| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.8634646519158122, | |
| "eval_steps": 100, | |
| "global_step": 400, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0021586616297895305, | |
| "eval_loss": 1.4607058763504028, | |
| "eval_runtime": 15.8449, | |
| "eval_samples_per_second": 12.307, | |
| "eval_steps_per_second": 6.185, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.010793308148947653, | |
| "grad_norm": 10.714120864868164, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 2.7678, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.021586616297895305, | |
| "grad_norm": 16.979589462280273, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 3.0718, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.032379924446842956, | |
| "grad_norm": 13.034200668334961, | |
| "learning_rate": 5e-05, | |
| "loss": 2.3931, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.04317323259579061, | |
| "grad_norm": 11.763636589050293, | |
| "learning_rate": 6.666666666666667e-05, | |
| "loss": 0.7464, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.053966540744738264, | |
| "grad_norm": 13.58657169342041, | |
| "learning_rate": 8.333333333333334e-05, | |
| "loss": 0.1455, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.06475984889368591, | |
| "grad_norm": 4.897015571594238, | |
| "learning_rate": 0.0001, | |
| "loss": 0.3329, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.07555315704263357, | |
| "grad_norm": 11.031070709228516, | |
| "learning_rate": 9.995494831023409e-05, | |
| "loss": 0.1538, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.08634646519158122, | |
| "grad_norm": 3.7399301528930664, | |
| "learning_rate": 9.981987442712633e-05, | |
| "loss": 0.1843, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.09713977334052887, | |
| "grad_norm": 13.98750114440918, | |
| "learning_rate": 9.959502176294383e-05, | |
| "loss": 0.1598, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.10793308148947653, | |
| "grad_norm": 4.850798606872559, | |
| "learning_rate": 9.928079551738543e-05, | |
| "loss": 0.1192, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.11872638963842418, | |
| "grad_norm": 4.3721442222595215, | |
| "learning_rate": 9.887776194738432e-05, | |
| "loss": 0.0531, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.12951969778737182, | |
| "grad_norm": 2.2560079097747803, | |
| "learning_rate": 9.838664734667495e-05, | |
| "loss": 0.0253, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.14031300593631948, | |
| "grad_norm": 0.05000563710927963, | |
| "learning_rate": 9.780833673696254e-05, | |
| "loss": 0.004, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.15110631408526715, | |
| "grad_norm": 24.105127334594727, | |
| "learning_rate": 9.714387227305422e-05, | |
| "loss": 0.029, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.16189962223421478, | |
| "grad_norm": 0.44868168234825134, | |
| "learning_rate": 9.639445136482548e-05, | |
| "loss": 0.0174, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.17269293038316244, | |
| "grad_norm": 0.05608593299984932, | |
| "learning_rate": 9.55614245194068e-05, | |
| "loss": 0.0005, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.1834862385321101, | |
| "grad_norm": 2.6503210067749023, | |
| "learning_rate": 9.464629290747842e-05, | |
| "loss": 0.0557, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.19427954668105774, | |
| "grad_norm": 24.43099021911621, | |
| "learning_rate": 9.365070565805941e-05, | |
| "loss": 0.0581, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.2050728548300054, | |
| "grad_norm": 0.019484924152493477, | |
| "learning_rate": 9.257645688666556e-05, | |
| "loss": 0.1835, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.21586616297895306, | |
| "grad_norm": 0.16496644914150238, | |
| "learning_rate": 9.142548246219212e-05, | |
| "loss": 0.0975, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.21586616297895306, | |
| "eval_loss": 0.007833840325474739, | |
| "eval_runtime": 16.1967, | |
| "eval_samples_per_second": 12.039, | |
| "eval_steps_per_second": 6.051, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.2266594711279007, | |
| "grad_norm": 0.5707283616065979, | |
| "learning_rate": 9.019985651834703e-05, | |
| "loss": 0.0485, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.23745277927684835, | |
| "grad_norm": 1.5518710613250732, | |
| "learning_rate": 8.890178771592199e-05, | |
| "loss": 0.0304, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.248246087425796, | |
| "grad_norm": 0.15097752213478088, | |
| "learning_rate": 8.753361526263621e-05, | |
| "loss": 0.0489, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.25903939557474365, | |
| "grad_norm": 0.09341863542795181, | |
| "learning_rate": 8.609780469772623e-05, | |
| "loss": 0.0663, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.26983270372369134, | |
| "grad_norm": 4.45965576171875, | |
| "learning_rate": 8.459694344887732e-05, | |
| "loss": 0.0634, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.28062601187263897, | |
| "grad_norm": 0.03669450059533119, | |
| "learning_rate": 8.303373616950408e-05, | |
| "loss": 0.0033, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.2914193200215866, | |
| "grad_norm": 0.00979743991047144, | |
| "learning_rate": 8.141099986478212e-05, | |
| "loss": 0.0031, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.3022126281705343, | |
| "grad_norm": 4.463639259338379, | |
| "learning_rate": 7.973165881521434e-05, | |
| "loss": 0.0866, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.3130059363194819, | |
| "grad_norm": 22.530567169189453, | |
| "learning_rate": 7.799873930687978e-05, | |
| "loss": 0.0558, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 0.32379924446842956, | |
| "grad_norm": 0.10999006032943726, | |
| "learning_rate": 7.621536417786159e-05, | |
| "loss": 0.0566, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.33459255261737725, | |
| "grad_norm": 1.8067408800125122, | |
| "learning_rate": 7.438474719068173e-05, | |
| "loss": 0.0178, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 0.3453858607663249, | |
| "grad_norm": 0.04744194447994232, | |
| "learning_rate": 7.251018724088367e-05, | |
| "loss": 0.0054, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.3561791689152725, | |
| "grad_norm": 1.6452178955078125, | |
| "learning_rate": 7.059506241219965e-05, | |
| "loss": 0.0072, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 0.3669724770642202, | |
| "grad_norm": 5.83623743057251, | |
| "learning_rate": 6.864282388901544e-05, | |
| "loss": 0.0357, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.37776578521316784, | |
| "grad_norm": 0.008103343658149242, | |
| "learning_rate": 6.665698973710288e-05, | |
| "loss": 0.0034, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.38855909336211547, | |
| "grad_norm": 0.008926725946366787, | |
| "learning_rate": 6.464113856382752e-05, | |
| "loss": 0.0003, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.39935240151106316, | |
| "grad_norm": 0.4430943727493286, | |
| "learning_rate": 6.259890306925627e-05, | |
| "loss": 0.0012, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 0.4101457096600108, | |
| "grad_norm": 1.2495324611663818, | |
| "learning_rate": 6.0533963499786314e-05, | |
| "loss": 0.0074, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.4209390178089584, | |
| "grad_norm": 0.028867028653621674, | |
| "learning_rate": 5.8450041016092464e-05, | |
| "loss": 0.0006, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 0.4317323259579061, | |
| "grad_norm": 0.28853410482406616, | |
| "learning_rate": 5.6350890987343944e-05, | |
| "loss": 0.0016, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.4317323259579061, | |
| "eval_loss": 0.002732636407017708, | |
| "eval_runtime": 16.2748, | |
| "eval_samples_per_second": 11.982, | |
| "eval_steps_per_second": 6.022, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.44252563410685375, | |
| "grad_norm": 0.9760282039642334, | |
| "learning_rate": 5.4240296223775465e-05, | |
| "loss": 0.0018, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 0.4533189422558014, | |
| "grad_norm": 0.012445294298231602, | |
| "learning_rate": 5.212206015980742e-05, | |
| "loss": 0.0012, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.46411225040474907, | |
| "grad_norm": 0.027542708441615105, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0004, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 0.4749055585536967, | |
| "grad_norm": 0.011909710243344307, | |
| "learning_rate": 4.78779398401926e-05, | |
| "loss": 0.0246, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.48569886670264434, | |
| "grad_norm": 3.801331043243408, | |
| "learning_rate": 4.575970377622456e-05, | |
| "loss": 0.0292, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.496492174851592, | |
| "grad_norm": 0.030169179663062096, | |
| "learning_rate": 4.364910901265606e-05, | |
| "loss": 0.036, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.5072854830005397, | |
| "grad_norm": 0.024300722405314445, | |
| "learning_rate": 4.1549958983907555e-05, | |
| "loss": 0.0002, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 0.5180787911494873, | |
| "grad_norm": 0.0078019872307777405, | |
| "learning_rate": 3.94660365002137e-05, | |
| "loss": 0.0023, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.5288720992984349, | |
| "grad_norm": 1.6584054231643677, | |
| "learning_rate": 3.740109693074375e-05, | |
| "loss": 0.0025, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 0.5396654074473827, | |
| "grad_norm": 16.714975357055664, | |
| "learning_rate": 3.5358861436172485e-05, | |
| "loss": 0.0681, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.5504587155963303, | |
| "grad_norm": 3.540844440460205, | |
| "learning_rate": 3.334301026289712e-05, | |
| "loss": 0.0674, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 0.5612520237452779, | |
| "grad_norm": 0.04804408550262451, | |
| "learning_rate": 3.135717611098458e-05, | |
| "loss": 0.0372, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.5720453318942256, | |
| "grad_norm": 0.014452624134719372, | |
| "learning_rate": 2.9404937587800375e-05, | |
| "loss": 0.0004, | |
| "step": 265 | |
| }, | |
| { | |
| "epoch": 0.5828386400431732, | |
| "grad_norm": 0.03764789551496506, | |
| "learning_rate": 2.748981275911633e-05, | |
| "loss": 0.0005, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.5936319481921208, | |
| "grad_norm": 0.08727376908063889, | |
| "learning_rate": 2.5615252809318284e-05, | |
| "loss": 0.0005, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.6044252563410686, | |
| "grad_norm": 0.025368178263306618, | |
| "learning_rate": 2.3784635822138424e-05, | |
| "loss": 0.0005, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.6152185644900162, | |
| "grad_norm": 0.033766236156225204, | |
| "learning_rate": 2.2001260693120233e-05, | |
| "loss": 0.0011, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 0.6260118726389639, | |
| "grad_norm": 1.3521661758422852, | |
| "learning_rate": 2.026834118478567e-05, | |
| "loss": 0.0021, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.6368051807879115, | |
| "grad_norm": 0.02496200241148472, | |
| "learning_rate": 1.858900013521788e-05, | |
| "loss": 0.0006, | |
| "step": 295 | |
| }, | |
| { | |
| "epoch": 0.6475984889368591, | |
| "grad_norm": 0.019956547766923904, | |
| "learning_rate": 1.6966263830495936e-05, | |
| "loss": 0.0009, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.6475984889368591, | |
| "eval_loss": 0.0026158045511692762, | |
| "eval_runtime": 16.2813, | |
| "eval_samples_per_second": 11.977, | |
| "eval_steps_per_second": 6.019, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.6583917970858068, | |
| "grad_norm": 0.23486492037773132, | |
| "learning_rate": 1.5403056551122697e-05, | |
| "loss": 0.0017, | |
| "step": 305 | |
| }, | |
| { | |
| "epoch": 0.6691851052347545, | |
| "grad_norm": 8.54863166809082, | |
| "learning_rate": 1.3902195302273779e-05, | |
| "loss": 0.0329, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.6799784133837021, | |
| "grad_norm": 0.19243143498897552, | |
| "learning_rate": 1.246638473736378e-05, | |
| "loss": 0.0098, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 0.6907717215326498, | |
| "grad_norm": 0.023138998076319695, | |
| "learning_rate": 1.1098212284078036e-05, | |
| "loss": 0.0364, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.7015650296815974, | |
| "grad_norm": 0.01772366650402546, | |
| "learning_rate": 9.800143481652979e-06, | |
| "loss": 0.0048, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 0.712358337830545, | |
| "grad_norm": 0.012253678403794765, | |
| "learning_rate": 8.574517537807897e-06, | |
| "loss": 0.0003, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.7231516459794927, | |
| "grad_norm": 0.4814263880252838, | |
| "learning_rate": 7.423543113334436e-06, | |
| "loss": 0.0143, | |
| "step": 335 | |
| }, | |
| { | |
| "epoch": 0.7339449541284404, | |
| "grad_norm": 0.007275492884218693, | |
| "learning_rate": 6.349294341940593e-06, | |
| "loss": 0.0016, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.744738262277388, | |
| "grad_norm": 0.6222500801086426, | |
| "learning_rate": 5.353707092521582e-06, | |
| "loss": 0.0009, | |
| "step": 345 | |
| }, | |
| { | |
| "epoch": 0.7555315704263357, | |
| "grad_norm": 95.98480224609375, | |
| "learning_rate": 4.43857548059321e-06, | |
| "loss": 0.0666, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.7663248785752833, | |
| "grad_norm": 1.1413521766662598, | |
| "learning_rate": 3.605548635174533e-06, | |
| "loss": 0.0031, | |
| "step": 355 | |
| }, | |
| { | |
| "epoch": 0.7771181867242309, | |
| "grad_norm": 7.497922420501709, | |
| "learning_rate": 2.85612772694579e-06, | |
| "loss": 0.0081, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.7879114948731786, | |
| "grad_norm": 0.18039092421531677, | |
| "learning_rate": 2.191663263037458e-06, | |
| "loss": 0.0416, | |
| "step": 365 | |
| }, | |
| { | |
| "epoch": 0.7987048030221263, | |
| "grad_norm": 0.9556758999824524, | |
| "learning_rate": 1.6133526533250565e-06, | |
| "loss": 0.0296, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.809498111171074, | |
| "grad_norm": 0.2905849516391754, | |
| "learning_rate": 1.1222380526156928e-06, | |
| "loss": 0.0012, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 0.8202914193200216, | |
| "grad_norm": 0.059283964335918427, | |
| "learning_rate": 7.192044826145771e-07, | |
| "loss": 0.001, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.8310847274689692, | |
| "grad_norm": 0.029324790462851524, | |
| "learning_rate": 4.049782370561583e-07, | |
| "loss": 0.0003, | |
| "step": 385 | |
| }, | |
| { | |
| "epoch": 0.8418780356179169, | |
| "grad_norm": 0.014780248515307903, | |
| "learning_rate": 1.8012557287367392e-07, | |
| "loss": 0.0002, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.8526713437668646, | |
| "grad_norm": 0.052729569375514984, | |
| "learning_rate": 4.5051689765929214e-08, | |
| "loss": 0.0009, | |
| "step": 395 | |
| }, | |
| { | |
| "epoch": 0.8634646519158122, | |
| "grad_norm": 0.5674512386322021, | |
| "learning_rate": 0.0, | |
| "loss": 0.0009, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.8634646519158122, | |
| "eval_loss": 0.0023802632931619883, | |
| "eval_runtime": 16.2847, | |
| "eval_samples_per_second": 11.974, | |
| "eval_steps_per_second": 6.018, | |
| "step": 400 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 400, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.402138244677632e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |