567368,
      "step": 305
    },
    {
      "epoch": 0.15967035797064125,
      "grad_norm": 0.3765491511973325,
      "learning_rate": 9.976014894458963e-05,
      "loss": 0.9007,
      "num_input_tokens_seen": 3625848,
      "step": 310
    },
    {
      "epoch": 0.16224568632500644,
      "grad_norm": 0.5264420727347517,
      "learning_rate": 9.973885749488589e-05,
      "loss": 0.9036,
      "num_input_tokens_seen": 3684336,
      "step": 315
    },
    {
      "epoch": 0.16482101467937163,
      "grad_norm": 0.24680747784235688,
      "learning_rate": 9.971666325125874e-05,
      "loss": 0.8936,
      "num_input_tokens_seen": 3742800,
      "step": 320
    },
    {
      "epoch": 0.1673963430337368,
      "grad_norm": 0.4982571051665039,
      "learning_rate": 9.969356661654876e-05,
      "loss": 0.8989,
      "num_input_tokens_seen": 3801280,
      "step": 325
    },
    {
      "epoch": 0.16997167138810199,
      "grad_norm": 0.49943012602572584,
      "learning_rate": 9.966956800997546e-05,
      "loss": 0.8983,
      "num_input_tokens_seen": 3859792,
      "step": 330
    },
    {
      "epoch": 0.17254699974246718,
      "grad_norm": 0.37381050353079964,
      "learning_rate": 9.964466786712984e-05,
      "loss": 0.9038,
      "num_input_tokens_seen": 3918272,
      "step": 335
    },
    {
      "epoch": 0.17512232809683234,
      "grad_norm": 0.7501484170811903,
      "learning_rate": 9.961886663996629e-05,
      "loss": 0.8947,
      "num_input_tokens_seen": 3976760,
      "step": 340
    },
    {
      "epoch": 0.17769765645119753,
      "grad_norm": 0.5623847203835772,
      "learning_rate": 9.959216479679458e-05,
      "loss": 0.9179,
      "num_input_tokens_seen": 4035240,
      "step": 345
    },
    {
      "epoch": 0.1802729848055627,
      "grad_norm": 0.34381878607605765,
      "learning_rate": 9.956456282227122e-05,
      "loss": 0.9059,
      "num_input_tokens_seen": 4093688,
      "step": 350
    },
    {
      "epoch": 0.1802729848055627,
      "eval_loss": 0.8960411548614502,
      "eval_runtime": 20.0734,
      "eval_samples_per_second": 2.989,
      "eval_steps_per_second": 0.747,
      "num_input_tokens_seen": 4093688,
      "step": 350
    },
    {
      "epoch": 0.1828483131599279,
      "grad_norm": 0.5786050277605497,
      "learning_rate": 9.953606121739074e-05,
      "loss": 0.8795,
      "num_input_tokens_seen": 4152160,
      "step": 355
    },
    {
      "epoch": 0.18542364151429308,
      "grad_norm": 1.3309062985901938,
      "learning_rate": 9.950666049947653e-05,
      "loss": 0.9143,
      "num_input_tokens_seen": 4210648,
      "step": 360
    },
    {
      "epoch": 0.18799896986865824,
      "grad_norm": 0.5010297124723248,
      "learning_rate": 9.947636120217155e-05,
      "loss": 0.9164,
      "num_input_tokens_seen": 4269136,
      "step": 365
    },
    {
      "epoch": 0.19057429822302344,
      "grad_norm": 0.7250841632803818,
      "learning_rate": 9.944516387542852e-05,
      "loss": 0.9061,
      "num_input_tokens_seen": 4327664,
      "step": 370
    },
    {
      "epoch": 0.19314962657738863,
      "grad_norm": 0.4506280653909736,
      "learning_rate": 9.941306908550005e-05,
      "loss": 0.8873,
      "num_input_tokens_seen": 4386120,
      "step": 375
    },
    {
      "epoch": 0.1957249549317538,
      "grad_norm": 0.6467175538087946,
      "learning_rate": 9.938007741492828e-05,
      "loss": 0.9038,
      "num_input_tokens_seen": 4444560,
      "step": 380
    },
    {
      "epoch": 0.19830028328611898,
      "grad_norm": 0.6742654767461002,
      "learning_rate": 9.934618946253437e-05,
      "loss": 0.9116,
      "num_input_tokens_seen": 4503016,
      "step": 385
    },
    {
      "epoch": 0.20087561164048418,
      "grad_norm": 0.418516338281364,
      "learning_rate": 9.931140584340761e-05,
      "loss": 0.9023,
      "num_input_tokens_seen": 4561496,
      "step": 390
    },
    {
      "epoch": 0.20345093999484934,
      "grad_norm": 0.5738919793445436,
      "learning_rate": 9.92757271888942e-05,
      "loss": 0.8901,
      "num_input_tokens_seen": 4619944,
      "step": 395
    },
    {
      "epoch": 0.20602626834921453,
      "grad_norm": 0.3600152815973316,
      "learning_rate": 9.923915414658587e-05,
      "loss": 0.9033,
      "num_input_tokens_seen": 4678384,
      "step": 400
    },
    {
      "epoch": 0.20602626834921453,
      "eval_loss": 0.906301736831665,
      "eval_runtime": 19.8079,
      "eval_samples_per_second": 3.029,
      "eval_steps_per_second": 0.757,
      "num_input_tokens_seen": 4678384,
      "step": 400
    },
    {
      "epoch": 0.2086015967035797,
      "grad_norm": 0.4854922483196253,
      "learning_rate": 9.920168738030807e-05,
      "loss": 0.8951,
      "num_input_tokens_seen": 4736904,
      "step": 405
    },
    {
      "epoch": 0.2111769250579449,
      "grad_norm": 0.4537032287993717,
      "learning_rate": 9.916332757010799e-05,
      "loss": 0.9131,
      "num_input_tokens_seen": 4795376,
      "step": 410
    },
    {
      "epoch": 0.21375225341231008,
      "grad_norm": 0.46655140937482126,
      "learning_rate": 9.912407541224213e-05,
      "loss": 0.8923,
      "num_input_tokens_seen": 4853880,
      "step": 415
    },
    {
      "epoch": 0.21632758176667524,
      "grad_norm": 0.3608301258843965,
      "learning_rate": 9.908393161916374e-05,
      "loss": 0.9026,
      "num_input_tokens_seen": 4912360,
      "step": 420
    },
    {
      "epoch": 0.21890291012104043,
      "grad_norm": 0.5055648531803498,
      "learning_rate": 9.904289691950979e-05,
      "loss": 0.905,
      "num_input_tokens_seen": 4970872,
      "step": 425
    },
    {
      "epoch": 0.22147823847540563,
      "grad_norm": 0.3171606869940592,
      "learning_rate": 9.900097205808789e-05,
      "loss": 0.8941,
      "num_input_tokens_seen": 5029304,
      "step": 430
    },
    {
      "epoch": 0.2240535668297708,
      "grad_norm": 0.5798722428230844,
      "learning_rate": 9.895815779586262e-05,
      "loss": 0.9031,
      "num_input_tokens_seen": 5087800,
      "step": 435
    },
    {
      "epoch": 0.22662889518413598,
      "grad_norm": 0.5751083474979835,
      "learning_rate": 9.891445490994182e-05,
      "loss": 0.8975,
      "num_input_tokens_seen": 5146312,
      "step": 440
    },
    {
      "epoch": 0.22920422353850115,
      "grad_norm": 1.2316619525123293,
      "learning_rate": 9.886986419356246e-05,
      "loss": 0.901,
      "num_input_tokens_seen": 5204800,
      "step": 445
    },
    {
      "epoch": 0.23177955189286634,
      "grad_norm": 0.5953254293558816,
      "learning_rate": 9.88243864560762e-05,
      "loss": 0.9062,
      "num_input_tokens_seen": 5263304,
      "step": 450
    },
    {
      "epoch": 0.23177955189286634,
      "eval_loss": 0.9007609486579895,
      "eval_runtime": 20.8827,
      "eval_samples_per_second": 2.873,
      "eval_steps_per_second": 0.718,
      "num_input_tokens_seen": 5263304,
      "step": 450
    },
    {
      "epoch": 0.23435488024723153,
      "grad_norm": 0.7549819464827967,
      "learning_rate": 9.877802252293474e-05,
      "loss": 0.8891,
      "num_input_tokens_seen": 5321760,
      "step": 455
    },
    {
      "epoch": 0.2369302086015967,
      "grad_norm": 0.9960909370043465,
      "learning_rate": 9.873077323567488e-05,
      "loss": 0.9026,
      "num_input_tokens_seen": 5380224,
      "step": 460
    },
    {
      "epoch": 0.23950553695596188,
      "grad_norm": 1.5888386521989892,
      "learning_rate": 9.868263945190312e-05,
      "loss": 0.8707,
      "num_input_tokens_seen": 5438704,
      "step": 465
    },
    {
      "epoch": 0.24208086531032708,
      "grad_norm": 3.0542478842411587,
      "learning_rate": 9.863362204528024e-05,
      "loss": 0.9051,
      "num_input_tokens_seen": 5497208,
      "step": 470
    },
    {
      "epoch": 0.24465619366469224,
      "grad_norm": 1.2908325061552137,
      "learning_rate": 9.858372190550533e-05,
      "loss": 0.8711,
      "num_input_tokens_seen": 5555704,
      "step": 475
    },
    {
      "epoch": 0.24723152201905743,
      "grad_norm": 3.1989324866235744,
      "learning_rate": 9.853293993829969e-05,
      "loss": 0.885,
      "num_input_tokens_seen": 5614160,
      "step": 480
    },
    {
      "epoch": 0.24980685037342262,
      "grad_norm": 3.591366302378185,
      "learning_rate": 9.848127706539039e-05,
      "loss": 0.8615,
      "num_input_tokens_seen": 5672640,
      "step": 485
    },
    {
      "epoch": 0.2523821787277878,
      "grad_norm": 2.053833335696007,
      "learning_rate": 9.842873422449354e-05,
      "loss": 0.9057,
      "num_input_tokens_seen": 5731072,
      "step": 490
    },
    {
      "epoch": 0.254957507082153,
      "grad_norm": 1.4501486574941083,
      "learning_rate": 9.837531236929726e-05,
      "loss": 0.8818,
      "num_input_tokens_seen": 5789544,
      "step": 495
    },
    {
      "epoch": 0.25753283543651817,
      "grad_norm": 2.1068404021122866,
      "learning_rate": 9.832101246944439e-05,
      "loss": 0.8576,
      "num_input_tokens_seen": 5848048,
      "step": 500
    },
    {
      "epoch": 0.25753283543651817,
      "eval_loss": 0.8268976211547852,
      "eval_runtime": 19.6346,
      "eval_samples_per_second": 3.056,
      "eval_steps_per_second": 0.764,
      "num_input_tokens_seen": 5848048,
      "step": 500
    },
    {
      "epoch": 0.26010816379088336,
      "grad_norm": 4.216936754020565,
      "learning_rate": 9.826583551051483e-05,
      "loss": 0.8566,
      "num_input_tokens_seen": 5906512,
      "step": 505
    },
    {
      "epoch": 0.2626834921452485,
      "grad_norm": 10.456282683777822,
      "learning_rate": 9.820978249400773e-05,
      "loss": 0.8365,
      "num_input_tokens_seen": 5965024,
      "step": 510
    },
    {
      "epoch": 0.2652588204996137,
      "grad_norm": 2.334974931865165,
      "learning_rate": 9.81528544373233e-05,
      "loss": 0.8882,
      "num_input_tokens_seen": 6023496,
      "step": 515
    },
    {
      "epoch": 0.2678341488539789,
      "grad_norm": 0.6948827424617825,
      "learning_rate": 9.809505237374426e-05,
      "loss": 0.8799,
      "num_input_tokens_seen": 6082000,
      "step": 520
    },
    {
      "epoch": 0.2704094772083441,
      "grad_norm": 0.8415524888602947,
      "learning_rate": 9.80363773524172e-05,
      "loss": 0.8758,
      "num_input_tokens_seen": 6140480,
      "step": 525
    },
    {
      "epoch": 0.27298480556270927,
      "grad_norm": 1.3536520282199265,
      "learning_rate": 9.797683043833345e-05,
      "loss": 0.8644,
      "num_input_tokens_seen": 6198968,
      "step": 530
    },
    {
      "epoch": 0.2755601339170744,
      "grad_norm": 4.556188528469967,
      "learning_rate": 9.791641271230982e-05,
      "loss": 0.8453,
      "num_input_tokens_seen": 6257464,
      "step": 535
    },
    {
      "epoch": 0.2781354622714396,
      "grad_norm": 2.890141630286954,
      "learning_rate": 9.78551252709689e-05,
      "loss": 0.8533,
      "num_input_tokens_seen": 6315944,
      "step": 540
    },
    {
      "epoch": 0.2807107906258048,
      "grad_norm": 4.471490037342243,
      "learning_rate": 9.779296922671923e-05,
      "loss": 0.8575,
      "num_input_tokens_seen": 6374408,
      "step": 545
    },
    {
      "epoch": 0.28328611898017,
      "grad_norm": 3.174906426420603,
      "learning_rate": 9.77299457077351e-05,
      "loss": 0.8666,
      "num_input_tokens_seen": 6432936,
      "step": 550
    },
    {
      "epoch": 0.28328611898017,
      "eval_loss": 0.7909801602363586,
      "eval_runtime": 19.8739,
      "eval_samples_per_second": 3.019,
      "eval_steps_per_second": 0.755,
      "num_input_tokens_seen": 6432936,
      "step": 550
    },
    {
      "epoch": 0.28586144733453517,
      "grad_norm": 3.529163852540611,
      "learning_rate": 9.7666055857936e-05,
      "loss": 0.8264,
      "num_input_tokens_seen": 6491400,
      "step": 555
    },
    {
      "epoch": 0.28843677568890036,
      "grad_norm": 4.044590312854015,
      "learning_rate": 9.760130083696595e-05,
      "loss": 0.8456,
      "num_input_tokens_seen": 6549872,
      "step": 560
    },
    {
      "epoch": 0.2910121040432655,
      "grad_norm": 4.650808013267891,
      "learning_rate": 9.75356818201724e-05,
      "loss": 0.8032,
      "num_input_tokens_seen": 6608296,
      "step": 565
    },
    {
      "epoch": 0.2935874323976307,
      "grad_norm": 6.577223054225459,
      "learning_rate": 9.746919999858492e-05,
      "loss": 0.8081,
      "num_input_tokens_seen": 6666768,
      "step": 570
    },
    {
      "epoch": 0.2961627607519959,
      "grad_norm": 7.1732684079932545,
      "learning_rate": 9.740185657889357e-05,
      "loss": 0.8398,
      "num_input_tokens_seen": 6725248,
      "step": 575
    },
    {
      "epoch": 0.29873808910636107,
      "grad_norm": 21.451661035438484,
      "learning_rate": 9.733365278342696e-05,
      "loss": 0.8908,
      "num_input_tokens_seen": 6783680,
      "step": 580
    },
    {
      "epoch": 0.30131341746072626,
      "grad_norm": 4.031699151478832,
      "learning_rate": 9.726458985013017e-05,
      "loss": 0.8248,
      "num_input_tokens_seen": 6842144,
      "step": 585
    },
    {
      "epoch": 0.3038887458150914,
      "grad_norm": 3.45579530759462,
      "learning_rate": 9.719466903254215e-05,
      "loss": 0.829,
      "num_input_tokens_seen": 6900656,
      "step": 590
    },
    {
      "epoch": 0.3064640741694566,
      "grad_norm": 4.518719062630672,
      "learning_rate": 9.712389159977307e-05,
      "loss": 0.8269,
      "num_input_tokens_seen": 6959128,
      "step": 595
    },
    {
      "epoch": 0.3090394025238218,
      "grad_norm": 12.728221405806083,
      "learning_rate": 9.705225883648121e-05,
      "loss": 0.7997,
      "num_input_tokens_seen": 7017576,
      "step": 600
    },
    {
      "epoch": 0.3090394025238218,
      "eval_loss": 0.787663459777832,
      "eval_runtime": 19.6121,
      "eval_samples_per_second": 3.059,
      "eval_steps_per_second": 0.765,
      "num_input_tokens_seen": 7017576,
      "step": 600
    },
    {
      "epoch": 0.311614730878187,
      "grad_norm": 6.69798043480266,
      "learning_rate": 9.697977204284973e-05,
      "loss": 0.8925,
      "num_input_tokens_seen": 7076032,
      "step": 605
    },
    {
      "epoch": 0.31419005923255217,
      "grad_norm": 5.067921055882507,
      "learning_rate": 9.690643253456297e-05,
      "loss": 0.8159,
      "num_input_tokens_seen": 7134536,
      "step": 610
    },
    {
      "epoch": 0.31676538758691736,
      "grad_norm": 7.400939684061883,
      "learning_rate": 9.683224164278264e-05,
      "loss": 0.826,
      "num_input_tokens_seen": 7193032,
      "step": 615
    },
    {
      "epoch": 0.3193407159412825,
      "grad_norm": 5.898525799199162,
      "learning_rate": 9.675720071412365e-05,
      "loss": 0.8187,
      "num_input_tokens_seen": 7251568,
      "step": 620
    },
    {
      "epoch": 0.3219160442956477,
      "grad_norm": 11.267105316774332,
      "learning_rate": 9.66813111106296e-05,
      "loss": 0.8524,
      "num_input_tokens_seen": 7310072,
      "step": 625
    },
    {
      "epoch": 0.3244913726500129,
      "grad_norm": 6.703970582399643,
      "learning_rate": 9.660457420974819e-05,
      "loss": 0.7966,
      "num_input_tokens_seen": 7368560,
      "step": 630
    },
    {
      "epoch": 0.32706670100437807,
      "grad_norm": 6.945445265294353,
      "learning_rate": 9.652699140430608e-05,
      "loss": 0.799,
      "num_input_tokens_seen": 7427040,
      "step": 635
    },
    {
      "epoch": 0.32964202935874326,
      "grad_norm": 7.0684293091171595,
      "learning_rate": 9.644856410248369e-05,
      "loss": 0.8477,
      "num_input_tokens_seen": 7485552,
      "step": 640
    },
    {
      "epoch": 0.3322173577131084,
      "grad_norm": 7.165086711244158,
      "learning_rate": 9.636929372778963e-05,
      "loss": 0.7867,
      "num_input_tokens_seen": 7544040,
      "step": 645
    },
    {
      "epoch": 0.3347926860674736,
      "grad_norm": 9.185933515393563,
      "learning_rate": 9.628918171903485e-05,
      "loss": 0.8367,
      "num_input_tokens_seen": 7602512,
      "step": 650
    },
    {
      "epoch": 0.3347926860674736,
      "eval_loss": 0.7940558791160583,
      "eval_runtime": 19.7641,
      "eval_samples_per_second": 3.036,
      "eval_steps_per_second": 0.759,
      "num_input_tokens_seen": 7602512,
      "step": 650
    },
    {
      "epoch": 0.3373680144218388,
      "grad_norm": 6.586425160827751,
      "learning_rate": 9.620822953030652e-05,
      "loss": 0.8131,
      "num_input_tokens_seen": 7660968,
      "step": 655
    },
    {
      "epoch": 0.33994334277620397,
      "grad_norm": 6.92970378602844,
      "learning_rate": 9.612643863094163e-05,
      "loss": 0.8348,
      "num_input_tokens_seen": 7719448,
      "step": 660
    },
    {
      "epoch": 0.34251867113056916,
      "grad_norm": 8.35228285894448,
      "learning_rate": 9.604381050550038e-05,
      "loss": 0.8289,
      "num_input_tokens_seen": 7777928,
      "step": 665
    },
    {
      "epoch": 0.34509399948493436,
      "grad_norm": 12.894782157020227,
      "learning_rate": 9.596034665373916e-05,
      "loss": 0.7758,
      "num_input_tokens_seen": 7836424,
      "step": 670
    },
    {
      "epoch": 0.3476693278392995,
      "grad_norm": 13.409694970235305,
      "learning_rate": 9.587604859058334e-05,
      "loss": 0.8189,
      "num_input_tokens_seen": 7894904,
      "step": 675
    },
    {
      "epoch": 0.3502446561936647,
      "grad_norm": 8.783205826578632,
      "learning_rate": 9.579091784609984e-05,
      "loss": 0.8221,
      "num_input_tokens_seen": 7953432,
      "step": 680
    },
    {
      "epoch": 0.3528199845480299,
      "grad_norm": 8.368380903445857,
      "learning_rate": 9.570495596546926e-05,
      "loss": 0.8378,
      "num_input_tokens_seen": 8011888,
      "step": 685
    },
    {
      "epoch": 0.35539531290239507,
      "grad_norm": 6.7086179135551145,
      "learning_rate": 9.561816450895793e-05,
      "loss": 0.7529,
      "num_input_tokens_seen": 8070344,
      "step": 690
    },
    {
      "epoch": 0.35797064125676026,
      "grad_norm": 8.476897088436242,
      "learning_rate": 9.55305450518895e-05,
      "loss": 0.7311,
      "num_input_tokens_seen": 8128816,
      "step": 695
    },
    {
      "epoch": 0.3605459696111254,
      "grad_norm": 10.427785019598666,
      "learning_rate": 9.544209918461642e-05,
      "loss": 0.774,
      "num_input_tokens_seen": 8187320,
      "step": 700
    },
    {
      "epoch": 0.3605459696111254,
      "eval_loss": 0.7318872809410095,
      "eval_runtime": 19.6917,
      "eval_samples_per_second": 3.047,
      "eval_steps_per_second": 0.762,
      "num_input_tokens_seen": 8187320,
      "step": 700
    }
  ],
  "logging_steps": 5,
  "max_steps": 3882,
  "num_input_tokens_seen": 8187320,
  "num_train_epochs": 2,
  "save_steps": 50,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 540175003287552.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}