| { | |
| "best_metric": 0.37855324149131775, | |
| "best_model_checkpoint": "xblock-large-patch3-224/checkpoint-1872", | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 1872, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.04006410256410257, | |
| "grad_norm": 8.251535415649414, | |
| "learning_rate": 6.3829787234042555e-06, | |
| "loss": 1.3524, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.08012820512820513, | |
| "grad_norm": 12.843944549560547, | |
| "learning_rate": 1.3031914893617023e-05, | |
| "loss": 0.8737, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.1201923076923077, | |
| "grad_norm": 15.241576194763184, | |
| "learning_rate": 1.9414893617021276e-05, | |
| "loss": 0.6916, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.16025641025641027, | |
| "grad_norm": 6.287885665893555, | |
| "learning_rate": 2.6063829787234046e-05, | |
| "loss": 0.8028, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.20032051282051283, | |
| "grad_norm": 4.597654819488525, | |
| "learning_rate": 3.271276595744681e-05, | |
| "loss": 0.7592, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.2403846153846154, | |
| "grad_norm": 3.4197440147399902, | |
| "learning_rate": 3.936170212765958e-05, | |
| "loss": 0.8431, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.28044871794871795, | |
| "grad_norm": 7.936722755432129, | |
| "learning_rate": 4.601063829787234e-05, | |
| "loss": 0.6737, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.32051282051282054, | |
| "grad_norm": 27.45574188232422, | |
| "learning_rate": 4.970308788598575e-05, | |
| "loss": 0.9851, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.3605769230769231, | |
| "grad_norm": 7.184787273406982, | |
| "learning_rate": 4.896080760095012e-05, | |
| "loss": 0.9612, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.40064102564102566, | |
| "grad_norm": 5.246083736419678, | |
| "learning_rate": 4.821852731591449e-05, | |
| "loss": 0.7614, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.4407051282051282, | |
| "grad_norm": 14.747861862182617, | |
| "learning_rate": 4.7476247030878864e-05, | |
| "loss": 0.9273, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.4807692307692308, | |
| "grad_norm": 10.44289779663086, | |
| "learning_rate": 4.673396674584323e-05, | |
| "loss": 0.8679, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.5208333333333334, | |
| "grad_norm": 11.92969799041748, | |
| "learning_rate": 4.5991686460807604e-05, | |
| "loss": 0.8816, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 0.5608974358974359, | |
| "grad_norm": 8.216315269470215, | |
| "learning_rate": 4.524940617577197e-05, | |
| "loss": 0.8063, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.6009615384615384, | |
| "grad_norm": 9.245816230773926, | |
| "learning_rate": 4.4507125890736345e-05, | |
| "loss": 0.8215, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 0.6410256410256411, | |
| "grad_norm": 8.622091293334961, | |
| "learning_rate": 4.376484560570072e-05, | |
| "loss": 0.9864, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.6810897435897436, | |
| "grad_norm": 4.5337419509887695, | |
| "learning_rate": 4.3022565320665086e-05, | |
| "loss": 0.6767, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 0.7211538461538461, | |
| "grad_norm": 6.817728519439697, | |
| "learning_rate": 4.228028503562946e-05, | |
| "loss": 0.8387, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.7612179487179487, | |
| "grad_norm": 10.498202323913574, | |
| "learning_rate": 4.153800475059383e-05, | |
| "loss": 0.7565, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 0.8012820512820513, | |
| "grad_norm": 13.513296127319336, | |
| "learning_rate": 4.07957244655582e-05, | |
| "loss": 0.6909, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.8413461538461539, | |
| "grad_norm": 13.574014663696289, | |
| "learning_rate": 4.005344418052257e-05, | |
| "loss": 0.757, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 0.8814102564102564, | |
| "grad_norm": 6.092533588409424, | |
| "learning_rate": 3.9311163895486934e-05, | |
| "loss": 0.8626, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.9214743589743589, | |
| "grad_norm": 14.241544723510742, | |
| "learning_rate": 3.856888361045131e-05, | |
| "loss": 0.6826, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 0.9615384615384616, | |
| "grad_norm": 4.413307189941406, | |
| "learning_rate": 3.7826603325415675e-05, | |
| "loss": 0.7986, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.844551282051282, | |
| "eval_f1_macro": 0.8160973181841822, | |
| "eval_f1_micro": 0.844551282051282, | |
| "eval_f1_weighted": 0.8402659980370851, | |
| "eval_loss": 0.44926634430885315, | |
| "eval_precision_macro": 0.8683442897703473, | |
| "eval_precision_micro": 0.844551282051282, | |
| "eval_precision_weighted": 0.8585860519825896, | |
| "eval_recall_macro": 0.7986028134342184, | |
| "eval_recall_micro": 0.844551282051282, | |
| "eval_recall_weighted": 0.844551282051282, | |
| "eval_runtime": 1324.3117, | |
| "eval_samples_per_second": 0.942, | |
| "eval_steps_per_second": 0.059, | |
| "step": 624 | |
| }, | |
| { | |
| "epoch": 1.001602564102564, | |
| "grad_norm": 4.381113052368164, | |
| "learning_rate": 3.708432304038005e-05, | |
| "loss": 0.861, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 1.0416666666666667, | |
| "grad_norm": 5.680240631103516, | |
| "learning_rate": 3.6342042755344416e-05, | |
| "loss": 0.6535, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.0817307692307692, | |
| "grad_norm": 5.435142993927002, | |
| "learning_rate": 3.559976247030879e-05, | |
| "loss": 0.7551, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 1.1217948717948718, | |
| "grad_norm": 18.080663681030273, | |
| "learning_rate": 3.4857482185273164e-05, | |
| "loss": 0.7618, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.1618589743589745, | |
| "grad_norm": 12.246950149536133, | |
| "learning_rate": 3.411520190023753e-05, | |
| "loss": 0.7952, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 1.2019230769230769, | |
| "grad_norm": 9.545283317565918, | |
| "learning_rate": 3.3372921615201904e-05, | |
| "loss": 0.6002, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.2419871794871795, | |
| "grad_norm": 8.819951057434082, | |
| "learning_rate": 3.263064133016627e-05, | |
| "loss": 0.7556, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 1.282051282051282, | |
| "grad_norm": 10.624255180358887, | |
| "learning_rate": 3.1888361045130645e-05, | |
| "loss": 0.7373, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.3221153846153846, | |
| "grad_norm": 7.088049411773682, | |
| "learning_rate": 3.114608076009501e-05, | |
| "loss": 0.6166, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 1.3621794871794872, | |
| "grad_norm": 7.568330764770508, | |
| "learning_rate": 3.0403800475059386e-05, | |
| "loss": 0.6451, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 1.4022435897435899, | |
| "grad_norm": 6.381913185119629, | |
| "learning_rate": 2.9661520190023756e-05, | |
| "loss": 0.6397, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 1.4423076923076923, | |
| "grad_norm": 9.472349166870117, | |
| "learning_rate": 2.8919239904988127e-05, | |
| "loss": 0.6227, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.482371794871795, | |
| "grad_norm": 4.770060062408447, | |
| "learning_rate": 2.8176959619952497e-05, | |
| "loss": 0.6374, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 1.5224358974358974, | |
| "grad_norm": 6.0947771072387695, | |
| "learning_rate": 2.7434679334916867e-05, | |
| "loss": 0.5955, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 1.5625, | |
| "grad_norm": 5.420393466949463, | |
| "learning_rate": 2.6692399049881234e-05, | |
| "loss": 0.7621, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 1.6025641025641026, | |
| "grad_norm": 8.125214576721191, | |
| "learning_rate": 2.5950118764845605e-05, | |
| "loss": 0.624, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.6426282051282053, | |
| "grad_norm": 11.498839378356934, | |
| "learning_rate": 2.5207838479809975e-05, | |
| "loss": 0.6889, | |
| "step": 1025 | |
| }, | |
| { | |
| "epoch": 1.6826923076923077, | |
| "grad_norm": 6.599166393280029, | |
| "learning_rate": 2.446555819477435e-05, | |
| "loss": 0.5743, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 1.7227564102564101, | |
| "grad_norm": 18.760013580322266, | |
| "learning_rate": 2.372327790973872e-05, | |
| "loss": 0.6392, | |
| "step": 1075 | |
| }, | |
| { | |
| "epoch": 1.7628205128205128, | |
| "grad_norm": 10.053047180175781, | |
| "learning_rate": 2.2980997624703086e-05, | |
| "loss": 0.7208, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.8028846153846154, | |
| "grad_norm": 6.171570777893066, | |
| "learning_rate": 2.223871733966746e-05, | |
| "loss": 0.5817, | |
| "step": 1125 | |
| }, | |
| { | |
| "epoch": 1.842948717948718, | |
| "grad_norm": 5.438927173614502, | |
| "learning_rate": 2.149643705463183e-05, | |
| "loss": 0.4576, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 1.8830128205128205, | |
| "grad_norm": 8.946006774902344, | |
| "learning_rate": 2.07541567695962e-05, | |
| "loss": 0.695, | |
| "step": 1175 | |
| }, | |
| { | |
| "epoch": 1.9230769230769231, | |
| "grad_norm": 7.02325439453125, | |
| "learning_rate": 2.001187648456057e-05, | |
| "loss": 0.6768, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.9631410256410255, | |
| "grad_norm": 4.542496204376221, | |
| "learning_rate": 1.926959619952494e-05, | |
| "loss": 0.6592, | |
| "step": 1225 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8717948717948718, | |
| "eval_f1_macro": 0.8420194226887088, | |
| "eval_f1_micro": 0.8717948717948718, | |
| "eval_f1_weighted": 0.8664623444122613, | |
| "eval_loss": 0.39590924978256226, | |
| "eval_precision_macro": 0.8724657298009468, | |
| "eval_precision_micro": 0.8717948717948718, | |
| "eval_precision_weighted": 0.8825043036411662, | |
| "eval_recall_macro": 0.8395777006090396, | |
| "eval_recall_micro": 0.8717948717948718, | |
| "eval_recall_weighted": 0.8717948717948718, | |
| "eval_runtime": 1293.5897, | |
| "eval_samples_per_second": 0.965, | |
| "eval_steps_per_second": 0.06, | |
| "step": 1248 | |
| }, | |
| { | |
| "epoch": 2.003205128205128, | |
| "grad_norm": 7.841921329498291, | |
| "learning_rate": 1.8527315914489312e-05, | |
| "loss": 0.6025, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 2.043269230769231, | |
| "grad_norm": 9.250267028808594, | |
| "learning_rate": 1.7785035629453682e-05, | |
| "loss": 0.4597, | |
| "step": 1275 | |
| }, | |
| { | |
| "epoch": 2.0833333333333335, | |
| "grad_norm": 3.1655519008636475, | |
| "learning_rate": 1.7042755344418056e-05, | |
| "loss": 0.3962, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 2.123397435897436, | |
| "grad_norm": 5.661818981170654, | |
| "learning_rate": 1.6300475059382423e-05, | |
| "loss": 0.6639, | |
| "step": 1325 | |
| }, | |
| { | |
| "epoch": 2.1634615384615383, | |
| "grad_norm": 3.282247304916382, | |
| "learning_rate": 1.5558194774346793e-05, | |
| "loss": 0.464, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 2.203525641025641, | |
| "grad_norm": 3.829854965209961, | |
| "learning_rate": 1.4815914489311164e-05, | |
| "loss": 0.4943, | |
| "step": 1375 | |
| }, | |
| { | |
| "epoch": 2.2435897435897436, | |
| "grad_norm": 5.670718193054199, | |
| "learning_rate": 1.4073634204275534e-05, | |
| "loss": 0.6288, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.2836538461538463, | |
| "grad_norm": 10.35056209564209, | |
| "learning_rate": 1.3331353919239906e-05, | |
| "loss": 0.4846, | |
| "step": 1425 | |
| }, | |
| { | |
| "epoch": 2.323717948717949, | |
| "grad_norm": 0.6631863117218018, | |
| "learning_rate": 1.2589073634204277e-05, | |
| "loss": 0.3622, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 2.363782051282051, | |
| "grad_norm": 6.619789123535156, | |
| "learning_rate": 1.1846793349168647e-05, | |
| "loss": 0.5013, | |
| "step": 1475 | |
| }, | |
| { | |
| "epoch": 2.4038461538461537, | |
| "grad_norm": 6.15729284286499, | |
| "learning_rate": 1.1104513064133017e-05, | |
| "loss": 0.4955, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.4439102564102564, | |
| "grad_norm": 0.05698273330926895, | |
| "learning_rate": 1.0362232779097388e-05, | |
| "loss": 0.4693, | |
| "step": 1525 | |
| }, | |
| { | |
| "epoch": 2.483974358974359, | |
| "grad_norm": 3.433866262435913, | |
| "learning_rate": 9.619952494061758e-06, | |
| "loss": 0.4221, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 2.5240384615384617, | |
| "grad_norm": 6.545916557312012, | |
| "learning_rate": 8.877672209026128e-06, | |
| "loss": 0.6492, | |
| "step": 1575 | |
| }, | |
| { | |
| "epoch": 2.564102564102564, | |
| "grad_norm": 12.447614669799805, | |
| "learning_rate": 8.135391923990499e-06, | |
| "loss": 0.4888, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.6041666666666665, | |
| "grad_norm": 6.291492462158203, | |
| "learning_rate": 7.393111638954869e-06, | |
| "loss": 0.4994, | |
| "step": 1625 | |
| }, | |
| { | |
| "epoch": 2.644230769230769, | |
| "grad_norm": 5.843363285064697, | |
| "learning_rate": 6.6508313539192404e-06, | |
| "loss": 0.6511, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 2.684294871794872, | |
| "grad_norm": 6.965985298156738, | |
| "learning_rate": 5.908551068883611e-06, | |
| "loss": 0.4697, | |
| "step": 1675 | |
| }, | |
| { | |
| "epoch": 2.7243589743589745, | |
| "grad_norm": 7.191352367401123, | |
| "learning_rate": 5.166270783847981e-06, | |
| "loss": 0.4224, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 2.7644230769230766, | |
| "grad_norm": 14.876649856567383, | |
| "learning_rate": 4.4239904988123516e-06, | |
| "loss": 0.6627, | |
| "step": 1725 | |
| }, | |
| { | |
| "epoch": 2.8044871794871797, | |
| "grad_norm": 6.894955158233643, | |
| "learning_rate": 3.681710213776722e-06, | |
| "loss": 0.5652, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 2.844551282051282, | |
| "grad_norm": 5.6652984619140625, | |
| "learning_rate": 2.9394299287410927e-06, | |
| "loss": 0.6411, | |
| "step": 1775 | |
| }, | |
| { | |
| "epoch": 2.8846153846153846, | |
| "grad_norm": 8.991388320922852, | |
| "learning_rate": 2.197149643705463e-06, | |
| "loss": 0.4483, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 2.9246794871794872, | |
| "grad_norm": 8.376466751098633, | |
| "learning_rate": 1.4548693586698337e-06, | |
| "loss": 0.4551, | |
| "step": 1825 | |
| }, | |
| { | |
| "epoch": 2.96474358974359, | |
| "grad_norm": 3.4173240661621094, | |
| "learning_rate": 7.422802850356294e-07, | |
| "loss": 0.4227, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8886217948717948, | |
| "eval_f1_macro": 0.86094260720702, | |
| "eval_f1_micro": 0.8886217948717948, | |
| "eval_f1_weighted": 0.883196165156119, | |
| "eval_loss": 0.37855324149131775, | |
| "eval_precision_macro": 0.8961444617693151, | |
| "eval_precision_micro": 0.8886217948717948, | |
| "eval_precision_weighted": 0.8922651559280282, | |
| "eval_recall_macro": 0.8524486181675118, | |
| "eval_recall_micro": 0.8886217948717948, | |
| "eval_recall_weighted": 0.8886217948717948, | |
| "eval_runtime": 1323.0713, | |
| "eval_samples_per_second": 0.943, | |
| "eval_steps_per_second": 0.059, | |
| "step": 1872 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 1872, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 4.1009193984929587e+18, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |