| { | |
| "best_metric": 0.7289302945137024, | |
| "best_model_checkpoint": "/output/zgt-longformer-base-4096-finetuned-Gu21schedule-BS256-10ep-unrolled-13772/checkpoint-31926", | |
| "epoch": 9.999417758369724, | |
| "global_step": 32200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.1, | |
| "eval_accuracy": 0.15377753335327157, | |
| "eval_loss": 6.225821018218994, | |
| "eval_runtime": 2282.6214, | |
| "eval_samples_per_second": 45.205, | |
| "eval_steps_per_second": 2.826, | |
| "step": 313 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 9.316770186335403e-05, | |
| "loss": 7.0555, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "eval_accuracy": 0.17207746116906447, | |
| "eval_loss": 5.971940994262695, | |
| "eval_runtime": 2280.3592, | |
| "eval_samples_per_second": 45.25, | |
| "eval_steps_per_second": 2.829, | |
| "step": 626 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "eval_accuracy": 0.2249201960138123, | |
| "eval_loss": 5.275129318237305, | |
| "eval_runtime": 2280.8737, | |
| "eval_samples_per_second": 45.24, | |
| "eval_steps_per_second": 2.828, | |
| "step": 939 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 0.00018633540372670805, | |
| "loss": 5.7705, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "eval_accuracy": 0.47743625094670755, | |
| "eval_loss": 3.3030917644500732, | |
| "eval_runtime": 2281.1425, | |
| "eval_samples_per_second": 45.235, | |
| "eval_steps_per_second": 2.828, | |
| "step": 1252 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 0.0002795031055900621, | |
| "loss": 3.5618, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "eval_accuracy": 0.6006654298264571, | |
| "eval_loss": 2.2308430671691895, | |
| "eval_runtime": 2282.0168, | |
| "eval_samples_per_second": 45.217, | |
| "eval_steps_per_second": 2.826, | |
| "step": 1565 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "eval_accuracy": 0.6511583497642377, | |
| "eval_loss": 1.8314521312713623, | |
| "eval_runtime": 2288.2913, | |
| "eval_samples_per_second": 45.093, | |
| "eval_steps_per_second": 2.819, | |
| "step": 1878 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 0.0003726708074534161, | |
| "loss": 2.0809, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "eval_accuracy": 0.6766689599727895, | |
| "eval_loss": 1.6349689960479736, | |
| "eval_runtime": 2284.6918, | |
| "eval_samples_per_second": 45.165, | |
| "eval_steps_per_second": 2.823, | |
| "step": 2191 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 0.0004658385093167702, | |
| "loss": 1.6957, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "eval_accuracy": 0.6925903806215541, | |
| "eval_loss": 1.5208783149719238, | |
| "eval_runtime": 2285.4871, | |
| "eval_samples_per_second": 45.149, | |
| "eval_steps_per_second": 2.822, | |
| "step": 2504 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "eval_accuracy": 0.7061222686146594, | |
| "eval_loss": 1.4304304122924805, | |
| "eval_runtime": 2286.3274, | |
| "eval_samples_per_second": 45.132, | |
| "eval_steps_per_second": 2.821, | |
| "step": 2817 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 0.0005590062111801242, | |
| "loss": 1.5265, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "eval_accuracy": 0.7137086705541928, | |
| "eval_loss": 1.3776779174804688, | |
| "eval_runtime": 2285.165, | |
| "eval_samples_per_second": 45.155, | |
| "eval_steps_per_second": 2.823, | |
| "step": 3130 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "eval_accuracy": 0.7221819099152214, | |
| "eval_loss": 1.3269994258880615, | |
| "eval_runtime": 2284.944, | |
| "eval_samples_per_second": 45.16, | |
| "eval_steps_per_second": 2.823, | |
| "step": 3443 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 0.0005942028985507245, | |
| "loss": 1.4318, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "eval_accuracy": 0.7294565866257247, | |
| "eval_loss": 1.2796943187713623, | |
| "eval_runtime": 2284.7192, | |
| "eval_samples_per_second": 45.164, | |
| "eval_steps_per_second": 2.823, | |
| "step": 3756 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 0.0005838509316770186, | |
| "loss": 1.3525, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "eval_accuracy": 0.7355410868863292, | |
| "eval_loss": 1.2408177852630615, | |
| "eval_runtime": 2284.293, | |
| "eval_samples_per_second": 45.172, | |
| "eval_steps_per_second": 2.824, | |
| "step": 4069 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "eval_accuracy": 0.7400509934425414, | |
| "eval_loss": 1.2121047973632812, | |
| "eval_runtime": 2284.0943, | |
| "eval_samples_per_second": 45.176, | |
| "eval_steps_per_second": 2.824, | |
| "step": 4382 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 0.0005734989648033125, | |
| "loss": 1.2955, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "eval_accuracy": 0.7442083806959977, | |
| "eval_loss": 1.1882882118225098, | |
| "eval_runtime": 2287.2324, | |
| "eval_samples_per_second": 45.114, | |
| "eval_steps_per_second": 2.82, | |
| "step": 4695 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "learning_rate": 0.0005631469979296066, | |
| "loss": 1.2534, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "eval_accuracy": 0.7477817783646259, | |
| "eval_loss": 1.1666710376739502, | |
| "eval_runtime": 2287.2927, | |
| "eval_samples_per_second": 45.113, | |
| "eval_steps_per_second": 2.82, | |
| "step": 5008 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "eval_accuracy": 0.7509729759994749, | |
| "eval_loss": 1.1466068029403687, | |
| "eval_runtime": 2286.8281, | |
| "eval_samples_per_second": 45.122, | |
| "eval_steps_per_second": 2.821, | |
| "step": 5321 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "learning_rate": 0.0005527950310559006, | |
| "loss": 1.2177, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "eval_accuracy": 0.7543796916990622, | |
| "eval_loss": 1.1255745887756348, | |
| "eval_runtime": 2287.311, | |
| "eval_samples_per_second": 45.113, | |
| "eval_steps_per_second": 2.82, | |
| "step": 5634 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "eval_accuracy": 0.7573199604626228, | |
| "eval_loss": 1.1089411973953247, | |
| "eval_runtime": 2287.9585, | |
| "eval_samples_per_second": 45.1, | |
| "eval_steps_per_second": 2.819, | |
| "step": 5947 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "learning_rate": 0.0005424430641821946, | |
| "loss": 1.1842, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "eval_accuracy": 0.7602460251356375, | |
| "eval_loss": 1.0928661823272705, | |
| "eval_runtime": 2287.121, | |
| "eval_samples_per_second": 45.117, | |
| "eval_steps_per_second": 2.82, | |
| "step": 6260 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 0.0005320910973084886, | |
| "loss": 1.1621, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "eval_accuracy": 0.7627146220262728, | |
| "eval_loss": 1.0785621404647827, | |
| "eval_runtime": 2287.2838, | |
| "eval_samples_per_second": 45.113, | |
| "eval_steps_per_second": 2.82, | |
| "step": 6573 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "eval_accuracy": 0.7646867902540654, | |
| "eval_loss": 1.0680384635925293, | |
| "eval_runtime": 2286.3444, | |
| "eval_samples_per_second": 45.132, | |
| "eval_steps_per_second": 2.821, | |
| "step": 6886 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "learning_rate": 0.0005217391304347826, | |
| "loss": 1.135, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "eval_accuracy": 0.7667706269089923, | |
| "eval_loss": 1.055375099182129, | |
| "eval_runtime": 2286.5038, | |
| "eval_samples_per_second": 45.129, | |
| "eval_steps_per_second": 2.821, | |
| "step": 7199 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "learning_rate": 0.0005113871635610765, | |
| "loss": 1.1177, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "eval_accuracy": 0.7681812681198601, | |
| "eval_loss": 1.0453423261642456, | |
| "eval_runtime": 2287.1445, | |
| "eval_samples_per_second": 45.116, | |
| "eval_steps_per_second": 2.82, | |
| "step": 7512 | |
| }, | |
| { | |
| "epoch": 2.43, | |
| "eval_accuracy": 0.7700268668327276, | |
| "eval_loss": 1.034972071647644, | |
| "eval_runtime": 2286.7661, | |
| "eval_samples_per_second": 45.124, | |
| "eval_steps_per_second": 2.821, | |
| "step": 7825 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "learning_rate": 0.0005010351966873705, | |
| "loss": 1.0962, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "eval_accuracy": 0.7721261102435742, | |
| "eval_loss": 1.0252578258514404, | |
| "eval_runtime": 2286.3022, | |
| "eval_samples_per_second": 45.133, | |
| "eval_steps_per_second": 2.821, | |
| "step": 8138 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "eval_accuracy": 0.7741078076971635, | |
| "eval_loss": 1.013240098953247, | |
| "eval_runtime": 2286.5811, | |
| "eval_samples_per_second": 45.127, | |
| "eval_steps_per_second": 2.821, | |
| "step": 8451 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "learning_rate": 0.0004906832298136645, | |
| "loss": 1.0815, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 2.72, | |
| "eval_accuracy": 0.7759812989947295, | |
| "eval_loss": 1.0042061805725098, | |
| "eval_runtime": 2285.9681, | |
| "eval_samples_per_second": 45.139, | |
| "eval_steps_per_second": 2.822, | |
| "step": 8764 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "learning_rate": 0.00048033126293995856, | |
| "loss": 1.0656, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 2.82, | |
| "eval_accuracy": 0.7772353193440255, | |
| "eval_loss": 0.9960096478462219, | |
| "eval_runtime": 2287.724, | |
| "eval_samples_per_second": 45.105, | |
| "eval_steps_per_second": 2.819, | |
| "step": 9077 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "eval_accuracy": 0.7787955466825444, | |
| "eval_loss": 0.9880508780479431, | |
| "eval_runtime": 2286.544, | |
| "eval_samples_per_second": 45.128, | |
| "eval_steps_per_second": 2.821, | |
| "step": 9390 | |
| }, | |
| { | |
| "epoch": 2.95, | |
| "learning_rate": 0.00046997929606625255, | |
| "loss": 1.0529, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "eval_accuracy": 0.7798999599716246, | |
| "eval_loss": 0.9820164442062378, | |
| "eval_runtime": 2286.6981, | |
| "eval_samples_per_second": 45.125, | |
| "eval_steps_per_second": 2.821, | |
| "step": 9703 | |
| }, | |
| { | |
| "epoch": 3.11, | |
| "learning_rate": 0.0004596273291925466, | |
| "loss": 1.0369, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 3.11, | |
| "eval_accuracy": 0.7809209831853372, | |
| "eval_loss": 0.9759008288383484, | |
| "eval_runtime": 2286.5121, | |
| "eval_samples_per_second": 45.129, | |
| "eval_steps_per_second": 2.821, | |
| "step": 10016 | |
| }, | |
| { | |
| "epoch": 3.21, | |
| "eval_accuracy": 0.7824648109671964, | |
| "eval_loss": 0.9672925472259521, | |
| "eval_runtime": 2286.3624, | |
| "eval_samples_per_second": 45.132, | |
| "eval_steps_per_second": 2.821, | |
| "step": 10329 | |
| }, | |
| { | |
| "epoch": 3.26, | |
| "learning_rate": 0.0004492753623188405, | |
| "loss": 1.0243, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 3.3, | |
| "eval_accuracy": 0.7833865666507331, | |
| "eval_loss": 0.9606760740280151, | |
| "eval_runtime": 2287.5123, | |
| "eval_samples_per_second": 45.109, | |
| "eval_steps_per_second": 2.82, | |
| "step": 10642 | |
| }, | |
| { | |
| "epoch": 3.4, | |
| "eval_accuracy": 0.7848175718314938, | |
| "eval_loss": 0.9529650807380676, | |
| "eval_runtime": 2285.505, | |
| "eval_samples_per_second": 45.148, | |
| "eval_steps_per_second": 2.822, | |
| "step": 10955 | |
| }, | |
| { | |
| "epoch": 3.42, | |
| "learning_rate": 0.0004389233954451345, | |
| "loss": 1.0122, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 3.5, | |
| "eval_accuracy": 0.7858604580968149, | |
| "eval_loss": 0.9482426047325134, | |
| "eval_runtime": 2286.9396, | |
| "eval_samples_per_second": 45.12, | |
| "eval_steps_per_second": 2.82, | |
| "step": 11268 | |
| }, | |
| { | |
| "epoch": 3.57, | |
| "learning_rate": 0.00042857142857142855, | |
| "loss": 1.0032, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "eval_accuracy": 0.7869931740974813, | |
| "eval_loss": 0.9418181777000427, | |
| "eval_runtime": 2286.0439, | |
| "eval_samples_per_second": 45.138, | |
| "eval_steps_per_second": 2.821, | |
| "step": 11581 | |
| }, | |
| { | |
| "epoch": 3.69, | |
| "eval_accuracy": 0.7881060845230868, | |
| "eval_loss": 0.935520350933075, | |
| "eval_runtime": 2287.1216, | |
| "eval_samples_per_second": 45.117, | |
| "eval_steps_per_second": 2.82, | |
| "step": 11894 | |
| }, | |
| { | |
| "epoch": 3.73, | |
| "learning_rate": 0.00041821946169772254, | |
| "loss": 0.9946, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 3.79, | |
| "eval_accuracy": 0.789029778259283, | |
| "eval_loss": 0.9287633299827576, | |
| "eval_runtime": 2287.6344, | |
| "eval_samples_per_second": 45.106, | |
| "eval_steps_per_second": 2.82, | |
| "step": 12207 | |
| }, | |
| { | |
| "epoch": 3.88, | |
| "learning_rate": 0.0004078674948240166, | |
| "loss": 0.9843, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 3.89, | |
| "eval_accuracy": 0.7905491091549709, | |
| "eval_loss": 0.9216060638427734, | |
| "eval_runtime": 2286.907, | |
| "eval_samples_per_second": 45.121, | |
| "eval_steps_per_second": 2.82, | |
| "step": 12520 | |
| }, | |
| { | |
| "epoch": 3.99, | |
| "eval_accuracy": 0.7915253922756511, | |
| "eval_loss": 0.9164555072784424, | |
| "eval_runtime": 2286.4984, | |
| "eval_samples_per_second": 45.129, | |
| "eval_steps_per_second": 2.821, | |
| "step": 12833 | |
| }, | |
| { | |
| "epoch": 4.04, | |
| "learning_rate": 0.0003975155279503105, | |
| "loss": 0.9732, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 4.08, | |
| "eval_accuracy": 0.79180229021036, | |
| "eval_loss": 0.9145598411560059, | |
| "eval_runtime": 2286.6882, | |
| "eval_samples_per_second": 45.125, | |
| "eval_steps_per_second": 2.821, | |
| "step": 13146 | |
| }, | |
| { | |
| "epoch": 4.18, | |
| "eval_accuracy": 0.7928778039058078, | |
| "eval_loss": 0.9099816679954529, | |
| "eval_runtime": 2286.2419, | |
| "eval_samples_per_second": 45.134, | |
| "eval_steps_per_second": 2.821, | |
| "step": 13459 | |
| }, | |
| { | |
| "epoch": 4.19, | |
| "learning_rate": 0.0003871635610766045, | |
| "loss": 0.9616, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 4.28, | |
| "eval_accuracy": 0.7939590053283955, | |
| "eval_loss": 0.9022164940834045, | |
| "eval_runtime": 2289.1496, | |
| "eval_samples_per_second": 45.077, | |
| "eval_steps_per_second": 2.818, | |
| "step": 13772 | |
| }, | |
| { | |
| "epoch": 4.35, | |
| "learning_rate": 0.00037681159420289854, | |
| "loss": 0.9944, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 4.37, | |
| "eval_accuracy": 0.7947964050782428, | |
| "eval_loss": 0.8981769680976868, | |
| "eval_runtime": 2284.4024, | |
| "eval_samples_per_second": 45.17, | |
| "eval_steps_per_second": 2.823, | |
| "step": 14085 | |
| }, | |
| { | |
| "epoch": 4.47, | |
| "eval_accuracy": 0.7957720898669994, | |
| "eval_loss": 0.8913259506225586, | |
| "eval_runtime": 2283.9356, | |
| "eval_samples_per_second": 45.179, | |
| "eval_steps_per_second": 2.824, | |
| "step": 14398 | |
| }, | |
| { | |
| "epoch": 4.5, | |
| "learning_rate": 0.0003664596273291925, | |
| "loss": 0.9478, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 4.57, | |
| "eval_accuracy": 0.7969175432650357, | |
| "eval_loss": 0.8867812752723694, | |
| "eval_runtime": 2285.4828, | |
| "eval_samples_per_second": 45.149, | |
| "eval_steps_per_second": 2.822, | |
| "step": 14711 | |
| }, | |
| { | |
| "epoch": 4.66, | |
| "learning_rate": 0.0003561076604554865, | |
| "loss": 0.9393, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 4.67, | |
| "eval_accuracy": 0.797533680364233, | |
| "eval_loss": 0.8840567469596863, | |
| "eval_runtime": 2285.1994, | |
| "eval_samples_per_second": 45.154, | |
| "eval_steps_per_second": 2.823, | |
| "step": 15024 | |
| }, | |
| { | |
| "epoch": 4.76, | |
| "eval_accuracy": 0.7979718683498008, | |
| "eval_loss": 0.8810065388679504, | |
| "eval_runtime": 2284.6667, | |
| "eval_samples_per_second": 45.165, | |
| "eval_steps_per_second": 2.823, | |
| "step": 15337 | |
| }, | |
| { | |
| "epoch": 4.81, | |
| "learning_rate": 0.0003457556935817805, | |
| "loss": 0.9318, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 4.86, | |
| "eval_accuracy": 0.7990604182276236, | |
| "eval_loss": 0.8742536306381226, | |
| "eval_runtime": 2285.2497, | |
| "eval_samples_per_second": 45.153, | |
| "eval_steps_per_second": 2.822, | |
| "step": 15650 | |
| }, | |
| { | |
| "epoch": 4.96, | |
| "eval_accuracy": 0.7999829388654707, | |
| "eval_loss": 0.8692484498023987, | |
| "eval_runtime": 2285.3369, | |
| "eval_samples_per_second": 45.152, | |
| "eval_steps_per_second": 2.822, | |
| "step": 15963 | |
| }, | |
| { | |
| "epoch": 4.97, | |
| "learning_rate": 0.0003354037267080745, | |
| "loss": 0.9244, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 5.05, | |
| "eval_accuracy": 0.8005920131667231, | |
| "eval_loss": 0.8657424449920654, | |
| "eval_runtime": 2284.9682, | |
| "eval_samples_per_second": 45.159, | |
| "eval_steps_per_second": 2.823, | |
| "step": 16276 | |
| }, | |
| { | |
| "epoch": 5.12, | |
| "learning_rate": 0.0003250517598343685, | |
| "loss": 0.9136, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 5.15, | |
| "eval_accuracy": 0.8014606980049568, | |
| "eval_loss": 0.8616141080856323, | |
| "eval_runtime": 2286.3782, | |
| "eval_samples_per_second": 45.131, | |
| "eval_steps_per_second": 2.821, | |
| "step": 16589 | |
| }, | |
| { | |
| "epoch": 5.25, | |
| "eval_accuracy": 0.8017829977599819, | |
| "eval_loss": 0.8604244589805603, | |
| "eval_runtime": 2285.0235, | |
| "eval_samples_per_second": 45.158, | |
| "eval_steps_per_second": 2.823, | |
| "step": 16902 | |
| }, | |
| { | |
| "epoch": 5.28, | |
| "learning_rate": 0.00031469979296066245, | |
| "loss": 1.0126, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 5.35, | |
| "eval_accuracy": 0.8022356256348646, | |
| "eval_loss": 0.857310950756073, | |
| "eval_runtime": 2286.9153, | |
| "eval_samples_per_second": 45.121, | |
| "eval_steps_per_second": 2.82, | |
| "step": 17215 | |
| }, | |
| { | |
| "epoch": 5.43, | |
| "learning_rate": 0.0003043478260869565, | |
| "loss": 0.9029, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 5.44, | |
| "eval_accuracy": 0.803324097225058, | |
| "eval_loss": 0.8512480854988098, | |
| "eval_runtime": 2286.0471, | |
| "eval_samples_per_second": 45.138, | |
| "eval_steps_per_second": 2.821, | |
| "step": 17528 | |
| }, | |
| { | |
| "epoch": 5.54, | |
| "eval_accuracy": 0.8043363001480831, | |
| "eval_loss": 0.846803605556488, | |
| "eval_runtime": 2285.0827, | |
| "eval_samples_per_second": 45.157, | |
| "eval_steps_per_second": 2.823, | |
| "step": 17841 | |
| }, | |
| { | |
| "epoch": 5.59, | |
| "learning_rate": 0.0002939958592132505, | |
| "loss": 0.8978, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 5.64, | |
| "eval_accuracy": 0.8043422672863869, | |
| "eval_loss": 0.8456587791442871, | |
| "eval_runtime": 2285.9389, | |
| "eval_samples_per_second": 45.14, | |
| "eval_steps_per_second": 2.822, | |
| "step": 18154 | |
| }, | |
| { | |
| "epoch": 5.73, | |
| "eval_accuracy": 0.8058391901420066, | |
| "eval_loss": 0.8390516638755798, | |
| "eval_runtime": 2287.2411, | |
| "eval_samples_per_second": 45.114, | |
| "eval_steps_per_second": 2.82, | |
| "step": 18467 | |
| }, | |
| { | |
| "epoch": 5.75, | |
| "learning_rate": 0.00028364389233954447, | |
| "loss": 0.8892, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 5.83, | |
| "eval_accuracy": 0.8061718215677711, | |
| "eval_loss": 0.8365983963012695, | |
| "eval_runtime": 2286.8653, | |
| "eval_samples_per_second": 45.122, | |
| "eval_steps_per_second": 2.82, | |
| "step": 18780 | |
| }, | |
| { | |
| "epoch": 5.9, | |
| "learning_rate": 0.0002732919254658385, | |
| "loss": 0.8823, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 5.93, | |
| "eval_accuracy": 0.8067812938706644, | |
| "eval_loss": 0.8338650465011597, | |
| "eval_runtime": 2283.6467, | |
| "eval_samples_per_second": 45.185, | |
| "eval_steps_per_second": 2.824, | |
| "step": 19093 | |
| }, | |
| { | |
| "epoch": 6.03, | |
| "eval_accuracy": 0.8076228111411446, | |
| "eval_loss": 0.8287978768348694, | |
| "eval_runtime": 2283.5755, | |
| "eval_samples_per_second": 45.187, | |
| "eval_steps_per_second": 2.825, | |
| "step": 19406 | |
| }, | |
| { | |
| "epoch": 6.06, | |
| "learning_rate": 0.0002629399585921325, | |
| "loss": 0.8767, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 6.12, | |
| "eval_accuracy": 0.8079247096194075, | |
| "eval_loss": 0.8258889317512512, | |
| "eval_runtime": 2283.9062, | |
| "eval_samples_per_second": 45.18, | |
| "eval_steps_per_second": 2.824, | |
| "step": 19719 | |
| }, | |
| { | |
| "epoch": 6.21, | |
| "learning_rate": 0.0002525879917184265, | |
| "loss": 0.8695, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 6.22, | |
| "eval_accuracy": 0.8087667947840288, | |
| "eval_loss": 0.8213385343551636, | |
| "eval_runtime": 2283.1428, | |
| "eval_samples_per_second": 45.195, | |
| "eval_steps_per_second": 2.825, | |
| "step": 20032 | |
| }, | |
| { | |
| "epoch": 6.32, | |
| "eval_accuracy": 0.8094497691620606, | |
| "eval_loss": 0.818403959274292, | |
| "eval_runtime": 2282.9543, | |
| "eval_samples_per_second": 45.199, | |
| "eval_steps_per_second": 2.825, | |
| "step": 20345 | |
| }, | |
| { | |
| "epoch": 6.37, | |
| "learning_rate": 0.00024223602484472047, | |
| "loss": 0.8638, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 6.42, | |
| "eval_accuracy": 0.8098454214492765, | |
| "eval_loss": 0.8152730464935303, | |
| "eval_runtime": 2283.8904, | |
| "eval_samples_per_second": 45.18, | |
| "eval_steps_per_second": 2.824, | |
| "step": 20658 | |
| }, | |
| { | |
| "epoch": 6.51, | |
| "eval_accuracy": 0.8105404823136922, | |
| "eval_loss": 0.8126282095909119, | |
| "eval_runtime": 2283.7674, | |
| "eval_samples_per_second": 45.183, | |
| "eval_steps_per_second": 2.824, | |
| "step": 20971 | |
| }, | |
| { | |
| "epoch": 6.52, | |
| "learning_rate": 0.00023188405797101448, | |
| "loss": 0.8581, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 6.61, | |
| "eval_accuracy": 0.8113756054828404, | |
| "eval_loss": 0.8074054718017578, | |
| "eval_runtime": 2286.9317, | |
| "eval_samples_per_second": 45.12, | |
| "eval_steps_per_second": 2.82, | |
| "step": 21284 | |
| }, | |
| { | |
| "epoch": 6.68, | |
| "learning_rate": 0.00022153209109730847, | |
| "loss": 0.852, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 6.71, | |
| "eval_accuracy": 0.8117181171348008, | |
| "eval_loss": 0.8063432574272156, | |
| "eval_runtime": 2287.1237, | |
| "eval_samples_per_second": 45.116, | |
| "eval_steps_per_second": 2.82, | |
| "step": 21597 | |
| }, | |
| { | |
| "epoch": 6.8, | |
| "eval_accuracy": 0.8124529403827395, | |
| "eval_loss": 0.8029435276985168, | |
| "eval_runtime": 2287.0757, | |
| "eval_samples_per_second": 45.117, | |
| "eval_steps_per_second": 2.82, | |
| "step": 21910 | |
| }, | |
| { | |
| "epoch": 6.83, | |
| "learning_rate": 0.00021118012422360248, | |
| "loss": 0.8471, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 6.9, | |
| "eval_accuracy": 0.8129203966068748, | |
| "eval_loss": 0.8003395199775696, | |
| "eval_runtime": 2284.3383, | |
| "eval_samples_per_second": 45.172, | |
| "eval_steps_per_second": 2.824, | |
| "step": 22223 | |
| }, | |
| { | |
| "epoch": 6.99, | |
| "learning_rate": 0.00020082815734989647, | |
| "loss": 0.8433, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.8138011440474225, | |
| "eval_loss": 0.7962095141410828, | |
| "eval_runtime": 2284.3684, | |
| "eval_samples_per_second": 45.171, | |
| "eval_steps_per_second": 2.824, | |
| "step": 22536 | |
| }, | |
| { | |
| "epoch": 7.1, | |
| "eval_accuracy": 0.8143908260747907, | |
| "eval_loss": 0.7925078272819519, | |
| "eval_runtime": 2283.4321, | |
| "eval_samples_per_second": 45.189, | |
| "eval_steps_per_second": 2.825, | |
| "step": 22849 | |
| }, | |
| { | |
| "epoch": 7.14, | |
| "learning_rate": 0.00019047619047619045, | |
| "loss": 0.833, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 7.19, | |
| "eval_accuracy": 0.8149341946879414, | |
| "eval_loss": 0.7910623550415039, | |
| "eval_runtime": 2282.6794, | |
| "eval_samples_per_second": 45.204, | |
| "eval_steps_per_second": 2.826, | |
| "step": 23162 | |
| }, | |
| { | |
| "epoch": 7.29, | |
| "eval_accuracy": 0.8155852939319885, | |
| "eval_loss": 0.787351131439209, | |
| "eval_runtime": 2280.6576, | |
| "eval_samples_per_second": 45.244, | |
| "eval_steps_per_second": 2.828, | |
| "step": 23475 | |
| }, | |
| { | |
| "epoch": 7.3, | |
| "learning_rate": 0.00018012422360248444, | |
| "loss": 0.8292, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 7.39, | |
| "eval_accuracy": 0.8161142535647913, | |
| "eval_loss": 0.7855063676834106, | |
| "eval_runtime": 2280.5342, | |
| "eval_samples_per_second": 45.247, | |
| "eval_steps_per_second": 2.828, | |
| "step": 23788 | |
| }, | |
| { | |
| "epoch": 7.45, | |
| "learning_rate": 0.00016977225672877845, | |
| "loss": 0.8239, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 7.48, | |
| "eval_accuracy": 0.8166872710027051, | |
| "eval_loss": 0.7811318039894104, | |
| "eval_runtime": 2280.9511, | |
| "eval_samples_per_second": 45.239, | |
| "eval_steps_per_second": 2.828, | |
| "step": 24101 | |
| }, | |
| { | |
| "epoch": 7.58, | |
| "eval_accuracy": 0.8169237775843518, | |
| "eval_loss": 0.7789031267166138, | |
| "eval_runtime": 2283.8263, | |
| "eval_samples_per_second": 45.182, | |
| "eval_steps_per_second": 2.824, | |
| "step": 24414 | |
| }, | |
| { | |
| "epoch": 7.61, | |
| "learning_rate": 0.00015942028985507247, | |
| "loss": 0.8199, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 7.68, | |
| "eval_accuracy": 0.817701750352434, | |
| "eval_loss": 0.7756628394126892, | |
| "eval_runtime": 2282.7729, | |
| "eval_samples_per_second": 45.202, | |
| "eval_steps_per_second": 2.826, | |
| "step": 24727 | |
| }, | |
| { | |
| "epoch": 7.76, | |
| "learning_rate": 0.00014906832298136645, | |
| "loss": 0.8161, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 7.78, | |
| "eval_accuracy": 0.8180780046913727, | |
| "eval_loss": 0.7731657028198242, | |
| "eval_runtime": 2282.4287, | |
| "eval_samples_per_second": 45.209, | |
| "eval_steps_per_second": 2.826, | |
| "step": 25040 | |
| }, | |
| { | |
| "epoch": 7.87, | |
| "eval_accuracy": 0.818685810759912, | |
| "eval_loss": 0.7693967223167419, | |
| "eval_runtime": 2283.0222, | |
| "eval_samples_per_second": 45.198, | |
| "eval_steps_per_second": 2.825, | |
| "step": 25353 | |
| }, | |
| { | |
| "epoch": 7.92, | |
| "learning_rate": 0.00013871635610766044, | |
| "loss": 0.8109, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 7.97, | |
| "eval_accuracy": 0.8192860535759192, | |
| "eval_loss": 0.7670938372612, | |
| "eval_runtime": 2281.3276, | |
| "eval_samples_per_second": 45.231, | |
| "eval_steps_per_second": 2.827, | |
| "step": 25666 | |
| }, | |
| { | |
| "epoch": 8.07, | |
| "eval_accuracy": 0.8195751214315984, | |
| "eval_loss": 0.765252947807312, | |
| "eval_runtime": 2281.8385, | |
| "eval_samples_per_second": 45.221, | |
| "eval_steps_per_second": 2.827, | |
| "step": 25979 | |
| }, | |
| { | |
| "epoch": 8.07, | |
| "learning_rate": 0.00012836438923395443, | |
| "loss": 0.8056, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 8.16, | |
| "eval_accuracy": 0.8201802172658903, | |
| "eval_loss": 0.7628767490386963, | |
| "eval_runtime": 2281.1128, | |
| "eval_samples_per_second": 45.235, | |
| "eval_steps_per_second": 2.828, | |
| "step": 26292 | |
| }, | |
| { | |
| "epoch": 8.23, | |
| "learning_rate": 0.00011801242236024844, | |
| "loss": 0.7985, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 8.26, | |
| "eval_accuracy": 0.8206215247730642, | |
| "eval_loss": 0.7607190608978271, | |
| "eval_runtime": 2281.4692, | |
| "eval_samples_per_second": 45.228, | |
| "eval_steps_per_second": 2.827, | |
| "step": 26605 | |
| }, | |
| { | |
| "epoch": 8.36, | |
| "eval_accuracy": 0.8213350638715384, | |
| "eval_loss": 0.7569999694824219, | |
| "eval_runtime": 2281.2083, | |
| "eval_samples_per_second": 45.233, | |
| "eval_steps_per_second": 2.827, | |
| "step": 26918 | |
| }, | |
| { | |
| "epoch": 8.38, | |
| "learning_rate": 0.00010766045548654244, | |
| "loss": 0.7974, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 8.46, | |
| "eval_accuracy": 0.8217071873214412, | |
| "eval_loss": 0.7556775808334351, | |
| "eval_runtime": 2284.1074, | |
| "eval_samples_per_second": 45.176, | |
| "eval_steps_per_second": 2.824, | |
| "step": 27231 | |
| }, | |
| { | |
| "epoch": 8.54, | |
| "learning_rate": 9.730848861283643e-05, | |
| "loss": 0.7904, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 8.55, | |
| "eval_accuracy": 0.8221692281949351, | |
| "eval_loss": 0.7519500255584717, | |
| "eval_runtime": 2284.0984, | |
| "eval_samples_per_second": 45.176, | |
| "eval_steps_per_second": 2.824, | |
| "step": 27544 | |
| }, | |
| { | |
| "epoch": 8.65, | |
| "eval_accuracy": 0.8223129100927267, | |
| "eval_loss": 0.7522034645080566, | |
| "eval_runtime": 2282.5801, | |
| "eval_samples_per_second": 45.206, | |
| "eval_steps_per_second": 2.826, | |
| "step": 27857 | |
| }, | |
| { | |
| "epoch": 8.7, | |
| "learning_rate": 8.695652173913043e-05, | |
| "loss": 0.786, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 8.75, | |
| "eval_accuracy": 0.822818443765433, | |
| "eval_loss": 0.7492069005966187, | |
| "eval_runtime": 2285.4968, | |
| "eval_samples_per_second": 45.149, | |
| "eval_steps_per_second": 2.822, | |
| "step": 28170 | |
| }, | |
| { | |
| "epoch": 8.85, | |
| "eval_accuracy": 0.8232270363634561, | |
| "eval_loss": 0.7467764019966125, | |
| "eval_runtime": 2283.855, | |
| "eval_samples_per_second": 45.181, | |
| "eval_steps_per_second": 2.824, | |
| "step": 28483 | |
| }, | |
| { | |
| "epoch": 8.85, | |
| "learning_rate": 7.660455486542441e-05, | |
| "loss": 0.7849, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 8.94, | |
| "eval_accuracy": 0.8238604187982002, | |
| "eval_loss": 0.7439348697662354, | |
| "eval_runtime": 2282.6461, | |
| "eval_samples_per_second": 45.205, | |
| "eval_steps_per_second": 2.826, | |
| "step": 28796 | |
| }, | |
| { | |
| "epoch": 9.01, | |
| "learning_rate": 6.625258799171841e-05, | |
| "loss": 0.7795, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 9.04, | |
| "eval_accuracy": 0.8241178859950476, | |
| "eval_loss": 0.7431699633598328, | |
| "eval_runtime": 2283.3834, | |
| "eval_samples_per_second": 45.19, | |
| "eval_steps_per_second": 2.825, | |
| "step": 29109 | |
| }, | |
| { | |
| "epoch": 9.14, | |
| "eval_accuracy": 0.8245760960749274, | |
| "eval_loss": 0.7410110235214233, | |
| "eval_runtime": 2282.9831, | |
| "eval_samples_per_second": 45.198, | |
| "eval_steps_per_second": 2.825, | |
| "step": 29422 | |
| }, | |
| { | |
| "epoch": 9.16, | |
| "learning_rate": 5.5900621118012414e-05, | |
| "loss": 0.7729, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 9.23, | |
| "eval_accuracy": 0.8251275339014902, | |
| "eval_loss": 0.7382827997207642, | |
| "eval_runtime": 2283.3756, | |
| "eval_samples_per_second": 45.191, | |
| "eval_steps_per_second": 2.825, | |
| "step": 29735 | |
| }, | |
| { | |
| "epoch": 9.32, | |
| "learning_rate": 4.554865424430642e-05, | |
| "loss": 0.769, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 9.33, | |
| "eval_accuracy": 0.8251445663318959, | |
| "eval_loss": 0.7376763224601746, | |
| "eval_runtime": 2285.0249, | |
| "eval_samples_per_second": 45.158, | |
| "eval_steps_per_second": 2.823, | |
| "step": 30048 | |
| }, | |
| { | |
| "epoch": 9.43, | |
| "eval_accuracy": 0.8256427450961197, | |
| "eval_loss": 0.7350980639457703, | |
| "eval_runtime": 2284.3614, | |
| "eval_samples_per_second": 45.171, | |
| "eval_steps_per_second": 2.824, | |
| "step": 30361 | |
| }, | |
| { | |
| "epoch": 9.47, | |
| "learning_rate": 3.5196687370600414e-05, | |
| "loss": 0.7685, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 9.53, | |
| "eval_accuracy": 0.825855438078354, | |
| "eval_loss": 0.7343987822532654, | |
| "eval_runtime": 2284.6411, | |
| "eval_samples_per_second": 45.166, | |
| "eval_steps_per_second": 2.823, | |
| "step": 30674 | |
| }, | |
| { | |
| "epoch": 9.62, | |
| "eval_accuracy": 0.8260118981852421, | |
| "eval_loss": 0.7326580286026001, | |
| "eval_runtime": 2284.6851, | |
| "eval_samples_per_second": 45.165, | |
| "eval_steps_per_second": 2.823, | |
| "step": 30987 | |
| }, | |
| { | |
| "epoch": 9.63, | |
| "learning_rate": 2.4844720496894407e-05, | |
| "loss": 0.7654, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 9.72, | |
| "eval_accuracy": 0.8263685390350757, | |
| "eval_loss": 0.7310478091239929, | |
| "eval_runtime": 2284.8075, | |
| "eval_samples_per_second": 45.162, | |
| "eval_steps_per_second": 2.823, | |
| "step": 31300 | |
| }, | |
| { | |
| "epoch": 9.78, | |
| "learning_rate": 1.4492753623188405e-05, | |
| "loss": 0.7627, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 9.82, | |
| "eval_accuracy": 0.8265398845224224, | |
| "eval_loss": 0.7310448884963989, | |
| "eval_runtime": 2283.7965, | |
| "eval_samples_per_second": 45.182, | |
| "eval_steps_per_second": 2.824, | |
| "step": 31613 | |
| }, | |
| { | |
| "epoch": 9.91, | |
| "eval_accuracy": 0.8269284900752123, | |
| "eval_loss": 0.7289302945137024, | |
| "eval_runtime": 2283.9473, | |
| "eval_samples_per_second": 45.179, | |
| "eval_steps_per_second": 2.824, | |
| "step": 31926 | |
| }, | |
| { | |
| "epoch": 9.94, | |
| "learning_rate": 4.140786749482401e-06, | |
| "loss": 0.7611, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 32200, | |
| "total_flos": 2.708513974181422e+18, | |
| "train_loss": 0.48268780489145596, | |
| "train_runtime": 544749.7789, | |
| "train_samples_per_second": 15.133, | |
| "train_steps_per_second": 0.059 | |
| } | |
| ], | |
| "max_steps": 32200, | |
| "num_train_epochs": 10, | |
| "total_flos": 2.708513974181422e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |