| { | |
| "best_metric": 0.7410179640718563, | |
| "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-rsna-2018/checkpoint-1002", | |
| "epoch": 29.820359281437124, | |
| "eval_steps": 500, | |
| "global_step": 2490, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.11976047904191617, | |
| "grad_norm": 5.50510835647583, | |
| "learning_rate": 2.0080321285140564e-06, | |
| "loss": 0.7371, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.23952095808383234, | |
| "grad_norm": 10.76452350616455, | |
| "learning_rate": 4.016064257028113e-06, | |
| "loss": 0.7203, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.3592814371257485, | |
| "grad_norm": 6.238582611083984, | |
| "learning_rate": 6.024096385542169e-06, | |
| "loss": 0.6998, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.47904191616766467, | |
| "grad_norm": 6.8226189613342285, | |
| "learning_rate": 8.032128514056226e-06, | |
| "loss": 0.6751, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.5988023952095808, | |
| "grad_norm": 8.670806884765625, | |
| "learning_rate": 1.0040160642570281e-05, | |
| "loss": 0.6305, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.718562874251497, | |
| "grad_norm": 9.462418556213379, | |
| "learning_rate": 1.2048192771084338e-05, | |
| "loss": 0.6519, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.8383233532934131, | |
| "grad_norm": 4.775774955749512, | |
| "learning_rate": 1.4056224899598394e-05, | |
| "loss": 0.6457, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.9580838323353293, | |
| "grad_norm": 5.374617099761963, | |
| "learning_rate": 1.606425702811245e-05, | |
| "loss": 0.6448, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.9940119760479041, | |
| "eval_accuracy": 0.6736526946107785, | |
| "eval_loss": 0.6735118627548218, | |
| "eval_runtime": 4.2254, | |
| "eval_samples_per_second": 158.092, | |
| "eval_steps_per_second": 4.97, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 1.0778443113772456, | |
| "grad_norm": 9.03003215789795, | |
| "learning_rate": 1.8072289156626505e-05, | |
| "loss": 0.7148, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.1976047904191618, | |
| "grad_norm": 8.153128623962402, | |
| "learning_rate": 2.0080321285140562e-05, | |
| "loss": 0.7236, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.3173652694610778, | |
| "grad_norm": 6.323952674865723, | |
| "learning_rate": 2.208835341365462e-05, | |
| "loss": 0.7318, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.437125748502994, | |
| "grad_norm": 8.060946464538574, | |
| "learning_rate": 2.4096385542168677e-05, | |
| "loss": 0.7131, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.55688622754491, | |
| "grad_norm": 8.46826457977295, | |
| "learning_rate": 2.6104417670682734e-05, | |
| "loss": 0.7513, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.6766467065868262, | |
| "grad_norm": 6.310857772827148, | |
| "learning_rate": 2.8112449799196788e-05, | |
| "loss": 0.7172, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.7964071856287425, | |
| "grad_norm": 7.297194957733154, | |
| "learning_rate": 3.012048192771085e-05, | |
| "loss": 0.7141, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.9161676646706587, | |
| "grad_norm": 5.612814903259277, | |
| "learning_rate": 3.21285140562249e-05, | |
| "loss": 0.736, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.655688622754491, | |
| "eval_loss": 0.6968684196472168, | |
| "eval_runtime": 4.7501, | |
| "eval_samples_per_second": 140.628, | |
| "eval_steps_per_second": 4.421, | |
| "step": 167 | |
| }, | |
| { | |
| "epoch": 2.035928143712575, | |
| "grad_norm": 6.799407482147217, | |
| "learning_rate": 3.413654618473896e-05, | |
| "loss": 0.7162, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 2.155688622754491, | |
| "grad_norm": 5.372466087341309, | |
| "learning_rate": 3.614457831325301e-05, | |
| "loss": 0.7359, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 2.2754491017964074, | |
| "grad_norm": 7.812543869018555, | |
| "learning_rate": 3.815261044176707e-05, | |
| "loss": 0.7076, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 2.3952095808383236, | |
| "grad_norm": 4.029458522796631, | |
| "learning_rate": 4.0160642570281125e-05, | |
| "loss": 0.7415, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.5149700598802394, | |
| "grad_norm": 4.32454776763916, | |
| "learning_rate": 4.2168674698795186e-05, | |
| "loss": 0.721, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 2.6347305389221556, | |
| "grad_norm": 4.972012042999268, | |
| "learning_rate": 4.417670682730924e-05, | |
| "loss": 0.7117, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 2.754491017964072, | |
| "grad_norm": 4.34082555770874, | |
| "learning_rate": 4.61847389558233e-05, | |
| "loss": 0.7245, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 2.874251497005988, | |
| "grad_norm": 6.3038787841796875, | |
| "learning_rate": 4.8192771084337354e-05, | |
| "loss": 0.7116, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 2.9940119760479043, | |
| "grad_norm": 5.848024368286133, | |
| "learning_rate": 4.9977688531905406e-05, | |
| "loss": 0.6895, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.9940119760479043, | |
| "eval_accuracy": 0.6916167664670658, | |
| "eval_loss": 0.6264948844909668, | |
| "eval_runtime": 5.127, | |
| "eval_samples_per_second": 130.29, | |
| "eval_steps_per_second": 4.096, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 3.1137724550898205, | |
| "grad_norm": 5.597372055053711, | |
| "learning_rate": 4.97545738509594e-05, | |
| "loss": 0.6981, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 3.2335329341317367, | |
| "grad_norm": 3.8474349975585938, | |
| "learning_rate": 4.953145917001339e-05, | |
| "loss": 0.6821, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 3.3532934131736525, | |
| "grad_norm": 4.591520309448242, | |
| "learning_rate": 4.930834448906738e-05, | |
| "loss": 0.6935, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 3.4730538922155687, | |
| "grad_norm": 7.39481258392334, | |
| "learning_rate": 4.908522980812137e-05, | |
| "loss": 0.7123, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 3.592814371257485, | |
| "grad_norm": 4.609263896942139, | |
| "learning_rate": 4.886211512717537e-05, | |
| "loss": 0.7209, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.712574850299401, | |
| "grad_norm": 6.359745502471924, | |
| "learning_rate": 4.8639000446229364e-05, | |
| "loss": 0.6891, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 3.8323353293413174, | |
| "grad_norm": 3.9733974933624268, | |
| "learning_rate": 4.8415885765283355e-05, | |
| "loss": 0.6796, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 3.9520958083832336, | |
| "grad_norm": 5.370658874511719, | |
| "learning_rate": 4.8192771084337354e-05, | |
| "loss": 0.6631, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.7155688622754491, | |
| "eval_loss": 0.627507209777832, | |
| "eval_runtime": 5.0645, | |
| "eval_samples_per_second": 131.899, | |
| "eval_steps_per_second": 4.147, | |
| "step": 334 | |
| }, | |
| { | |
| "epoch": 4.07185628742515, | |
| "grad_norm": 4.173786163330078, | |
| "learning_rate": 4.7969656403391346e-05, | |
| "loss": 0.6847, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 4.191616766467066, | |
| "grad_norm": 8.743853569030762, | |
| "learning_rate": 4.774654172244534e-05, | |
| "loss": 0.6653, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 4.311377245508982, | |
| "grad_norm": 3.8112645149230957, | |
| "learning_rate": 4.7523427041499336e-05, | |
| "loss": 0.682, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 4.431137724550898, | |
| "grad_norm": 4.91067361831665, | |
| "learning_rate": 4.730031236055333e-05, | |
| "loss": 0.6849, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 4.550898203592815, | |
| "grad_norm": 7.876720428466797, | |
| "learning_rate": 4.707719767960732e-05, | |
| "loss": 0.665, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 4.6706586826347305, | |
| "grad_norm": 3.3110787868499756, | |
| "learning_rate": 4.685408299866131e-05, | |
| "loss": 0.6923, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 4.790419161676647, | |
| "grad_norm": 4.038461208343506, | |
| "learning_rate": 4.663096831771531e-05, | |
| "loss": 0.6776, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 4.910179640718563, | |
| "grad_norm": 4.028420448303223, | |
| "learning_rate": 4.64078536367693e-05, | |
| "loss": 0.6725, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 4.994011976047904, | |
| "eval_accuracy": 0.7125748502994012, | |
| "eval_loss": 0.6311057806015015, | |
| "eval_runtime": 5.1897, | |
| "eval_samples_per_second": 128.717, | |
| "eval_steps_per_second": 4.046, | |
| "step": 417 | |
| }, | |
| { | |
| "epoch": 5.029940119760479, | |
| "grad_norm": 4.9200215339660645, | |
| "learning_rate": 4.61847389558233e-05, | |
| "loss": 0.6848, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 5.149700598802395, | |
| "grad_norm": 11.384115219116211, | |
| "learning_rate": 4.596162427487729e-05, | |
| "loss": 0.6651, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 5.269461077844311, | |
| "grad_norm": 4.320120811462402, | |
| "learning_rate": 4.5738509593931284e-05, | |
| "loss": 0.6627, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 5.389221556886228, | |
| "grad_norm": 3.2249197959899902, | |
| "learning_rate": 4.5515394912985275e-05, | |
| "loss": 0.695, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 5.508982035928144, | |
| "grad_norm": 6.424835681915283, | |
| "learning_rate": 4.529228023203927e-05, | |
| "loss": 0.665, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 5.62874251497006, | |
| "grad_norm": 3.735926389694214, | |
| "learning_rate": 4.506916555109326e-05, | |
| "loss": 0.6484, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 5.748502994011976, | |
| "grad_norm": 6.180431365966797, | |
| "learning_rate": 4.484605087014726e-05, | |
| "loss": 0.6623, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 5.868263473053892, | |
| "grad_norm": 4.55112886428833, | |
| "learning_rate": 4.4622936189201256e-05, | |
| "loss": 0.6954, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 5.9880239520958085, | |
| "grad_norm": 5.922323226928711, | |
| "learning_rate": 4.439982150825525e-05, | |
| "loss": 0.6778, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.7065868263473054, | |
| "eval_loss": 0.619443953037262, | |
| "eval_runtime": 4.9491, | |
| "eval_samples_per_second": 134.974, | |
| "eval_steps_per_second": 4.243, | |
| "step": 501 | |
| }, | |
| { | |
| "epoch": 6.107784431137724, | |
| "grad_norm": 4.259535789489746, | |
| "learning_rate": 4.417670682730924e-05, | |
| "loss": 0.6365, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 6.227544910179641, | |
| "grad_norm": 3.808413028717041, | |
| "learning_rate": 4.395359214636323e-05, | |
| "loss": 0.6913, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 6.347305389221557, | |
| "grad_norm": 4.0178632736206055, | |
| "learning_rate": 4.373047746541722e-05, | |
| "loss": 0.7112, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 6.467065868263473, | |
| "grad_norm": 3.5464377403259277, | |
| "learning_rate": 4.350736278447122e-05, | |
| "loss": 0.6287, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 6.586826347305389, | |
| "grad_norm": 5.3495612144470215, | |
| "learning_rate": 4.328424810352521e-05, | |
| "loss": 0.6441, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 6.706586826347305, | |
| "grad_norm": 3.6895763874053955, | |
| "learning_rate": 4.306113342257921e-05, | |
| "loss": 0.6492, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 6.826347305389222, | |
| "grad_norm": 6.391328811645508, | |
| "learning_rate": 4.2838018741633203e-05, | |
| "loss": 0.6347, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 6.946107784431137, | |
| "grad_norm": 3.929858922958374, | |
| "learning_rate": 4.2614904060687195e-05, | |
| "loss": 0.6734, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 6.994011976047904, | |
| "eval_accuracy": 0.7140718562874252, | |
| "eval_loss": 0.602377712726593, | |
| "eval_runtime": 5.4236, | |
| "eval_samples_per_second": 123.166, | |
| "eval_steps_per_second": 3.872, | |
| "step": 584 | |
| }, | |
| { | |
| "epoch": 7.065868263473054, | |
| "grad_norm": 3.6292643547058105, | |
| "learning_rate": 4.239178937974119e-05, | |
| "loss": 0.6651, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 7.18562874251497, | |
| "grad_norm": 5.194599628448486, | |
| "learning_rate": 4.2168674698795186e-05, | |
| "loss": 0.643, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 7.3053892215568865, | |
| "grad_norm": 4.0095953941345215, | |
| "learning_rate": 4.194556001784918e-05, | |
| "loss": 0.6436, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 7.425149700598802, | |
| "grad_norm": 4.267141819000244, | |
| "learning_rate": 4.172244533690317e-05, | |
| "loss": 0.6114, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 7.544910179640718, | |
| "grad_norm": 3.5301904678344727, | |
| "learning_rate": 4.149933065595716e-05, | |
| "loss": 0.6326, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 7.664670658682635, | |
| "grad_norm": 3.8866894245147705, | |
| "learning_rate": 4.127621597501116e-05, | |
| "loss": 0.6331, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 7.7844311377245505, | |
| "grad_norm": 3.902667284011841, | |
| "learning_rate": 4.105310129406515e-05, | |
| "loss": 0.6515, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 7.904191616766467, | |
| "grad_norm": 4.829390525817871, | |
| "learning_rate": 4.082998661311915e-05, | |
| "loss": 0.6231, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.7230538922155688, | |
| "eval_loss": 0.6081866025924683, | |
| "eval_runtime": 5.7887, | |
| "eval_samples_per_second": 115.397, | |
| "eval_steps_per_second": 3.628, | |
| "step": 668 | |
| }, | |
| { | |
| "epoch": 8.023952095808383, | |
| "grad_norm": 4.755038261413574, | |
| "learning_rate": 4.060687193217314e-05, | |
| "loss": 0.6261, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 8.1437125748503, | |
| "grad_norm": 3.4586455821990967, | |
| "learning_rate": 4.038375725122713e-05, | |
| "loss": 0.6534, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 8.263473053892216, | |
| "grad_norm": 8.194857597351074, | |
| "learning_rate": 4.0160642570281125e-05, | |
| "loss": 0.6329, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 8.383233532934131, | |
| "grad_norm": 2.9734175205230713, | |
| "learning_rate": 3.993752788933512e-05, | |
| "loss": 0.6676, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 8.502994011976048, | |
| "grad_norm": 5.66069221496582, | |
| "learning_rate": 3.9714413208389115e-05, | |
| "loss": 0.6178, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 8.622754491017965, | |
| "grad_norm": 3.4166412353515625, | |
| "learning_rate": 3.949129852744311e-05, | |
| "loss": 0.6399, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 8.74251497005988, | |
| "grad_norm": 5.076518535614014, | |
| "learning_rate": 3.9268183846497105e-05, | |
| "loss": 0.6163, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 8.862275449101796, | |
| "grad_norm": 3.32446551322937, | |
| "learning_rate": 3.90450691655511e-05, | |
| "loss": 0.6152, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 8.982035928143713, | |
| "grad_norm": 4.711836338043213, | |
| "learning_rate": 3.882195448460509e-05, | |
| "loss": 0.6164, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 8.994011976047904, | |
| "eval_accuracy": 0.7170658682634731, | |
| "eval_loss": 0.5845786333084106, | |
| "eval_runtime": 4.984, | |
| "eval_samples_per_second": 134.029, | |
| "eval_steps_per_second": 4.213, | |
| "step": 751 | |
| }, | |
| { | |
| "epoch": 9.10179640718563, | |
| "grad_norm": 4.9026055335998535, | |
| "learning_rate": 3.859883980365908e-05, | |
| "loss": 0.623, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 9.221556886227544, | |
| "grad_norm": 5.39790678024292, | |
| "learning_rate": 3.837572512271307e-05, | |
| "loss": 0.5958, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 9.341317365269461, | |
| "grad_norm": 4.951222896575928, | |
| "learning_rate": 3.815261044176707e-05, | |
| "loss": 0.6206, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 9.461077844311378, | |
| "grad_norm": 3.8360514640808105, | |
| "learning_rate": 3.792949576082106e-05, | |
| "loss": 0.638, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 9.580838323353294, | |
| "grad_norm": 4.05393123626709, | |
| "learning_rate": 3.770638107987506e-05, | |
| "loss": 0.6561, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 9.70059880239521, | |
| "grad_norm": 5.898914337158203, | |
| "learning_rate": 3.748326639892905e-05, | |
| "loss": 0.6197, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 9.820359281437126, | |
| "grad_norm": 4.737303733825684, | |
| "learning_rate": 3.7260151717983045e-05, | |
| "loss": 0.6068, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 9.940119760479043, | |
| "grad_norm": 3.769287347793579, | |
| "learning_rate": 3.7037037037037037e-05, | |
| "loss": 0.6261, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.7380239520958084, | |
| "eval_loss": 0.568150520324707, | |
| "eval_runtime": 5.0718, | |
| "eval_samples_per_second": 131.708, | |
| "eval_steps_per_second": 4.141, | |
| "step": 835 | |
| }, | |
| { | |
| "epoch": 10.059880239520957, | |
| "grad_norm": 4.2985920906066895, | |
| "learning_rate": 3.6813922356091035e-05, | |
| "loss": 0.6455, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 10.179640718562874, | |
| "grad_norm": 4.344922065734863, | |
| "learning_rate": 3.659080767514503e-05, | |
| "loss": 0.6438, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 10.29940119760479, | |
| "grad_norm": 4.293480396270752, | |
| "learning_rate": 3.636769299419902e-05, | |
| "loss": 0.6161, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 10.419161676646706, | |
| "grad_norm": 5.124499797821045, | |
| "learning_rate": 3.614457831325301e-05, | |
| "loss": 0.5976, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 10.538922155688622, | |
| "grad_norm": 3.7405636310577393, | |
| "learning_rate": 3.592146363230701e-05, | |
| "loss": 0.6125, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 10.658682634730539, | |
| "grad_norm": 3.614593744277954, | |
| "learning_rate": 3.5698348951361e-05, | |
| "loss": 0.6159, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 10.778443113772456, | |
| "grad_norm": 3.6515111923217773, | |
| "learning_rate": 3.5475234270415e-05, | |
| "loss": 0.6095, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 10.89820359281437, | |
| "grad_norm": 4.4123215675354, | |
| "learning_rate": 3.525211958946899e-05, | |
| "loss": 0.6153, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 10.994011976047904, | |
| "eval_accuracy": 0.718562874251497, | |
| "eval_loss": 0.6006675362586975, | |
| "eval_runtime": 4.9154, | |
| "eval_samples_per_second": 135.9, | |
| "eval_steps_per_second": 4.272, | |
| "step": 918 | |
| }, | |
| { | |
| "epoch": 11.017964071856287, | |
| "grad_norm": 3.9871232509613037, | |
| "learning_rate": 3.502900490852298e-05, | |
| "loss": 0.6124, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 11.137724550898204, | |
| "grad_norm": 4.505560874938965, | |
| "learning_rate": 3.4805890227576974e-05, | |
| "loss": 0.602, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 11.25748502994012, | |
| "grad_norm": 4.445052623748779, | |
| "learning_rate": 3.4582775546630966e-05, | |
| "loss": 0.6002, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 11.377245508982035, | |
| "grad_norm": 3.532015562057495, | |
| "learning_rate": 3.4359660865684965e-05, | |
| "loss": 0.6242, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 11.497005988023952, | |
| "grad_norm": 6.202284812927246, | |
| "learning_rate": 3.413654618473896e-05, | |
| "loss": 0.6113, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 11.616766467065869, | |
| "grad_norm": 5.8437089920043945, | |
| "learning_rate": 3.3913431503792955e-05, | |
| "loss": 0.6037, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 11.736526946107784, | |
| "grad_norm": 4.447215557098389, | |
| "learning_rate": 3.369031682284695e-05, | |
| "loss": 0.6356, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 11.8562874251497, | |
| "grad_norm": 4.645685195922852, | |
| "learning_rate": 3.346720214190094e-05, | |
| "loss": 0.6309, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 11.976047904191617, | |
| "grad_norm": 4.300328731536865, | |
| "learning_rate": 3.324408746095493e-05, | |
| "loss": 0.6046, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.7410179640718563, | |
| "eval_loss": 0.5744568705558777, | |
| "eval_runtime": 4.8113, | |
| "eval_samples_per_second": 138.84, | |
| "eval_steps_per_second": 4.365, | |
| "step": 1002 | |
| }, | |
| { | |
| "epoch": 12.095808383233534, | |
| "grad_norm": 3.7963852882385254, | |
| "learning_rate": 3.302097278000892e-05, | |
| "loss": 0.5979, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 12.215568862275449, | |
| "grad_norm": 5.136913776397705, | |
| "learning_rate": 3.279785809906292e-05, | |
| "loss": 0.6082, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 12.335329341317365, | |
| "grad_norm": 5.207279205322266, | |
| "learning_rate": 3.257474341811691e-05, | |
| "loss": 0.5884, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 12.455089820359282, | |
| "grad_norm": 4.387267589569092, | |
| "learning_rate": 3.235162873717091e-05, | |
| "loss": 0.6156, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 12.574850299401197, | |
| "grad_norm": 3.0785038471221924, | |
| "learning_rate": 3.21285140562249e-05, | |
| "loss": 0.6093, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 12.694610778443113, | |
| "grad_norm": 3.2741942405700684, | |
| "learning_rate": 3.1905399375278894e-05, | |
| "loss": 0.6033, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 12.81437125748503, | |
| "grad_norm": 4.547226428985596, | |
| "learning_rate": 3.1682284694332886e-05, | |
| "loss": 0.6157, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 12.934131736526947, | |
| "grad_norm": 4.921385765075684, | |
| "learning_rate": 3.1459170013386885e-05, | |
| "loss": 0.5679, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 12.994011976047904, | |
| "eval_accuracy": 0.7230538922155688, | |
| "eval_loss": 0.595708429813385, | |
| "eval_runtime": 4.9064, | |
| "eval_samples_per_second": 136.149, | |
| "eval_steps_per_second": 4.28, | |
| "step": 1085 | |
| }, | |
| { | |
| "epoch": 13.053892215568862, | |
| "grad_norm": 6.123495578765869, | |
| "learning_rate": 3.1236055332440876e-05, | |
| "loss": 0.5848, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 13.173652694610778, | |
| "grad_norm": 3.932276964187622, | |
| "learning_rate": 3.101294065149487e-05, | |
| "loss": 0.6018, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 13.293413173652695, | |
| "grad_norm": 4.827797889709473, | |
| "learning_rate": 3.078982597054887e-05, | |
| "loss": 0.6069, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 13.41317365269461, | |
| "grad_norm": 3.6165199279785156, | |
| "learning_rate": 3.056671128960286e-05, | |
| "loss": 0.6036, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 13.532934131736527, | |
| "grad_norm": 4.21213960647583, | |
| "learning_rate": 3.034359660865685e-05, | |
| "loss": 0.5872, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 13.652694610778443, | |
| "grad_norm": 6.651663780212402, | |
| "learning_rate": 3.012048192771085e-05, | |
| "loss": 0.6057, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 13.77245508982036, | |
| "grad_norm": 5.351555824279785, | |
| "learning_rate": 2.989736724676484e-05, | |
| "loss": 0.5951, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 13.892215568862275, | |
| "grad_norm": 5.491767406463623, | |
| "learning_rate": 2.9674252565818832e-05, | |
| "loss": 0.6027, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.7215568862275449, | |
| "eval_loss": 0.5884155631065369, | |
| "eval_runtime": 4.7316, | |
| "eval_samples_per_second": 141.178, | |
| "eval_steps_per_second": 4.438, | |
| "step": 1169 | |
| }, | |
| { | |
| "epoch": 14.011976047904191, | |
| "grad_norm": 4.720607757568359, | |
| "learning_rate": 2.9451137884872827e-05, | |
| "loss": 0.5779, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 14.131736526946108, | |
| "grad_norm": 4.377676963806152, | |
| "learning_rate": 2.922802320392682e-05, | |
| "loss": 0.5992, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 14.251497005988025, | |
| "grad_norm": 4.529723644256592, | |
| "learning_rate": 2.900490852298081e-05, | |
| "loss": 0.6127, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 14.37125748502994, | |
| "grad_norm": 3.385350465774536, | |
| "learning_rate": 2.878179384203481e-05, | |
| "loss": 0.5876, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 14.491017964071856, | |
| "grad_norm": 5.14049768447876, | |
| "learning_rate": 2.85586791610888e-05, | |
| "loss": 0.5685, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 14.610778443113773, | |
| "grad_norm": 6.226632118225098, | |
| "learning_rate": 2.8335564480142796e-05, | |
| "loss": 0.5957, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 14.730538922155688, | |
| "grad_norm": 4.943429470062256, | |
| "learning_rate": 2.8112449799196788e-05, | |
| "loss": 0.5709, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 14.850299401197605, | |
| "grad_norm": 3.945502758026123, | |
| "learning_rate": 2.788933511825078e-05, | |
| "loss": 0.5807, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 14.970059880239521, | |
| "grad_norm": 5.3703718185424805, | |
| "learning_rate": 2.7666220437304775e-05, | |
| "loss": 0.6249, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 14.994011976047904, | |
| "eval_accuracy": 0.7365269461077845, | |
| "eval_loss": 0.5808472037315369, | |
| "eval_runtime": 4.8318, | |
| "eval_samples_per_second": 138.251, | |
| "eval_steps_per_second": 4.346, | |
| "step": 1252 | |
| }, | |
| { | |
| "epoch": 15.089820359281438, | |
| "grad_norm": 3.6692726612091064, | |
| "learning_rate": 2.7443105756358774e-05, | |
| "loss": 0.5944, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 15.209580838323353, | |
| "grad_norm": 4.581055164337158, | |
| "learning_rate": 2.7219991075412765e-05, | |
| "loss": 0.5599, | |
| "step": 1270 | |
| }, | |
| { | |
| "epoch": 15.32934131736527, | |
| "grad_norm": 3.810741901397705, | |
| "learning_rate": 2.6996876394466757e-05, | |
| "loss": 0.577, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 15.449101796407186, | |
| "grad_norm": 4.873391151428223, | |
| "learning_rate": 2.6773761713520752e-05, | |
| "loss": 0.565, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 15.568862275449101, | |
| "grad_norm": 4.216405391693115, | |
| "learning_rate": 2.6550647032574744e-05, | |
| "loss": 0.5709, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 15.688622754491018, | |
| "grad_norm": 4.651778697967529, | |
| "learning_rate": 2.6327532351628736e-05, | |
| "loss": 0.5984, | |
| "step": 1310 | |
| }, | |
| { | |
| "epoch": 15.808383233532934, | |
| "grad_norm": 4.126067638397217, | |
| "learning_rate": 2.6104417670682734e-05, | |
| "loss": 0.5718, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 15.928143712574851, | |
| "grad_norm": 3.843979597091675, | |
| "learning_rate": 2.5881302989736726e-05, | |
| "loss": 0.6059, | |
| "step": 1330 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.7350299401197605, | |
| "eval_loss": 0.5699232816696167, | |
| "eval_runtime": 5.1217, | |
| "eval_samples_per_second": 130.426, | |
| "eval_steps_per_second": 4.1, | |
| "step": 1336 | |
| }, | |
| { | |
| "epoch": 16.047904191616766, | |
| "grad_norm": 3.6033108234405518, | |
| "learning_rate": 2.565818830879072e-05, | |
| "loss": 0.5773, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 16.16766467065868, | |
| "grad_norm": 5.372034549713135, | |
| "learning_rate": 2.5435073627844713e-05, | |
| "loss": 0.5818, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 16.2874251497006, | |
| "grad_norm": 4.0521955490112305, | |
| "learning_rate": 2.5211958946898705e-05, | |
| "loss": 0.5517, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 16.407185628742514, | |
| "grad_norm": 4.455647945404053, | |
| "learning_rate": 2.4988844265952703e-05, | |
| "loss": 0.5904, | |
| "step": 1370 | |
| }, | |
| { | |
| "epoch": 16.526946107784433, | |
| "grad_norm": 4.8622589111328125, | |
| "learning_rate": 2.4765729585006695e-05, | |
| "loss": 0.5943, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 16.646706586826348, | |
| "grad_norm": 5.169972896575928, | |
| "learning_rate": 2.4542614904060687e-05, | |
| "loss": 0.5841, | |
| "step": 1390 | |
| }, | |
| { | |
| "epoch": 16.766467065868262, | |
| "grad_norm": 4.88759708404541, | |
| "learning_rate": 2.4319500223114682e-05, | |
| "loss": 0.5639, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 16.88622754491018, | |
| "grad_norm": 5.843952178955078, | |
| "learning_rate": 2.4096385542168677e-05, | |
| "loss": 0.5776, | |
| "step": 1410 | |
| }, | |
| { | |
| "epoch": 16.994011976047904, | |
| "eval_accuracy": 0.7320359281437125, | |
| "eval_loss": 0.5769894123077393, | |
| "eval_runtime": 5.3536, | |
| "eval_samples_per_second": 124.775, | |
| "eval_steps_per_second": 3.923, | |
| "step": 1419 | |
| }, | |
| { | |
| "epoch": 17.005988023952096, | |
| "grad_norm": 8.036992073059082, | |
| "learning_rate": 2.387327086122267e-05, | |
| "loss": 0.5775, | |
| "step": 1420 | |
| }, | |
| { | |
| "epoch": 17.12574850299401, | |
| "grad_norm": 8.761232376098633, | |
| "learning_rate": 2.3650156180276664e-05, | |
| "loss": 0.56, | |
| "step": 1430 | |
| }, | |
| { | |
| "epoch": 17.24550898203593, | |
| "grad_norm": 3.9877655506134033, | |
| "learning_rate": 2.3427041499330656e-05, | |
| "loss": 0.5409, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 17.365269461077844, | |
| "grad_norm": 8.304950714111328, | |
| "learning_rate": 2.320392681838465e-05, | |
| "loss": 0.5696, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 17.48502994011976, | |
| "grad_norm": 4.155704498291016, | |
| "learning_rate": 2.2980812137438646e-05, | |
| "loss": 0.5888, | |
| "step": 1460 | |
| }, | |
| { | |
| "epoch": 17.604790419161677, | |
| "grad_norm": 4.174530029296875, | |
| "learning_rate": 2.2757697456492638e-05, | |
| "loss": 0.5566, | |
| "step": 1470 | |
| }, | |
| { | |
| "epoch": 17.724550898203592, | |
| "grad_norm": 6.5803093910217285, | |
| "learning_rate": 2.253458277554663e-05, | |
| "loss": 0.5385, | |
| "step": 1480 | |
| }, | |
| { | |
| "epoch": 17.84431137724551, | |
| "grad_norm": 4.418693542480469, | |
| "learning_rate": 2.2311468094600628e-05, | |
| "loss": 0.5411, | |
| "step": 1490 | |
| }, | |
| { | |
| "epoch": 17.964071856287426, | |
| "grad_norm": 4.162361145019531, | |
| "learning_rate": 2.208835341365462e-05, | |
| "loss": 0.5903, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.7215568862275449, | |
| "eval_loss": 0.5806027054786682, | |
| "eval_runtime": 4.6322, | |
| "eval_samples_per_second": 144.209, | |
| "eval_steps_per_second": 4.534, | |
| "step": 1503 | |
| }, | |
| { | |
| "epoch": 18.08383233532934, | |
| "grad_norm": 4.274399280548096, | |
| "learning_rate": 2.186523873270861e-05, | |
| "loss": 0.5612, | |
| "step": 1510 | |
| }, | |
| { | |
| "epoch": 18.20359281437126, | |
| "grad_norm": 5.013104438781738, | |
| "learning_rate": 2.1642124051762607e-05, | |
| "loss": 0.5654, | |
| "step": 1520 | |
| }, | |
| { | |
| "epoch": 18.323353293413174, | |
| "grad_norm": 4.541128635406494, | |
| "learning_rate": 2.1419009370816602e-05, | |
| "loss": 0.5492, | |
| "step": 1530 | |
| }, | |
| { | |
| "epoch": 18.44311377245509, | |
| "grad_norm": 4.409793853759766, | |
| "learning_rate": 2.1195894689870593e-05, | |
| "loss": 0.564, | |
| "step": 1540 | |
| }, | |
| { | |
| "epoch": 18.562874251497007, | |
| "grad_norm": 3.6504085063934326, | |
| "learning_rate": 2.097278000892459e-05, | |
| "loss": 0.5531, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 18.682634730538922, | |
| "grad_norm": 4.4471235275268555, | |
| "learning_rate": 2.074966532797858e-05, | |
| "loss": 0.5911, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 18.802395209580837, | |
| "grad_norm": 4.53286600112915, | |
| "learning_rate": 2.0526550647032576e-05, | |
| "loss": 0.572, | |
| "step": 1570 | |
| }, | |
| { | |
| "epoch": 18.922155688622755, | |
| "grad_norm": 4.156210899353027, | |
| "learning_rate": 2.030343596608657e-05, | |
| "loss": 0.5633, | |
| "step": 1580 | |
| }, | |
| { | |
| "epoch": 18.994011976047904, | |
| "eval_accuracy": 0.7380239520958084, | |
| "eval_loss": 0.576755166053772, | |
| "eval_runtime": 5.1497, | |
| "eval_samples_per_second": 129.715, | |
| "eval_steps_per_second": 4.078, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 19.04191616766467, | |
| "grad_norm": 4.36007022857666, | |
| "learning_rate": 2.0080321285140562e-05, | |
| "loss": 0.5304, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 19.161676646706585, | |
| "grad_norm": 6.453935623168945, | |
| "learning_rate": 1.9857206604194558e-05, | |
| "loss": 0.5595, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 19.281437125748504, | |
| "grad_norm": 4.35880708694458, | |
| "learning_rate": 1.9634091923248553e-05, | |
| "loss": 0.5578, | |
| "step": 1610 | |
| }, | |
| { | |
| "epoch": 19.40119760479042, | |
| "grad_norm": 4.454617500305176, | |
| "learning_rate": 1.9410977242302544e-05, | |
| "loss": 0.532, | |
| "step": 1620 | |
| }, | |
| { | |
| "epoch": 19.520958083832337, | |
| "grad_norm": 4.35628604888916, | |
| "learning_rate": 1.9187862561356536e-05, | |
| "loss": 0.5433, | |
| "step": 1630 | |
| }, | |
| { | |
| "epoch": 19.64071856287425, | |
| "grad_norm": 4.488490104675293, | |
| "learning_rate": 1.896474788041053e-05, | |
| "loss": 0.5545, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 19.760479041916167, | |
| "grad_norm": 5.309743881225586, | |
| "learning_rate": 1.8741633199464527e-05, | |
| "loss": 0.5793, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 19.880239520958085, | |
| "grad_norm": 4.599639415740967, | |
| "learning_rate": 1.8518518518518518e-05, | |
| "loss": 0.55, | |
| "step": 1660 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 4.7093892097473145, | |
| "learning_rate": 1.8295403837572513e-05, | |
| "loss": 0.5544, | |
| "step": 1670 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.7350299401197605, | |
| "eval_loss": 0.5829503536224365, | |
| "eval_runtime": 4.8934, | |
| "eval_samples_per_second": 136.511, | |
| "eval_steps_per_second": 4.292, | |
| "step": 1670 | |
| }, | |
| { | |
| "epoch": 20.119760479041915, | |
| "grad_norm": 4.780275344848633, | |
| "learning_rate": 1.8072289156626505e-05, | |
| "loss": 0.5465, | |
| "step": 1680 | |
| }, | |
| { | |
| "epoch": 20.239520958083833, | |
| "grad_norm": 4.53477668762207, | |
| "learning_rate": 1.78491744756805e-05, | |
| "loss": 0.525, | |
| "step": 1690 | |
| }, | |
| { | |
| "epoch": 20.35928143712575, | |
| "grad_norm": 5.508608818054199, | |
| "learning_rate": 1.7626059794734495e-05, | |
| "loss": 0.5768, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 20.479041916167663, | |
| "grad_norm": 5.174022197723389, | |
| "learning_rate": 1.7402945113788487e-05, | |
| "loss": 0.5308, | |
| "step": 1710 | |
| }, | |
| { | |
| "epoch": 20.59880239520958, | |
| "grad_norm": 4.175419330596924, | |
| "learning_rate": 1.7179830432842482e-05, | |
| "loss": 0.5609, | |
| "step": 1720 | |
| }, | |
| { | |
| "epoch": 20.718562874251496, | |
| "grad_norm": 4.944879055023193, | |
| "learning_rate": 1.6956715751896478e-05, | |
| "loss": 0.5323, | |
| "step": 1730 | |
| }, | |
| { | |
| "epoch": 20.83832335329341, | |
| "grad_norm": 4.967810153961182, | |
| "learning_rate": 1.673360107095047e-05, | |
| "loss": 0.5234, | |
| "step": 1740 | |
| }, | |
| { | |
| "epoch": 20.95808383233533, | |
| "grad_norm": 5.018576622009277, | |
| "learning_rate": 1.651048639000446e-05, | |
| "loss": 0.5515, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 20.994011976047904, | |
| "eval_accuracy": 0.7260479041916168, | |
| "eval_loss": 0.5965989232063293, | |
| "eval_runtime": 4.6792, | |
| "eval_samples_per_second": 142.759, | |
| "eval_steps_per_second": 4.488, | |
| "step": 1753 | |
| }, | |
| { | |
| "epoch": 21.077844311377245, | |
| "grad_norm": 4.204094886779785, | |
| "learning_rate": 1.6287371709058456e-05, | |
| "loss": 0.5672, | |
| "step": 1760 | |
| }, | |
| { | |
| "epoch": 21.197604790419163, | |
| "grad_norm": 4.336850643157959, | |
| "learning_rate": 1.606425702811245e-05, | |
| "loss": 0.5011, | |
| "step": 1770 | |
| }, | |
| { | |
| "epoch": 21.317365269461078, | |
| "grad_norm": 5.5377583503723145, | |
| "learning_rate": 1.5841142347166443e-05, | |
| "loss": 0.5755, | |
| "step": 1780 | |
| }, | |
| { | |
| "epoch": 21.437125748502993, | |
| "grad_norm": 5.438154697418213, | |
| "learning_rate": 1.5618027666220438e-05, | |
| "loss": 0.5349, | |
| "step": 1790 | |
| }, | |
| { | |
| "epoch": 21.55688622754491, | |
| "grad_norm": 5.496122360229492, | |
| "learning_rate": 1.5394912985274433e-05, | |
| "loss": 0.5754, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 21.676646706586826, | |
| "grad_norm": 5.726380348205566, | |
| "learning_rate": 1.5171798304328425e-05, | |
| "loss": 0.5778, | |
| "step": 1810 | |
| }, | |
| { | |
| "epoch": 21.79640718562874, | |
| "grad_norm": 4.841021537780762, | |
| "learning_rate": 1.494868362338242e-05, | |
| "loss": 0.5269, | |
| "step": 1820 | |
| }, | |
| { | |
| "epoch": 21.91616766467066, | |
| "grad_norm": 4.8706583976745605, | |
| "learning_rate": 1.4725568942436414e-05, | |
| "loss": 0.5249, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_accuracy": 0.7335329341317365, | |
| "eval_loss": 0.6078537106513977, | |
| "eval_runtime": 4.6418, | |
| "eval_samples_per_second": 143.909, | |
| "eval_steps_per_second": 4.524, | |
| "step": 1837 | |
| }, | |
| { | |
| "epoch": 22.035928143712574, | |
| "grad_norm": 6.536875247955322, | |
| "learning_rate": 1.4502454261490405e-05, | |
| "loss": 0.5403, | |
| "step": 1840 | |
| }, | |
| { | |
| "epoch": 22.15568862275449, | |
| "grad_norm": 5.156835079193115, | |
| "learning_rate": 1.42793395805444e-05, | |
| "loss": 0.5547, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 22.275449101796408, | |
| "grad_norm": 4.475517749786377, | |
| "learning_rate": 1.4056224899598394e-05, | |
| "loss": 0.5474, | |
| "step": 1860 | |
| }, | |
| { | |
| "epoch": 22.395209580838323, | |
| "grad_norm": 5.913077354431152, | |
| "learning_rate": 1.3833110218652387e-05, | |
| "loss": 0.5242, | |
| "step": 1870 | |
| }, | |
| { | |
| "epoch": 22.51497005988024, | |
| "grad_norm": 4.135039806365967, | |
| "learning_rate": 1.3609995537706383e-05, | |
| "loss": 0.527, | |
| "step": 1880 | |
| }, | |
| { | |
| "epoch": 22.634730538922156, | |
| "grad_norm": 4.773129940032959, | |
| "learning_rate": 1.3386880856760376e-05, | |
| "loss": 0.5828, | |
| "step": 1890 | |
| }, | |
| { | |
| "epoch": 22.75449101796407, | |
| "grad_norm": 4.826302528381348, | |
| "learning_rate": 1.3163766175814368e-05, | |
| "loss": 0.5407, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 22.87425149700599, | |
| "grad_norm": 5.572017669677734, | |
| "learning_rate": 1.2940651494868363e-05, | |
| "loss": 0.5329, | |
| "step": 1910 | |
| }, | |
| { | |
| "epoch": 22.994011976047904, | |
| "grad_norm": 5.162022113800049, | |
| "learning_rate": 1.2717536813922356e-05, | |
| "loss": 0.5212, | |
| "step": 1920 | |
| }, | |
| { | |
| "epoch": 22.994011976047904, | |
| "eval_accuracy": 0.7245508982035929, | |
| "eval_loss": 0.5972306132316589, | |
| "eval_runtime": 5.0266, | |
| "eval_samples_per_second": 132.892, | |
| "eval_steps_per_second": 4.178, | |
| "step": 1920 | |
| }, | |
| { | |
| "epoch": 23.11377245508982, | |
| "grad_norm": 5.251833915710449, | |
| "learning_rate": 1.2494422132976352e-05, | |
| "loss": 0.5422, | |
| "step": 1930 | |
| }, | |
| { | |
| "epoch": 23.233532934131738, | |
| "grad_norm": 4.400450706481934, | |
| "learning_rate": 1.2271307452030343e-05, | |
| "loss": 0.512, | |
| "step": 1940 | |
| }, | |
| { | |
| "epoch": 23.353293413173652, | |
| "grad_norm": 5.726296901702881, | |
| "learning_rate": 1.2048192771084338e-05, | |
| "loss": 0.5167, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 23.473053892215567, | |
| "grad_norm": 5.346691131591797, | |
| "learning_rate": 1.1825078090138332e-05, | |
| "loss": 0.5554, | |
| "step": 1960 | |
| }, | |
| { | |
| "epoch": 23.592814371257486, | |
| "grad_norm": 6.358211994171143, | |
| "learning_rate": 1.1601963409192325e-05, | |
| "loss": 0.525, | |
| "step": 1970 | |
| }, | |
| { | |
| "epoch": 23.7125748502994, | |
| "grad_norm": 4.755873680114746, | |
| "learning_rate": 1.1378848728246319e-05, | |
| "loss": 0.5479, | |
| "step": 1980 | |
| }, | |
| { | |
| "epoch": 23.83233532934132, | |
| "grad_norm": 6.708542346954346, | |
| "learning_rate": 1.1155734047300314e-05, | |
| "loss": 0.5457, | |
| "step": 1990 | |
| }, | |
| { | |
| "epoch": 23.952095808383234, | |
| "grad_norm": 4.143124103546143, | |
| "learning_rate": 1.0932619366354306e-05, | |
| "loss": 0.5268, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.7230538922155688, | |
| "eval_loss": 0.5921865701675415, | |
| "eval_runtime": 5.3047, | |
| "eval_samples_per_second": 125.926, | |
| "eval_steps_per_second": 3.959, | |
| "step": 2004 | |
| }, | |
| { | |
| "epoch": 24.07185628742515, | |
| "grad_norm": 5.585824489593506, | |
| "learning_rate": 1.0709504685408301e-05, | |
| "loss": 0.4723, | |
| "step": 2010 | |
| }, | |
| { | |
| "epoch": 24.191616766467067, | |
| "grad_norm": 5.241277694702148, | |
| "learning_rate": 1.0486390004462294e-05, | |
| "loss": 0.5317, | |
| "step": 2020 | |
| }, | |
| { | |
| "epoch": 24.311377245508982, | |
| "grad_norm": 5.840533256530762, | |
| "learning_rate": 1.0263275323516288e-05, | |
| "loss": 0.5458, | |
| "step": 2030 | |
| }, | |
| { | |
| "epoch": 24.431137724550897, | |
| "grad_norm": 5.158961772918701, | |
| "learning_rate": 1.0040160642570281e-05, | |
| "loss": 0.5169, | |
| "step": 2040 | |
| }, | |
| { | |
| "epoch": 24.550898203592816, | |
| "grad_norm": 4.132058620452881, | |
| "learning_rate": 9.817045961624276e-06, | |
| "loss": 0.531, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 24.67065868263473, | |
| "grad_norm": 5.0692830085754395, | |
| "learning_rate": 9.593931280678268e-06, | |
| "loss": 0.5462, | |
| "step": 2060 | |
| }, | |
| { | |
| "epoch": 24.790419161676645, | |
| "grad_norm": 5.38627815246582, | |
| "learning_rate": 9.370816599732263e-06, | |
| "loss": 0.5151, | |
| "step": 2070 | |
| }, | |
| { | |
| "epoch": 24.910179640718564, | |
| "grad_norm": 5.605633735656738, | |
| "learning_rate": 9.147701918786257e-06, | |
| "loss": 0.5406, | |
| "step": 2080 | |
| }, | |
| { | |
| "epoch": 24.994011976047904, | |
| "eval_accuracy": 0.7350299401197605, | |
| "eval_loss": 0.609959602355957, | |
| "eval_runtime": 5.5754, | |
| "eval_samples_per_second": 119.811, | |
| "eval_steps_per_second": 3.767, | |
| "step": 2087 | |
| }, | |
| { | |
| "epoch": 25.02994011976048, | |
| "grad_norm": 5.846259117126465, | |
| "learning_rate": 8.92458723784025e-06, | |
| "loss": 0.5213, | |
| "step": 2090 | |
| }, | |
| { | |
| "epoch": 25.149700598802394, | |
| "grad_norm": 7.1593708992004395, | |
| "learning_rate": 8.701472556894244e-06, | |
| "loss": 0.5329, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 25.269461077844312, | |
| "grad_norm": 5.0290985107421875, | |
| "learning_rate": 8.478357875948239e-06, | |
| "loss": 0.5042, | |
| "step": 2110 | |
| }, | |
| { | |
| "epoch": 25.389221556886227, | |
| "grad_norm": 4.644502639770508, | |
| "learning_rate": 8.25524319500223e-06, | |
| "loss": 0.5227, | |
| "step": 2120 | |
| }, | |
| { | |
| "epoch": 25.508982035928145, | |
| "grad_norm": 5.292252540588379, | |
| "learning_rate": 8.032128514056226e-06, | |
| "loss": 0.5137, | |
| "step": 2130 | |
| }, | |
| { | |
| "epoch": 25.62874251497006, | |
| "grad_norm": 4.673940658569336, | |
| "learning_rate": 7.809013833110219e-06, | |
| "loss": 0.5117, | |
| "step": 2140 | |
| }, | |
| { | |
| "epoch": 25.748502994011975, | |
| "grad_norm": 5.361245632171631, | |
| "learning_rate": 7.5858991521642126e-06, | |
| "loss": 0.5067, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 25.868263473053894, | |
| "grad_norm": 4.769536018371582, | |
| "learning_rate": 7.362784471218207e-06, | |
| "loss": 0.5562, | |
| "step": 2160 | |
| }, | |
| { | |
| "epoch": 25.98802395209581, | |
| "grad_norm": 5.589327335357666, | |
| "learning_rate": 7.1396697902722e-06, | |
| "loss": 0.5257, | |
| "step": 2170 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_accuracy": 0.7305389221556886, | |
| "eval_loss": 0.6003913879394531, | |
| "eval_runtime": 5.6113, | |
| "eval_samples_per_second": 119.045, | |
| "eval_steps_per_second": 3.742, | |
| "step": 2171 | |
| }, | |
| { | |
| "epoch": 26.107784431137723, | |
| "grad_norm": 5.76840353012085, | |
| "learning_rate": 6.916555109326194e-06, | |
| "loss": 0.5305, | |
| "step": 2180 | |
| }, | |
| { | |
| "epoch": 26.227544910179642, | |
| "grad_norm": 4.287968158721924, | |
| "learning_rate": 6.693440428380188e-06, | |
| "loss": 0.5028, | |
| "step": 2190 | |
| }, | |
| { | |
| "epoch": 26.347305389221557, | |
| "grad_norm": 7.93202543258667, | |
| "learning_rate": 6.4703257474341815e-06, | |
| "loss": 0.5173, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 26.46706586826347, | |
| "grad_norm": 5.515824794769287, | |
| "learning_rate": 6.247211066488176e-06, | |
| "loss": 0.5111, | |
| "step": 2210 | |
| }, | |
| { | |
| "epoch": 26.58682634730539, | |
| "grad_norm": 7.206600666046143, | |
| "learning_rate": 6.024096385542169e-06, | |
| "loss": 0.5112, | |
| "step": 2220 | |
| }, | |
| { | |
| "epoch": 26.706586826347305, | |
| "grad_norm": 4.152039527893066, | |
| "learning_rate": 5.800981704596163e-06, | |
| "loss": 0.4995, | |
| "step": 2230 | |
| }, | |
| { | |
| "epoch": 26.82634730538922, | |
| "grad_norm": 5.948368072509766, | |
| "learning_rate": 5.577867023650157e-06, | |
| "loss": 0.5431, | |
| "step": 2240 | |
| }, | |
| { | |
| "epoch": 26.94610778443114, | |
| "grad_norm": 4.189924240112305, | |
| "learning_rate": 5.3547523427041504e-06, | |
| "loss": 0.5152, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 26.994011976047904, | |
| "eval_accuracy": 0.7320359281437125, | |
| "eval_loss": 0.6091659665107727, | |
| "eval_runtime": 4.8264, | |
| "eval_samples_per_second": 138.405, | |
| "eval_steps_per_second": 4.351, | |
| "step": 2254 | |
| }, | |
| { | |
| "epoch": 27.065868263473053, | |
| "grad_norm": 5.771645545959473, | |
| "learning_rate": 5.131637661758144e-06, | |
| "loss": 0.5083, | |
| "step": 2260 | |
| }, | |
| { | |
| "epoch": 27.18562874251497, | |
| "grad_norm": 5.393200397491455, | |
| "learning_rate": 4.908522980812138e-06, | |
| "loss": 0.4818, | |
| "step": 2270 | |
| }, | |
| { | |
| "epoch": 27.305389221556887, | |
| "grad_norm": 5.9255828857421875, | |
| "learning_rate": 4.685408299866132e-06, | |
| "loss": 0.5215, | |
| "step": 2280 | |
| }, | |
| { | |
| "epoch": 27.4251497005988, | |
| "grad_norm": 5.309273719787598, | |
| "learning_rate": 4.462293618920125e-06, | |
| "loss": 0.4627, | |
| "step": 2290 | |
| }, | |
| { | |
| "epoch": 27.54491017964072, | |
| "grad_norm": 6.227340221405029, | |
| "learning_rate": 4.239178937974119e-06, | |
| "loss": 0.5067, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 27.664670658682635, | |
| "grad_norm": 4.3603105545043945, | |
| "learning_rate": 4.016064257028113e-06, | |
| "loss": 0.5189, | |
| "step": 2310 | |
| }, | |
| { | |
| "epoch": 27.78443113772455, | |
| "grad_norm": 5.04020357131958, | |
| "learning_rate": 3.7929495760821063e-06, | |
| "loss": 0.5311, | |
| "step": 2320 | |
| }, | |
| { | |
| "epoch": 27.904191616766468, | |
| "grad_norm": 4.995678901672363, | |
| "learning_rate": 3.5698348951361e-06, | |
| "loss": 0.4858, | |
| "step": 2330 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.7230538922155688, | |
| "eval_loss": 0.6100460886955261, | |
| "eval_runtime": 5.2608, | |
| "eval_samples_per_second": 126.978, | |
| "eval_steps_per_second": 3.992, | |
| "step": 2338 | |
| }, | |
| { | |
| "epoch": 28.023952095808383, | |
| "grad_norm": 3.8030707836151123, | |
| "learning_rate": 3.346720214190094e-06, | |
| "loss": 0.5329, | |
| "step": 2340 | |
| }, | |
| { | |
| "epoch": 28.143712574850298, | |
| "grad_norm": 4.550446033477783, | |
| "learning_rate": 3.123605533244088e-06, | |
| "loss": 0.4682, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 28.263473053892216, | |
| "grad_norm": 4.607069492340088, | |
| "learning_rate": 2.9004908522980813e-06, | |
| "loss": 0.4928, | |
| "step": 2360 | |
| }, | |
| { | |
| "epoch": 28.38323353293413, | |
| "grad_norm": 4.880204677581787, | |
| "learning_rate": 2.6773761713520752e-06, | |
| "loss": 0.5187, | |
| "step": 2370 | |
| }, | |
| { | |
| "epoch": 28.50299401197605, | |
| "grad_norm": 4.848081111907959, | |
| "learning_rate": 2.454261490406069e-06, | |
| "loss": 0.5051, | |
| "step": 2380 | |
| }, | |
| { | |
| "epoch": 28.622754491017965, | |
| "grad_norm": 5.9451189041137695, | |
| "learning_rate": 2.2311468094600625e-06, | |
| "loss": 0.5156, | |
| "step": 2390 | |
| }, | |
| { | |
| "epoch": 28.74251497005988, | |
| "grad_norm": 5.852599143981934, | |
| "learning_rate": 2.0080321285140564e-06, | |
| "loss": 0.4913, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 28.862275449101798, | |
| "grad_norm": 5.628122329711914, | |
| "learning_rate": 1.78491744756805e-06, | |
| "loss": 0.4711, | |
| "step": 2410 | |
| }, | |
| { | |
| "epoch": 28.982035928143713, | |
| "grad_norm": 5.2325358390808105, | |
| "learning_rate": 1.561802766622044e-06, | |
| "loss": 0.5412, | |
| "step": 2420 | |
| }, | |
| { | |
| "epoch": 28.994011976047904, | |
| "eval_accuracy": 0.7350299401197605, | |
| "eval_loss": 0.6115620136260986, | |
| "eval_runtime": 5.4339, | |
| "eval_samples_per_second": 122.932, | |
| "eval_steps_per_second": 3.865, | |
| "step": 2421 | |
| }, | |
| { | |
| "epoch": 29.101796407185628, | |
| "grad_norm": 5.715780258178711, | |
| "learning_rate": 1.3386880856760376e-06, | |
| "loss": 0.528, | |
| "step": 2430 | |
| }, | |
| { | |
| "epoch": 29.221556886227546, | |
| "grad_norm": 5.140947341918945, | |
| "learning_rate": 1.1155734047300313e-06, | |
| "loss": 0.5015, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 29.34131736526946, | |
| "grad_norm": 5.15585470199585, | |
| "learning_rate": 8.92458723784025e-07, | |
| "loss": 0.4934, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 29.461077844311376, | |
| "grad_norm": 3.845532178878784, | |
| "learning_rate": 6.693440428380188e-07, | |
| "loss": 0.4809, | |
| "step": 2460 | |
| }, | |
| { | |
| "epoch": 29.580838323353294, | |
| "grad_norm": 5.481026649475098, | |
| "learning_rate": 4.462293618920125e-07, | |
| "loss": 0.4645, | |
| "step": 2470 | |
| }, | |
| { | |
| "epoch": 29.70059880239521, | |
| "grad_norm": 4.100111484527588, | |
| "learning_rate": 2.2311468094600626e-07, | |
| "loss": 0.4856, | |
| "step": 2480 | |
| }, | |
| { | |
| "epoch": 29.820359281437124, | |
| "grad_norm": 5.421600818634033, | |
| "learning_rate": 0.0, | |
| "loss": 0.4972, | |
| "step": 2490 | |
| }, | |
| { | |
| "epoch": 29.820359281437124, | |
| "eval_accuracy": 0.7290419161676647, | |
| "eval_loss": 0.6119701862335205, | |
| "eval_runtime": 4.8411, | |
| "eval_samples_per_second": 137.985, | |
| "eval_steps_per_second": 4.338, | |
| "step": 2490 | |
| }, | |
| { | |
| "epoch": 29.820359281437124, | |
| "step": 2490, | |
| "total_flos": 7.910788670992908e+18, | |
| "train_loss": 0.5937201630159554, | |
| "train_runtime": 3590.0203, | |
| "train_samples_per_second": 89.181, | |
| "train_steps_per_second": 0.694 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 2490, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 30, | |
| "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": 7.910788670992908e+18, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |