| { | |
| "best_metric": 0.9948348999023438, | |
| "best_model_checkpoint": "xblock-social-screenshots-2/checkpoint-1737", | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 1737, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 12.944127082824707, | |
| "learning_rate": 7.4712643678160925e-06, | |
| "loss": 2.5595, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 10.593692779541016, | |
| "learning_rate": 1.5517241379310346e-05, | |
| "loss": 2.1472, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 9.828369140625, | |
| "learning_rate": 2.327586206896552e-05, | |
| "loss": 1.7244, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "grad_norm": 8.050122261047363, | |
| "learning_rate": 3.132183908045977e-05, | |
| "loss": 1.5718, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 7.566161155700684, | |
| "learning_rate": 3.936781609195402e-05, | |
| "loss": 1.4015, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "grad_norm": 8.954880714416504, | |
| "learning_rate": 4.741379310344828e-05, | |
| "loss": 1.3705, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "grad_norm": 8.316436767578125, | |
| "learning_rate": 4.9392194497760715e-05, | |
| "loss": 1.3727, | |
| "step": 196 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "grad_norm": 4.580837726593018, | |
| "learning_rate": 4.8496481126039675e-05, | |
| "loss": 1.2824, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "grad_norm": 12.312744140625, | |
| "learning_rate": 4.760076775431862e-05, | |
| "loss": 1.3119, | |
| "step": 252 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 10.076079368591309, | |
| "learning_rate": 4.670505438259757e-05, | |
| "loss": 1.252, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "grad_norm": 9.073230743408203, | |
| "learning_rate": 4.580934101087652e-05, | |
| "loss": 1.3154, | |
| "step": 308 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "grad_norm": 9.829570770263672, | |
| "learning_rate": 4.491362763915547e-05, | |
| "loss": 1.3688, | |
| "step": 336 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "grad_norm": 7.165198802947998, | |
| "learning_rate": 4.4017914267434426e-05, | |
| "loss": 1.4376, | |
| "step": 364 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "grad_norm": 8.911142349243164, | |
| "learning_rate": 4.312220089571337e-05, | |
| "loss": 1.3533, | |
| "step": 392 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "grad_norm": 10.684996604919434, | |
| "learning_rate": 4.2226487523992326e-05, | |
| "loss": 1.2825, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "grad_norm": 10.35905647277832, | |
| "learning_rate": 4.133077415227127e-05, | |
| "loss": 1.3382, | |
| "step": 448 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "grad_norm": 9.137449264526367, | |
| "learning_rate": 4.0435060780550225e-05, | |
| "loss": 1.3054, | |
| "step": 476 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "grad_norm": 6.174429416656494, | |
| "learning_rate": 3.953934740882918e-05, | |
| "loss": 1.4294, | |
| "step": 504 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "grad_norm": 9.742380142211914, | |
| "learning_rate": 3.864363403710813e-05, | |
| "loss": 1.2502, | |
| "step": 532 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "grad_norm": 7.022644519805908, | |
| "learning_rate": 3.774792066538708e-05, | |
| "loss": 1.2926, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5591026747195859, | |
| "eval_f1_macro": 0.17116779008569874, | |
| "eval_f1_micro": 0.5591026747195859, | |
| "eval_f1_weighted": 0.47555675422704724, | |
| "eval_loss": 1.1608415842056274, | |
| "eval_precision_macro": 0.17146675898922856, | |
| "eval_precision_micro": 0.5591026747195859, | |
| "eval_precision_weighted": 0.4635728502463956, | |
| "eval_recall_macro": 0.1928392780592314, | |
| "eval_recall_micro": 0.5591026747195859, | |
| "eval_recall_weighted": 0.5591026747195859, | |
| "eval_runtime": 47.359, | |
| "eval_samples_per_second": 24.473, | |
| "eval_steps_per_second": 1.541, | |
| "step": 579 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "grad_norm": 7.127846717834473, | |
| "learning_rate": 3.685220729366603e-05, | |
| "loss": 1.161, | |
| "step": 588 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "grad_norm": 6.150296211242676, | |
| "learning_rate": 3.5956493921944976e-05, | |
| "loss": 1.1594, | |
| "step": 616 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "grad_norm": 6.171022415161133, | |
| "learning_rate": 3.506078055022393e-05, | |
| "loss": 1.1253, | |
| "step": 644 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "grad_norm": 12.212858200073242, | |
| "learning_rate": 3.416506717850288e-05, | |
| "loss": 1.2149, | |
| "step": 672 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "grad_norm": 8.228123664855957, | |
| "learning_rate": 3.3269353806781835e-05, | |
| "loss": 1.1905, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "grad_norm": 6.095876693725586, | |
| "learning_rate": 3.237364043506078e-05, | |
| "loss": 1.1387, | |
| "step": 728 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "grad_norm": 10.43623161315918, | |
| "learning_rate": 3.1477927063339734e-05, | |
| "loss": 1.2243, | |
| "step": 756 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "grad_norm": 6.94878625869751, | |
| "learning_rate": 3.058221369161868e-05, | |
| "loss": 1.0566, | |
| "step": 784 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 10.651735305786133, | |
| "learning_rate": 2.9686500319897637e-05, | |
| "loss": 1.1425, | |
| "step": 812 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "grad_norm": 8.426492691040039, | |
| "learning_rate": 2.8790786948176586e-05, | |
| "loss": 1.173, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "grad_norm": 7.550307273864746, | |
| "learning_rate": 2.789507357645554e-05, | |
| "loss": 1.2052, | |
| "step": 868 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "grad_norm": 9.533821105957031, | |
| "learning_rate": 2.6999360204734485e-05, | |
| "loss": 1.2043, | |
| "step": 896 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 7.776467323303223, | |
| "learning_rate": 2.6103646833013435e-05, | |
| "loss": 1.0199, | |
| "step": 924 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "grad_norm": 9.101223945617676, | |
| "learning_rate": 2.5207933461292388e-05, | |
| "loss": 1.1681, | |
| "step": 952 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "grad_norm": 7.713497161865234, | |
| "learning_rate": 2.4312220089571338e-05, | |
| "loss": 1.1177, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "grad_norm": 7.732503414154053, | |
| "learning_rate": 2.341650671785029e-05, | |
| "loss": 1.1099, | |
| "step": 1008 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "grad_norm": 9.252326965332031, | |
| "learning_rate": 2.252079334612924e-05, | |
| "loss": 1.2187, | |
| "step": 1036 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "grad_norm": 7.301546096801758, | |
| "learning_rate": 2.162507997440819e-05, | |
| "loss": 1.0414, | |
| "step": 1064 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "grad_norm": 8.406270027160645, | |
| "learning_rate": 2.0729366602687143e-05, | |
| "loss": 1.0479, | |
| "step": 1092 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "grad_norm": 7.392611980438232, | |
| "learning_rate": 1.9833653230966092e-05, | |
| "loss": 1.2113, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "grad_norm": 7.991569995880127, | |
| "learning_rate": 1.8937939859245045e-05, | |
| "loss": 1.1279, | |
| "step": 1148 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.548748921484038, | |
| "eval_f1_macro": 0.23354766712716346, | |
| "eval_f1_micro": 0.548748921484038, | |
| "eval_f1_weighted": 0.5206705245021934, | |
| "eval_loss": 1.076399326324463, | |
| "eval_precision_macro": 0.2591242803710448, | |
| "eval_precision_micro": 0.548748921484038, | |
| "eval_precision_weighted": 0.5086032820438354, | |
| "eval_recall_macro": 0.2392300811955355, | |
| "eval_recall_micro": 0.548748921484038, | |
| "eval_recall_weighted": 0.548748921484038, | |
| "eval_runtime": 45.7915, | |
| "eval_samples_per_second": 25.31, | |
| "eval_steps_per_second": 1.594, | |
| "step": 1158 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "grad_norm": 8.247747421264648, | |
| "learning_rate": 1.8042226487523995e-05, | |
| "loss": 1.0308, | |
| "step": 1176 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "grad_norm": 5.560178279876709, | |
| "learning_rate": 1.7146513115802944e-05, | |
| "loss": 1.0764, | |
| "step": 1204 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "grad_norm": 6.093762397766113, | |
| "learning_rate": 1.6250799744081894e-05, | |
| "loss": 1.1365, | |
| "step": 1232 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "grad_norm": 4.060959815979004, | |
| "learning_rate": 1.5355086372360844e-05, | |
| "loss": 1.0535, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "grad_norm": 7.9812822341918945, | |
| "learning_rate": 1.4459373000639795e-05, | |
| "loss": 1.0241, | |
| "step": 1288 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "grad_norm": 5.996217250823975, | |
| "learning_rate": 1.3563659628918746e-05, | |
| "loss": 1.0137, | |
| "step": 1316 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "grad_norm": 10.663773536682129, | |
| "learning_rate": 1.2667946257197696e-05, | |
| "loss": 1.0274, | |
| "step": 1344 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "grad_norm": 9.829933166503906, | |
| "learning_rate": 1.1772232885476649e-05, | |
| "loss": 0.9893, | |
| "step": 1372 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "grad_norm": 9.35245418548584, | |
| "learning_rate": 1.0876519513755598e-05, | |
| "loss": 1.0497, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "grad_norm": 9.116876602172852, | |
| "learning_rate": 9.98080614203455e-06, | |
| "loss": 0.972, | |
| "step": 1428 | |
| }, | |
| { | |
| "epoch": 2.51, | |
| "grad_norm": 8.650375366210938, | |
| "learning_rate": 9.085092770313499e-06, | |
| "loss": 1.0058, | |
| "step": 1456 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "grad_norm": 9.76453971862793, | |
| "learning_rate": 8.18937939859245e-06, | |
| "loss": 1.0451, | |
| "step": 1484 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "grad_norm": 5.405215263366699, | |
| "learning_rate": 7.293666026871402e-06, | |
| "loss": 0.9628, | |
| "step": 1512 | |
| }, | |
| { | |
| "epoch": 2.66, | |
| "grad_norm": 10.716626167297363, | |
| "learning_rate": 6.397952655150352e-06, | |
| "loss": 1.1384, | |
| "step": 1540 | |
| }, | |
| { | |
| "epoch": 2.71, | |
| "grad_norm": 4.47056770324707, | |
| "learning_rate": 5.502239283429303e-06, | |
| "loss": 0.9374, | |
| "step": 1568 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "grad_norm": 7.823177814483643, | |
| "learning_rate": 4.606525911708254e-06, | |
| "loss": 1.019, | |
| "step": 1596 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "grad_norm": 10.679405212402344, | |
| "learning_rate": 3.7108125399872046e-06, | |
| "loss": 1.0615, | |
| "step": 1624 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "grad_norm": 16.2053165435791, | |
| "learning_rate": 2.815099168266155e-06, | |
| "loss": 0.959, | |
| "step": 1652 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "grad_norm": 8.753532409667969, | |
| "learning_rate": 1.9193857965451054e-06, | |
| "loss": 1.0041, | |
| "step": 1680 | |
| }, | |
| { | |
| "epoch": 2.95, | |
| "grad_norm": 7.493074417114258, | |
| "learning_rate": 1.0236724248240563e-06, | |
| "loss": 0.998, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 7.273109436035156, | |
| "learning_rate": 1.2795905310300704e-07, | |
| "loss": 0.9851, | |
| "step": 1736 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.5823986194995686, | |
| "eval_f1_macro": 0.23526955685803672, | |
| "eval_f1_micro": 0.5823986194995686, | |
| "eval_f1_weighted": 0.5233223208450835, | |
| "eval_loss": 0.9948348999023438, | |
| "eval_precision_macro": 0.2973178657776674, | |
| "eval_precision_micro": 0.5823986194995686, | |
| "eval_precision_weighted": 0.5462756602202016, | |
| "eval_recall_macro": 0.24224828041633692, | |
| "eval_recall_micro": 0.5823986194995686, | |
| "eval_recall_weighted": 0.5823986194995686, | |
| "eval_runtime": 46.2253, | |
| "eval_samples_per_second": 25.073, | |
| "eval_steps_per_second": 1.579, | |
| "step": 1737 | |
| } | |
| ], | |
| "logging_steps": 28, | |
| "max_steps": 1737, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 1.0764602068237517e+18, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |