| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.016036655211912942, | |
| "eval_steps": 34, | |
| "global_step": 238, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 6.738090425173506e-05, | |
| "eval_loss": 1.0482484102249146, | |
| "eval_runtime": 1771.3022, | |
| "eval_samples_per_second": 14.111, | |
| "eval_steps_per_second": 1.764, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00020214271275520516, | |
| "grad_norm": 2.1362860202789307, | |
| "learning_rate": 1.5e-05, | |
| "loss": 1.0166, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0004042854255104103, | |
| "grad_norm": 1.772956371307373, | |
| "learning_rate": 3e-05, | |
| "loss": 1.0004, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0006064281382656155, | |
| "grad_norm": 1.3759489059448242, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.9564, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0008085708510208206, | |
| "grad_norm": 1.2244105339050293, | |
| "learning_rate": 4.999675562428437e-05, | |
| "loss": 0.891, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.001010713563776026, | |
| "grad_norm": 1.1622825860977173, | |
| "learning_rate": 4.9979724954289244e-05, | |
| "loss": 0.8578, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.001212856276531231, | |
| "grad_norm": 1.089287281036377, | |
| "learning_rate": 4.994810682835951e-05, | |
| "loss": 0.832, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0014149989892864362, | |
| "grad_norm": 1.685937762260437, | |
| "learning_rate": 4.990191971059033e-05, | |
| "loss": 0.8445, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.0016171417020416413, | |
| "grad_norm": 0.9353536367416382, | |
| "learning_rate": 4.984119057295783e-05, | |
| "loss": 0.8481, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.0018192844147968466, | |
| "grad_norm": 0.9442921876907349, | |
| "learning_rate": 4.976595487956823e-05, | |
| "loss": 0.8389, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.002021427127552052, | |
| "grad_norm": 1.0226161479949951, | |
| "learning_rate": 4.967625656594782e-05, | |
| "loss": 0.8224, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.002223569840307257, | |
| "grad_norm": 0.8120137453079224, | |
| "learning_rate": 4.957214801338581e-05, | |
| "loss": 0.849, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.002290950744558992, | |
| "eval_loss": 0.8384992480278015, | |
| "eval_runtime": 1781.1941, | |
| "eval_samples_per_second": 14.033, | |
| "eval_steps_per_second": 1.754, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.002425712553062462, | |
| "grad_norm": 0.8147669434547424, | |
| "learning_rate": 4.9453690018345144e-05, | |
| "loss": 0.8287, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.0026278552658176675, | |
| "grad_norm": 0.9256265163421631, | |
| "learning_rate": 4.932095175695911e-05, | |
| "loss": 0.8112, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.0028299979785728724, | |
| "grad_norm": 0.9979642033576965, | |
| "learning_rate": 4.917401074463441e-05, | |
| "loss": 0.8657, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.0030321406913280777, | |
| "grad_norm": 0.7800766825675964, | |
| "learning_rate": 4.901295279078431e-05, | |
| "loss": 0.7922, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.0032342834040832826, | |
| "grad_norm": 0.7466667294502258, | |
| "learning_rate": 4.883787194871841e-05, | |
| "loss": 0.8455, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.003436426116838488, | |
| "grad_norm": 0.7748194932937622, | |
| "learning_rate": 4.864887046071813e-05, | |
| "loss": 0.8259, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.0036385688295936932, | |
| "grad_norm": 1.103950023651123, | |
| "learning_rate": 4.8446058698330115e-05, | |
| "loss": 0.8387, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.003840711542348898, | |
| "grad_norm": 0.7845460176467896, | |
| "learning_rate": 4.822955509791233e-05, | |
| "loss": 0.8067, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.004042854255104104, | |
| "grad_norm": 0.7884831428527832, | |
| "learning_rate": 4.799948609147061e-05, | |
| "loss": 0.8231, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.004244996967859308, | |
| "grad_norm": 0.9247403144836426, | |
| "learning_rate": 4.7755986032825864e-05, | |
| "loss": 0.8159, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.004447139680614514, | |
| "grad_norm": 0.7747366428375244, | |
| "learning_rate": 4.74991971191553e-05, | |
| "loss": 0.8132, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.004581901489117984, | |
| "eval_loss": 0.8286266922950745, | |
| "eval_runtime": 1781.5654, | |
| "eval_samples_per_second": 14.03, | |
| "eval_steps_per_second": 1.754, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.004649282393369719, | |
| "grad_norm": 0.7278714776039124, | |
| "learning_rate": 4.7229269307953235e-05, | |
| "loss": 0.741, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.004851425106124924, | |
| "grad_norm": 0.9090484380722046, | |
| "learning_rate": 4.694636022946012e-05, | |
| "loss": 0.8075, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.00505356781888013, | |
| "grad_norm": 0.8673194050788879, | |
| "learning_rate": 4.665063509461097e-05, | |
| "loss": 0.8395, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.005255710531635335, | |
| "grad_norm": 0.7307804822921753, | |
| "learning_rate": 4.6342266598556814e-05, | |
| "loss": 0.7995, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.005457853244390539, | |
| "grad_norm": 0.9184255003929138, | |
| "learning_rate": 4.6021434819815555e-05, | |
| "loss": 0.8318, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.005659995957145745, | |
| "grad_norm": 0.7924419045448303, | |
| "learning_rate": 4.568832711511125e-05, | |
| "loss": 0.8095, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.00586213866990095, | |
| "grad_norm": 0.7330142259597778, | |
| "learning_rate": 4.534313800996299e-05, | |
| "loss": 0.7674, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.006064281382656155, | |
| "grad_norm": 0.7591224908828735, | |
| "learning_rate": 4.498606908508754e-05, | |
| "loss": 0.8409, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.006266424095411361, | |
| "grad_norm": 0.7741730213165283, | |
| "learning_rate": 4.46173288586818e-05, | |
| "loss": 0.8541, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.006468566808166565, | |
| "grad_norm": 0.7202086448669434, | |
| "learning_rate": 4.4237132664654154e-05, | |
| "loss": 0.85, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.0066707095209217705, | |
| "grad_norm": 0.7738878726959229, | |
| "learning_rate": 4.384570252687542e-05, | |
| "loss": 0.8571, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.006872852233676976, | |
| "grad_norm": 0.7431773543357849, | |
| "learning_rate": 4.344326702952326e-05, | |
| "loss": 0.8264, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 0.006872852233676976, | |
| "eval_loss": 0.8239989280700684, | |
| "eval_runtime": 1781.4353, | |
| "eval_samples_per_second": 14.031, | |
| "eval_steps_per_second": 1.754, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 0.007074994946432181, | |
| "grad_norm": 0.7040189504623413, | |
| "learning_rate": 4.303006118359537e-05, | |
| "loss": 0.8247, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.0072771376591873865, | |
| "grad_norm": 0.7915964722633362, | |
| "learning_rate": 4.260632628966974e-05, | |
| "loss": 0.8551, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 0.007479280371942592, | |
| "grad_norm": 0.7852084040641785, | |
| "learning_rate": 4.217230979699188e-05, | |
| "loss": 0.8425, | |
| "step": 111 | |
| }, | |
| { | |
| "epoch": 0.007681423084697796, | |
| "grad_norm": 0.6728894114494324, | |
| "learning_rate": 4.172826515897146e-05, | |
| "loss": 0.8141, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 0.007883565797453002, | |
| "grad_norm": 0.7391681671142578, | |
| "learning_rate": 4.12744516851726e-05, | |
| "loss": 0.7987, | |
| "step": 117 | |
| }, | |
| { | |
| "epoch": 0.008085708510208208, | |
| "grad_norm": 0.7469043135643005, | |
| "learning_rate": 4.0811134389884433e-05, | |
| "loss": 0.7909, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.008287851222963412, | |
| "grad_norm": 0.7632879614830017, | |
| "learning_rate": 4.0338583837360225e-05, | |
| "loss": 0.8031, | |
| "step": 123 | |
| }, | |
| { | |
| "epoch": 0.008489993935718617, | |
| "grad_norm": 0.7656901478767395, | |
| "learning_rate": 3.985707598381544e-05, | |
| "loss": 0.843, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 0.008692136648473823, | |
| "grad_norm": 0.8024786114692688, | |
| "learning_rate": 3.9366892016277096e-05, | |
| "loss": 0.8403, | |
| "step": 129 | |
| }, | |
| { | |
| "epoch": 0.008894279361229027, | |
| "grad_norm": 0.6944208145141602, | |
| "learning_rate": 3.886831818837847e-05, | |
| "loss": 0.7908, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 0.009096422073984234, | |
| "grad_norm": 0.719901442527771, | |
| "learning_rate": 3.8361645653195026e-05, | |
| "loss": 0.8151, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.009163802978235968, | |
| "eval_loss": 0.8194563388824463, | |
| "eval_runtime": 1782.0949, | |
| "eval_samples_per_second": 14.026, | |
| "eval_steps_per_second": 1.754, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 0.009298564786739438, | |
| "grad_norm": 0.6918753981590271, | |
| "learning_rate": 3.784717029321922e-05, | |
| "loss": 0.8194, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 0.009500707499494642, | |
| "grad_norm": 0.7483247518539429, | |
| "learning_rate": 3.732519254757344e-05, | |
| "loss": 0.8422, | |
| "step": 141 | |
| }, | |
| { | |
| "epoch": 0.009702850212249849, | |
| "grad_norm": 0.7642280459403992, | |
| "learning_rate": 3.679601723656205e-05, | |
| "loss": 0.8222, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 0.009904992925005053, | |
| "grad_norm": 0.7145370244979858, | |
| "learning_rate": 3.625995338366492e-05, | |
| "loss": 0.8073, | |
| "step": 147 | |
| }, | |
| { | |
| "epoch": 0.01010713563776026, | |
| "grad_norm": 0.732183039188385, | |
| "learning_rate": 3.5717314035076355e-05, | |
| "loss": 0.8163, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.010309278350515464, | |
| "grad_norm": 0.6954637765884399, | |
| "learning_rate": 3.516841607689501e-05, | |
| "loss": 0.7573, | |
| "step": 153 | |
| }, | |
| { | |
| "epoch": 0.01051142106327067, | |
| "grad_norm": 0.7373840808868408, | |
| "learning_rate": 3.461358005007128e-05, | |
| "loss": 0.7868, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 0.010713563776025874, | |
| "grad_norm": 0.7047626376152039, | |
| "learning_rate": 3.405312996322042e-05, | |
| "loss": 0.821, | |
| "step": 159 | |
| }, | |
| { | |
| "epoch": 0.010915706488781079, | |
| "grad_norm": 0.7702988982200623, | |
| "learning_rate": 3.348739310341068e-05, | |
| "loss": 0.8194, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 0.011117849201536285, | |
| "grad_norm": 0.7867685556411743, | |
| "learning_rate": 3.2916699845036816e-05, | |
| "loss": 0.7898, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 0.01131999191429149, | |
| "grad_norm": 0.7021005153656006, | |
| "learning_rate": 3.234138345689077e-05, | |
| "loss": 0.7621, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 0.011454753722794959, | |
| "eval_loss": 0.8163909316062927, | |
| "eval_runtime": 1780.9274, | |
| "eval_samples_per_second": 14.035, | |
| "eval_steps_per_second": 1.755, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.011522134627046696, | |
| "grad_norm": 0.7096220850944519, | |
| "learning_rate": 3.17617799075421e-05, | |
| "loss": 0.7807, | |
| "step": 171 | |
| }, | |
| { | |
| "epoch": 0.0117242773398019, | |
| "grad_norm": 0.7657400369644165, | |
| "learning_rate": 3.1178227669141744e-05, | |
| "loss": 0.7858, | |
| "step": 174 | |
| }, | |
| { | |
| "epoch": 0.011926420052557105, | |
| "grad_norm": 0.8024412393569946, | |
| "learning_rate": 3.0591067519763895e-05, | |
| "loss": 0.8122, | |
| "step": 177 | |
| }, | |
| { | |
| "epoch": 0.01212856276531231, | |
| "grad_norm": 0.6976025700569153, | |
| "learning_rate": 3.0000642344401113e-05, | |
| "loss": 0.8288, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.012330705478067515, | |
| "grad_norm": 0.6966779828071594, | |
| "learning_rate": 2.9407296934729227e-05, | |
| "loss": 0.793, | |
| "step": 183 | |
| }, | |
| { | |
| "epoch": 0.012532848190822721, | |
| "grad_norm": 0.7219818830490112, | |
| "learning_rate": 2.8811377787758636e-05, | |
| "loss": 0.7883, | |
| "step": 186 | |
| }, | |
| { | |
| "epoch": 0.012734990903577926, | |
| "grad_norm": 0.8189945816993713, | |
| "learning_rate": 2.8213232903489865e-05, | |
| "loss": 0.885, | |
| "step": 189 | |
| }, | |
| { | |
| "epoch": 0.01293713361633313, | |
| "grad_norm": 0.902603805065155, | |
| "learning_rate": 2.761321158169134e-05, | |
| "loss": 0.8383, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 0.013139276329088337, | |
| "grad_norm": 0.8128630518913269, | |
| "learning_rate": 2.7011664217918154e-05, | |
| "loss": 0.852, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 0.013341419041843541, | |
| "grad_norm": 0.7031587958335876, | |
| "learning_rate": 2.6408942098890936e-05, | |
| "loss": 0.8622, | |
| "step": 198 | |
| }, | |
| { | |
| "epoch": 0.013543561754598747, | |
| "grad_norm": 0.7614731788635254, | |
| "learning_rate": 2.580539719735433e-05, | |
| "loss": 0.8162, | |
| "step": 201 | |
| }, | |
| { | |
| "epoch": 0.013745704467353952, | |
| "grad_norm": 0.6810929179191589, | |
| "learning_rate": 2.5201381966534748e-05, | |
| "loss": 0.8271, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 0.013745704467353952, | |
| "eval_loss": 0.8147265315055847, | |
| "eval_runtime": 1782.1355, | |
| "eval_samples_per_second": 14.025, | |
| "eval_steps_per_second": 1.754, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 0.013947847180109158, | |
| "grad_norm": 0.7248020768165588, | |
| "learning_rate": 2.459724913431772e-05, | |
| "loss": 0.814, | |
| "step": 207 | |
| }, | |
| { | |
| "epoch": 0.014149989892864362, | |
| "grad_norm": 0.7375376224517822, | |
| "learning_rate": 2.399335149726463e-05, | |
| "loss": 0.8381, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.014352132605619567, | |
| "grad_norm": 0.75850510597229, | |
| "learning_rate": 2.3390041714589514e-05, | |
| "loss": 0.7851, | |
| "step": 213 | |
| }, | |
| { | |
| "epoch": 0.014554275318374773, | |
| "grad_norm": 0.711068332195282, | |
| "learning_rate": 2.2787672102216042e-05, | |
| "loss": 0.7992, | |
| "step": 216 | |
| }, | |
| { | |
| "epoch": 0.014756418031129977, | |
| "grad_norm": 0.7301695346832275, | |
| "learning_rate": 2.2186594427034864e-05, | |
| "loss": 0.8506, | |
| "step": 219 | |
| }, | |
| { | |
| "epoch": 0.014958560743885184, | |
| "grad_norm": 0.720683753490448, | |
| "learning_rate": 2.1587159701481716e-05, | |
| "loss": 0.8061, | |
| "step": 222 | |
| }, | |
| { | |
| "epoch": 0.015160703456640388, | |
| "grad_norm": 0.665138304233551, | |
| "learning_rate": 2.098971797855599e-05, | |
| "loss": 0.7454, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.015362846169395592, | |
| "grad_norm": 0.6854159235954285, | |
| "learning_rate": 2.0394618147399713e-05, | |
| "loss": 0.828, | |
| "step": 228 | |
| }, | |
| { | |
| "epoch": 0.015564988882150799, | |
| "grad_norm": 0.7191194891929626, | |
| "learning_rate": 1.980220772955602e-05, | |
| "loss": 0.7885, | |
| "step": 231 | |
| }, | |
| { | |
| "epoch": 0.015767131594906003, | |
| "grad_norm": 0.7301977276802063, | |
| "learning_rate": 1.921283267602643e-05, | |
| "loss": 0.81, | |
| "step": 234 | |
| }, | |
| { | |
| "epoch": 0.015969274307661208, | |
| "grad_norm": 0.7239346504211426, | |
| "learning_rate": 1.8626837165245165e-05, | |
| "loss": 0.787, | |
| "step": 237 | |
| }, | |
| { | |
| "epoch": 0.016036655211912942, | |
| "eval_loss": 0.8127490878105164, | |
| "eval_runtime": 1781.9808, | |
| "eval_samples_per_second": 14.027, | |
| "eval_steps_per_second": 1.754, | |
| "step": 238 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 400, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 34, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.346898330352353e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |