| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.05800398066533978, | |
| "eval_steps": 34, | |
| "global_step": 204, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0002843332385555871, | |
| "eval_loss": 1.5018354654312134, | |
| "eval_runtime": 166.295, | |
| "eval_samples_per_second": 35.623, | |
| "eval_steps_per_second": 4.456, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0008529997156667614, | |
| "grad_norm": 0.7210565209388733, | |
| "learning_rate": 1.5e-05, | |
| "loss": 1.5223, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0017059994313335229, | |
| "grad_norm": 0.6832892298698425, | |
| "learning_rate": 3e-05, | |
| "loss": 1.5221, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0025589991470002845, | |
| "grad_norm": 0.7228020429611206, | |
| "learning_rate": 4.5e-05, | |
| "loss": 1.408, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0034119988626670457, | |
| "grad_norm": 0.6384875774383545, | |
| "learning_rate": 4.999675562428437e-05, | |
| "loss": 1.4404, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.004264998578333807, | |
| "grad_norm": 0.3351251482963562, | |
| "learning_rate": 4.9979724954289244e-05, | |
| "loss": 1.3736, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.005117998294000569, | |
| "grad_norm": 0.182565838098526, | |
| "learning_rate": 4.994810682835951e-05, | |
| "loss": 1.3079, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.00597099800966733, | |
| "grad_norm": 0.19113591313362122, | |
| "learning_rate": 4.990191971059033e-05, | |
| "loss": 1.316, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.0068239977253340914, | |
| "grad_norm": 0.1948336660861969, | |
| "learning_rate": 4.984119057295783e-05, | |
| "loss": 1.2426, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.007676997441000853, | |
| "grad_norm": 0.18684643507003784, | |
| "learning_rate": 4.976595487956823e-05, | |
| "loss": 1.2503, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.008529997156667614, | |
| "grad_norm": 0.18414735794067383, | |
| "learning_rate": 4.967625656594782e-05, | |
| "loss": 1.2323, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.009382996872334376, | |
| "grad_norm": 0.17294025421142578, | |
| "learning_rate": 4.957214801338581e-05, | |
| "loss": 1.1942, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.009667330110889964, | |
| "eval_loss": 1.1813108921051025, | |
| "eval_runtime": 167.7438, | |
| "eval_samples_per_second": 35.316, | |
| "eval_steps_per_second": 4.417, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.010235996588001138, | |
| "grad_norm": 0.16702412068843842, | |
| "learning_rate": 4.9453690018345144e-05, | |
| "loss": 1.1981, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.011088996303667898, | |
| "grad_norm": 0.1968175172805786, | |
| "learning_rate": 4.932095175695911e-05, | |
| "loss": 1.1675, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.01194199601933466, | |
| "grad_norm": 0.18244469165802002, | |
| "learning_rate": 4.917401074463441e-05, | |
| "loss": 1.1584, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.01279499573500142, | |
| "grad_norm": 0.16749081015586853, | |
| "learning_rate": 4.901295279078431e-05, | |
| "loss": 1.1134, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.013647995450668183, | |
| "grad_norm": 0.17398597300052643, | |
| "learning_rate": 4.883787194871841e-05, | |
| "loss": 1.1139, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.014500995166334945, | |
| "grad_norm": 0.17164087295532227, | |
| "learning_rate": 4.864887046071813e-05, | |
| "loss": 1.079, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.015353994882001705, | |
| "grad_norm": 0.1644001007080078, | |
| "learning_rate": 4.8446058698330115e-05, | |
| "loss": 1.0646, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.01620699459766847, | |
| "grad_norm": 0.16490623354911804, | |
| "learning_rate": 4.822955509791233e-05, | |
| "loss": 1.0739, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.017059994313335228, | |
| "grad_norm": 0.17708458006381989, | |
| "learning_rate": 4.799948609147061e-05, | |
| "loss": 1.0897, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.01791299402900199, | |
| "grad_norm": 0.15597032010555267, | |
| "learning_rate": 4.7755986032825864e-05, | |
| "loss": 1.0566, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.018765993744668752, | |
| "grad_norm": 0.17728550732135773, | |
| "learning_rate": 4.74991971191553e-05, | |
| "loss": 1.0275, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.019334660221779928, | |
| "eval_loss": 1.0011852979660034, | |
| "eval_runtime": 168.1174, | |
| "eval_samples_per_second": 35.237, | |
| "eval_steps_per_second": 4.408, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.019618993460335514, | |
| "grad_norm": 0.16994404792785645, | |
| "learning_rate": 4.7229269307953235e-05, | |
| "loss": 0.9841, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.020471993176002276, | |
| "grad_norm": 0.17181497812271118, | |
| "learning_rate": 4.694636022946012e-05, | |
| "loss": 0.9944, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.021324992891669035, | |
| "grad_norm": 0.21411621570587158, | |
| "learning_rate": 4.665063509461097e-05, | |
| "loss": 1.0013, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.022177992607335797, | |
| "grad_norm": 0.19594340026378632, | |
| "learning_rate": 4.6342266598556814e-05, | |
| "loss": 0.9969, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.02303099232300256, | |
| "grad_norm": 0.2057276964187622, | |
| "learning_rate": 4.6021434819815555e-05, | |
| "loss": 0.9808, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.02388399203866932, | |
| "grad_norm": 0.21778051555156708, | |
| "learning_rate": 4.568832711511125e-05, | |
| "loss": 0.9456, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.024736991754336083, | |
| "grad_norm": 0.21349306404590607, | |
| "learning_rate": 4.534313800996299e-05, | |
| "loss": 0.953, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.02558999147000284, | |
| "grad_norm": 0.21818219125270844, | |
| "learning_rate": 4.498606908508754e-05, | |
| "loss": 0.9133, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.026442991185669604, | |
| "grad_norm": 0.2510223388671875, | |
| "learning_rate": 4.46173288586818e-05, | |
| "loss": 0.9125, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.027295990901336366, | |
| "grad_norm": 0.24031595885753632, | |
| "learning_rate": 4.4237132664654154e-05, | |
| "loss": 0.8784, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.028148990617003128, | |
| "grad_norm": 0.2543729543685913, | |
| "learning_rate": 4.384570252687542e-05, | |
| "loss": 0.8984, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.02900199033266989, | |
| "grad_norm": 0.27020716667175293, | |
| "learning_rate": 4.344326702952326e-05, | |
| "loss": 0.8719, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 0.02900199033266989, | |
| "eval_loss": 0.8574855923652649, | |
| "eval_runtime": 168.0651, | |
| "eval_samples_per_second": 35.248, | |
| "eval_steps_per_second": 4.409, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 0.029854990048336652, | |
| "grad_norm": 0.2670081853866577, | |
| "learning_rate": 4.303006118359537e-05, | |
| "loss": 0.8789, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.03070798976400341, | |
| "grad_norm": 0.3037591874599457, | |
| "learning_rate": 4.260632628966974e-05, | |
| "loss": 0.8526, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 0.031560989479670176, | |
| "grad_norm": 0.28440290689468384, | |
| "learning_rate": 4.217230979699188e-05, | |
| "loss": 0.8274, | |
| "step": 111 | |
| }, | |
| { | |
| "epoch": 0.03241398919533694, | |
| "grad_norm": 0.49433842301368713, | |
| "learning_rate": 4.172826515897146e-05, | |
| "loss": 0.8104, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 0.03326698891100369, | |
| "grad_norm": 0.2911880314350128, | |
| "learning_rate": 4.12744516851726e-05, | |
| "loss": 0.8392, | |
| "step": 117 | |
| }, | |
| { | |
| "epoch": 0.034119988626670456, | |
| "grad_norm": 0.3204285502433777, | |
| "learning_rate": 4.0811134389884433e-05, | |
| "loss": 0.788, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.03497298834233722, | |
| "grad_norm": 0.35650599002838135, | |
| "learning_rate": 4.0338583837360225e-05, | |
| "loss": 0.8004, | |
| "step": 123 | |
| }, | |
| { | |
| "epoch": 0.03582598805800398, | |
| "grad_norm": 0.310162216424942, | |
| "learning_rate": 3.985707598381544e-05, | |
| "loss": 0.8085, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 0.03667898777367074, | |
| "grad_norm": 0.3370768129825592, | |
| "learning_rate": 3.9366892016277096e-05, | |
| "loss": 0.7317, | |
| "step": 129 | |
| }, | |
| { | |
| "epoch": 0.037531987489337504, | |
| "grad_norm": 0.4600171446800232, | |
| "learning_rate": 3.886831818837847e-05, | |
| "loss": 0.7626, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 0.038384987205004266, | |
| "grad_norm": 0.43600377440452576, | |
| "learning_rate": 3.8361645653195026e-05, | |
| "loss": 0.7679, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.038669320443559856, | |
| "eval_loss": 0.7138456106185913, | |
| "eval_runtime": 168.0686, | |
| "eval_samples_per_second": 35.248, | |
| "eval_steps_per_second": 4.409, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 0.03923798692067103, | |
| "grad_norm": 0.4538515508174896, | |
| "learning_rate": 3.784717029321922e-05, | |
| "loss": 0.7585, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 0.04009098663633779, | |
| "grad_norm": 0.5923985838890076, | |
| "learning_rate": 3.732519254757344e-05, | |
| "loss": 0.7211, | |
| "step": 141 | |
| }, | |
| { | |
| "epoch": 0.04094398635200455, | |
| "grad_norm": 0.36592337489128113, | |
| "learning_rate": 3.679601723656205e-05, | |
| "loss": 0.6604, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 0.04179698606767131, | |
| "grad_norm": 0.44677990674972534, | |
| "learning_rate": 3.625995338366492e-05, | |
| "loss": 0.7846, | |
| "step": 147 | |
| }, | |
| { | |
| "epoch": 0.04264998578333807, | |
| "grad_norm": 0.4921860098838806, | |
| "learning_rate": 3.5717314035076355e-05, | |
| "loss": 0.6709, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.04350298549900483, | |
| "grad_norm": 0.5457685589790344, | |
| "learning_rate": 3.516841607689501e-05, | |
| "loss": 0.6974, | |
| "step": 153 | |
| }, | |
| { | |
| "epoch": 0.044355985214671594, | |
| "grad_norm": 0.46299344301223755, | |
| "learning_rate": 3.461358005007128e-05, | |
| "loss": 0.6643, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 0.045208984930338356, | |
| "grad_norm": 0.44877251982688904, | |
| "learning_rate": 3.405312996322042e-05, | |
| "loss": 0.6035, | |
| "step": 159 | |
| }, | |
| { | |
| "epoch": 0.04606198464600512, | |
| "grad_norm": 0.3635808527469635, | |
| "learning_rate": 3.348739310341068e-05, | |
| "loss": 0.6781, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 0.04691498436167188, | |
| "grad_norm": 0.4172018766403198, | |
| "learning_rate": 3.2916699845036816e-05, | |
| "loss": 0.6195, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 0.04776798407733864, | |
| "grad_norm": 0.36372795701026917, | |
| "learning_rate": 3.234138345689077e-05, | |
| "loss": 0.6599, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 0.048336650554449814, | |
| "eval_loss": 0.606797456741333, | |
| "eval_runtime": 168.0542, | |
| "eval_samples_per_second": 35.251, | |
| "eval_steps_per_second": 4.409, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.048620983793005404, | |
| "grad_norm": 0.4114479720592499, | |
| "learning_rate": 3.17617799075421e-05, | |
| "loss": 0.6287, | |
| "step": 171 | |
| }, | |
| { | |
| "epoch": 0.049473983508672166, | |
| "grad_norm": 0.36733385920524597, | |
| "learning_rate": 3.1178227669141744e-05, | |
| "loss": 0.621, | |
| "step": 174 | |
| }, | |
| { | |
| "epoch": 0.05032698322433893, | |
| "grad_norm": 0.3716995418071747, | |
| "learning_rate": 3.0591067519763895e-05, | |
| "loss": 0.5532, | |
| "step": 177 | |
| }, | |
| { | |
| "epoch": 0.05117998294000568, | |
| "grad_norm": 0.4997945725917816, | |
| "learning_rate": 3.0000642344401113e-05, | |
| "loss": 0.6024, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.052032982655672445, | |
| "grad_norm": 0.43555817008018494, | |
| "learning_rate": 2.9407296934729227e-05, | |
| "loss": 0.5092, | |
| "step": 183 | |
| }, | |
| { | |
| "epoch": 0.05288598237133921, | |
| "grad_norm": 0.49766504764556885, | |
| "learning_rate": 2.8811377787758636e-05, | |
| "loss": 0.6556, | |
| "step": 186 | |
| }, | |
| { | |
| "epoch": 0.05373898208700597, | |
| "grad_norm": 0.6171467304229736, | |
| "learning_rate": 2.8213232903489865e-05, | |
| "loss": 0.5736, | |
| "step": 189 | |
| }, | |
| { | |
| "epoch": 0.05459198180267273, | |
| "grad_norm": 0.8651450276374817, | |
| "learning_rate": 2.761321158169134e-05, | |
| "loss": 0.5568, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 0.055444981518339494, | |
| "grad_norm": 0.4486936926841736, | |
| "learning_rate": 2.7011664217918154e-05, | |
| "loss": 0.6145, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 0.056297981234006256, | |
| "grad_norm": 0.5348999500274658, | |
| "learning_rate": 2.6408942098890936e-05, | |
| "loss": 0.591, | |
| "step": 198 | |
| }, | |
| { | |
| "epoch": 0.05715098094967302, | |
| "grad_norm": 0.4498997628688812, | |
| "learning_rate": 2.580539719735433e-05, | |
| "loss": 0.5572, | |
| "step": 201 | |
| }, | |
| { | |
| "epoch": 0.05800398066533978, | |
| "grad_norm": 0.5493082404136658, | |
| "learning_rate": 2.5201381966534748e-05, | |
| "loss": 0.5316, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 0.05800398066533978, | |
| "eval_loss": 0.5341343879699707, | |
| "eval_runtime": 167.9533, | |
| "eval_samples_per_second": 35.272, | |
| "eval_steps_per_second": 4.412, | |
| "step": 204 | |
| } | |
| ], | |
| "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": 6.532800359222477e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |