| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.9999762505640492, | |
| "global_step": 50000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0, | |
| "learning_rate": 1.2e-06, | |
| "loss": 12.4223, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "learning_rate": 0.0003, | |
| "loss": 2.6675, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 0.0006, | |
| "loss": 1.2324, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 0.0005999622383021625, | |
| "loss": 1.1509, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 0.0005998489627149555, | |
| "loss": 1.1126, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 0.0005996602017549024, | |
| "loss": 1.0883, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 0.0005993960029415653, | |
| "loss": 1.0704, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 0.0005990564327855827, | |
| "loss": 1.057, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 0.0005986415767719254, | |
| "loss": 1.0441, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 0.0005981515393383762, | |
| "loss": 1.035, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 0.0005975864438492385, | |
| "loss": 1.0263, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "eval_loss": 0.9570981860160828, | |
| "eval_runtime": 730.5149, | |
| "eval_samples_per_second": 140.175, | |
| "eval_steps_per_second": 4.38, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 0.0005969464325642798, | |
| "loss": 1.0184, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 0.0005962316666029183, | |
| "loss": 1.0113, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 0.0005954423259036624, | |
| "loss": 1.0046, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 0.0005945786091788119, | |
| "loss": 0.9991, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 0.0005936407338644336, | |
| "loss": 0.9945, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 0.0005926289360656221, | |
| "loss": 0.9888, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 0.0005915434704970625, | |
| "loss": 0.9842, | |
| "step": 4250 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 0.0005903846104189068, | |
| "loss": 0.9801, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 0.0005891526475679825, | |
| "loss": 0.9767, | |
| "step": 4750 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 0.0005878478920843492, | |
| "loss": 0.9721, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "eval_loss": 0.9183236956596375, | |
| "eval_runtime": 729.0743, | |
| "eval_samples_per_second": 140.452, | |
| "eval_steps_per_second": 4.389, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 0.0005864706724332221, | |
| "loss": 0.9676, | |
| "step": 5250 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 0.0005850213353222835, | |
| "loss": 0.9645, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 0.0005835002456144005, | |
| "loss": 0.9605, | |
| "step": 5750 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 0.0005819077862357724, | |
| "loss": 0.9581, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 0.000580244358079532, | |
| "loss": 0.9546, | |
| "step": 6250 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 0.0005785103799048218, | |
| "loss": 0.9522, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 0.0005767062882313744, | |
| "loss": 0.9495, | |
| "step": 6750 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 0.0005748325372296208, | |
| "loss": 0.9469, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 0.0005728895986063555, | |
| "loss": 0.9439, | |
| "step": 7250 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.0005708779614859863, | |
| "loss": 0.9416, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "eval_loss": 0.8962864875793457, | |
| "eval_runtime": 729.2354, | |
| "eval_samples_per_second": 140.421, | |
| "eval_steps_per_second": 4.388, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.0005687981322874007, | |
| "loss": 0.9409, | |
| "step": 7750 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 0.000566650634596477, | |
| "loss": 0.9379, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 0.0005644360090342746, | |
| "loss": 0.936, | |
| "step": 8250 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 0.0005621548131209354, | |
| "loss": 0.9333, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 0.0005598076211353316, | |
| "loss": 0.9319, | |
| "step": 8750 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 0.000557395023970493, | |
| "loss": 0.93, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 0.0005549176289848543, | |
| "loss": 0.9278, | |
| "step": 9250 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 0.0005523760598493544, | |
| "loss": 0.9263, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 0.0005497709563904314, | |
| "loss": 0.9248, | |
| "step": 9750 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 0.0005471029744289498, | |
| "loss": 0.9239, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_loss": 0.8806753754615784, | |
| "eval_runtime": 883.4007, | |
| "eval_samples_per_second": 115.916, | |
| "eval_steps_per_second": 3.622, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 0.0005443727856151006, | |
| "loss": 0.9223, | |
| "step": 10250 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 0.0005415810772593175, | |
| "loss": 0.921, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 0.0005387285521592496, | |
| "loss": 0.9192, | |
| "step": 10750 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.0005358159284228363, | |
| "loss": 0.9175, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.000532843939287527, | |
| "loss": 0.9157, | |
| "step": 11250 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 0.0005298133329356933, | |
| "loss": 0.9147, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 0.0005267248723062775, | |
| "loss": 0.9133, | |
| "step": 11750 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 0.0005235793349027264, | |
| "loss": 0.913, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 0.0005203775125972599, | |
| "loss": 0.9112, | |
| "step": 12250 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 0.000517120211431521, | |
| "loss": 0.9111, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "eval_loss": 0.8689142465591431, | |
| "eval_runtime": 886.9621, | |
| "eval_samples_per_second": 115.45, | |
| "eval_steps_per_second": 3.608, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 0.0005138082514136589, | |
| "loss": 0.9088, | |
| "step": 12750 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 0.0005104424663118964, | |
| "loss": 0.9077, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 0.0005070237034446336, | |
| "loss": 0.9066, | |
| "step": 13250 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 0.0005035528234671396, | |
| "loss": 0.9049, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 0.0005000307001548875, | |
| "loss": 0.9038, | |
| "step": 13750 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 0.0004964582201835855, | |
| "loss": 0.9034, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 0.0004928362829059618, | |
| "loss": 0.9023, | |
| "step": 14250 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 0.0004891658001253567, | |
| "loss": 0.9012, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 0.00048544769586618153, | |
| "loss": 0.9001, | |
| "step": 14750 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 0.00048168290614129995, | |
| "loss": 0.8996, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "eval_loss": 0.8599066138267517, | |
| "eval_runtime": 890.9527, | |
| "eval_samples_per_second": 114.933, | |
| "eval_steps_per_second": 3.592, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 0.00047787237871639213, | |
| "loss": 0.898, | |
| "step": 15250 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 0.0004740170728713594, | |
| "loss": 0.8978, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 0.0004701179591588311, | |
| "loss": 0.8967, | |
| "step": 15750 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 0.00046617601915983307, | |
| "loss": 0.8956, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 0.00046219224523667927, | |
| "loss": 0.8937, | |
| "step": 16250 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 0.00045816764028315066, | |
| "loss": 0.894, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 0.0004541032174720219, | |
| "loss": 0.8929, | |
| "step": 16750 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 0.00045, | |
| "loss": 0.8927, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 0.00044585902083014057, | |
| "loss": 0.8905, | |
| "step": 17250 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 0.0004416813224318048, | |
| "loss": 0.8909, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "eval_loss": 0.8517666459083557, | |
| "eval_runtime": 888.2193, | |
| "eval_samples_per_second": 115.287, | |
| "eval_steps_per_second": 3.603, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 0.00043746795651822306, | |
| "loss": 0.8893, | |
| "step": 17750 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.0004332199837817322, | |
| "loss": 0.8891, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.0004289384736267515, | |
| "loss": 0.8889, | |
| "step": 18250 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 0.00042462450390056593, | |
| "loss": 0.8883, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 0.0004202791606219841, | |
| "loss": 0.8873, | |
| "step": 18750 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 0.0004159035377079385, | |
| "loss": 0.8868, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 0.0004114987366980982, | |
| "loss": 0.8857, | |
| "step": 19250 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 0.0004070658664775615, | |
| "loss": 0.8838, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 0.00040260604299770063, | |
| "loss": 0.8838, | |
| "step": 19750 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 0.0003981203889952265, | |
| "loss": 0.8837, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "eval_loss": 0.8464268445968628, | |
| "eval_runtime": 890.045, | |
| "eval_samples_per_second": 115.05, | |
| "eval_steps_per_second": 3.595, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 0.0003936100337095461, | |
| "loss": 0.8831, | |
| "step": 20250 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 0.0003890761125984825, | |
| "loss": 0.8821, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 0.0003845197670524289, | |
| "loss": 0.8813, | |
| "step": 20750 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 0.0003799421441070104, | |
| "loss": 0.8817, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 0.0003753443961543237, | |
| "loss": 0.88, | |
| "step": 21250 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 0.0003707276806528282, | |
| "loss": 0.8802, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 0.0003660931598359622, | |
| "loss": 0.8792, | |
| "step": 21750 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.0003614420004195572, | |
| "loss": 0.8785, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.000356775373308123, | |
| "loss": 0.878, | |
| "step": 22250 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 0.0003520944533000791, | |
| "loss": 0.8775, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "eval_loss": 0.8403845429420471, | |
| "eval_runtime": 891.8066, | |
| "eval_samples_per_second": 114.823, | |
| "eval_steps_per_second": 3.588, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 0.00034740041879200497, | |
| "loss": 0.8772, | |
| "step": 22750 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 0.00034269445148198553, | |
| "loss": 0.876, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 0.00033797773607212474, | |
| "loss": 0.8746, | |
| "step": 23250 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 0.0003332514599703033, | |
| "loss": 0.8755, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 0.0003285168129912547, | |
| "loss": 0.875, | |
| "step": 23750 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 0.0003237749870570365, | |
| "loss": 0.8745, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 0.0003190271758969693, | |
| "loss": 0.8733, | |
| "step": 24250 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 0.00031427457474712274, | |
| "loss": 0.8731, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 0.0003095183800494203, | |
| "loss": 0.8732, | |
| "step": 24750 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 0.00030475978915044235, | |
| "loss": 0.8716, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "eval_loss": 0.8353695869445801, | |
| "eval_runtime": 897.249, | |
| "eval_samples_per_second": 114.127, | |
| "eval_steps_per_second": 3.566, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 0.0003, | |
| "loss": 0.871, | |
| "step": 25250 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 0.0002952402108495576, | |
| "loss": 0.8709, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 0.00029048161995057974, | |
| "loss": 0.87, | |
| "step": 25750 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 0.0002857254252528773, | |
| "loss": 0.8699, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 0.00028097282410303066, | |
| "loss": 0.869, | |
| "step": 26250 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 0.0002762250129429634, | |
| "loss": 0.8684, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 0.00027148318700874523, | |
| "loss": 0.8687, | |
| "step": 26750 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 0.0002667485400296967, | |
| "loss": 0.8686, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 0.00026202226392787515, | |
| "loss": 0.8681, | |
| "step": 27250 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 0.0002573055485180145, | |
| "loss": 0.8669, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "eval_loss": 0.8309236764907837, | |
| "eval_runtime": 889.4471, | |
| "eval_samples_per_second": 115.128, | |
| "eval_steps_per_second": 3.598, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 0.000252599581207995, | |
| "loss": 0.8677, | |
| "step": 27750 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 0.0002479055466999209, | |
| "loss": 0.8658, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 0.00024322462669187702, | |
| "loss": 0.866, | |
| "step": 28250 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 0.0002385579995804428, | |
| "loss": 0.8655, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 0.00023390684016403777, | |
| "loss": 0.8652, | |
| "step": 28750 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 0.00022927231934717176, | |
| "loss": 0.865, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 0.00022465560384567624, | |
| "loss": 0.8641, | |
| "step": 29250 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 0.00022005785589298952, | |
| "loss": 0.8639, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 0.00021548023294757105, | |
| "loss": 0.8628, | |
| "step": 29750 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 0.00021092388740151762, | |
| "loss": 0.8633, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "eval_loss": 0.8268718719482422, | |
| "eval_runtime": 886.7236, | |
| "eval_samples_per_second": 115.481, | |
| "eval_steps_per_second": 3.609, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 0.00020638996629045387, | |
| "loss": 0.8624, | |
| "step": 30250 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 0.0002018796110047735, | |
| "loss": 0.8614, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 0.00019739395700229937, | |
| "loss": 0.8612, | |
| "step": 30750 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 0.00019293413352243846, | |
| "loss": 0.8608, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 0.00018850126330190176, | |
| "loss": 0.8606, | |
| "step": 31250 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 0.00018409646229206137, | |
| "loss": 0.8601, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 0.00017972083937801593, | |
| "loss": 0.8608, | |
| "step": 31750 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 0.0001753754960994341, | |
| "loss": 0.8597, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 0.0001710615263732485, | |
| "loss": 0.8593, | |
| "step": 32250 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 0.00016678001621826772, | |
| "loss": 0.8581, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "eval_loss": 0.8240156769752502, | |
| "eval_runtime": 887.3331, | |
| "eval_samples_per_second": 115.402, | |
| "eval_steps_per_second": 3.606, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 0.00016253204348177686, | |
| "loss": 0.8594, | |
| "step": 32750 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 0.00015831867756819522, | |
| "loss": 0.8591, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 0.00015414097916985944, | |
| "loss": 0.8581, | |
| "step": 33250 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 0.00015000000000000004, | |
| "loss": 0.8573, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 0.00014589678252797817, | |
| "loss": 0.8576, | |
| "step": 33750 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 0.00014183235971684924, | |
| "loss": 0.8571, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 0.00013780775476332082, | |
| "loss": 0.8572, | |
| "step": 34250 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 0.0001338239808401669, | |
| "loss": 0.8566, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 0.0001298820408411688, | |
| "loss": 0.8558, | |
| "step": 34750 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 0.00012598292712864058, | |
| "loss": 0.8565, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "eval_loss": 0.820695698261261, | |
| "eval_runtime": 867.1933, | |
| "eval_samples_per_second": 118.082, | |
| "eval_steps_per_second": 3.69, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 0.0001221276212836079, | |
| "loss": 0.8564, | |
| "step": 35250 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 0.00011831709385870004, | |
| "loss": 0.8553, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 0.0001145523041338184, | |
| "loss": 0.8548, | |
| "step": 35750 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 0.00011083419987464334, | |
| "loss": 0.8551, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 0.00010716371709403818, | |
| "loss": 0.8543, | |
| "step": 36250 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 0.00010354177981641449, | |
| "loss": 0.8547, | |
| "step": 36500 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 9.996929984511254e-05, | |
| "loss": 0.8538, | |
| "step": 36750 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 9.644717653286037e-05, | |
| "loss": 0.8539, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 9.297629655536644e-05, | |
| "loss": 0.8547, | |
| "step": 37250 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 8.955753368810358e-05, | |
| "loss": 0.8531, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "eval_loss": 0.8180103302001953, | |
| "eval_runtime": 728.2589, | |
| "eval_samples_per_second": 140.609, | |
| "eval_steps_per_second": 4.394, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 8.619174858634122e-05, | |
| "loss": 0.8533, | |
| "step": 37750 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 8.287978856847894e-05, | |
| "loss": 0.8533, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 7.962248740274003e-05, | |
| "loss": 0.8531, | |
| "step": 38250 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 7.642066509727359e-05, | |
| "loss": 0.8523, | |
| "step": 38500 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 7.327512769372254e-05, | |
| "loss": 0.8523, | |
| "step": 38750 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 7.018666706430662e-05, | |
| "loss": 0.8524, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 6.715606071247291e-05, | |
| "loss": 0.8519, | |
| "step": 39250 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 6.418407157716381e-05, | |
| "loss": 0.851, | |
| "step": 39500 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 6.127144784075033e-05, | |
| "loss": 0.8512, | |
| "step": 39750 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 5.841892274068241e-05, | |
| "loss": 0.8508, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_loss": 0.8166452050209045, | |
| "eval_runtime": 727.869, | |
| "eval_samples_per_second": 140.685, | |
| "eval_steps_per_second": 4.396, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 5.562721438489928e-05, | |
| "loss": 0.8508, | |
| "step": 40250 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 5.2897025571050186e-05, | |
| "loss": 0.8512, | |
| "step": 40500 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 5.022904360956861e-05, | |
| "loss": 0.8513, | |
| "step": 40750 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 4.76239401506456e-05, | |
| "loss": 0.8503, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 4.5082371015145716e-05, | |
| "loss": 0.8506, | |
| "step": 41250 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 4.260497602950688e-05, | |
| "loss": 0.8495, | |
| "step": 41500 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 4.019237886466838e-05, | |
| "loss": 0.8504, | |
| "step": 41750 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 3.784518687906452e-05, | |
| "loss": 0.85, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 3.556399096572541e-05, | |
| "loss": 0.8498, | |
| "step": 42250 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 3.3349365403522986e-05, | |
| "loss": 0.8494, | |
| "step": 42500 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "eval_loss": 0.8152065873146057, | |
| "eval_runtime": 727.6431, | |
| "eval_samples_per_second": 140.728, | |
| "eval_steps_per_second": 4.398, | |
| "step": 42500 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 3.120186771259927e-05, | |
| "loss": 0.8499, | |
| "step": 42750 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 2.9122038514013678e-05, | |
| "loss": 0.8494, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 2.7110401393644464e-05, | |
| "loss": 0.8494, | |
| "step": 43250 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 2.516746277037912e-05, | |
| "loss": 0.8487, | |
| "step": 43500 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 2.329371176862562e-05, | |
| "loss": 0.8497, | |
| "step": 43750 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 2.148962009517823e-05, | |
| "loss": 0.8483, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 1.9755641920468003e-05, | |
| "loss": 0.8494, | |
| "step": 44250 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 1.8092213764227503e-05, | |
| "loss": 0.8479, | |
| "step": 44500 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 1.6499754385599462e-05, | |
| "loss": 0.8487, | |
| "step": 44750 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 1.4978664677716402e-05, | |
| "loss": 0.8483, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "eval_loss": 0.814320981502533, | |
| "eval_runtime": 728.0287, | |
| "eval_samples_per_second": 140.654, | |
| "eval_steps_per_second": 4.395, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 1.3529327566777836e-05, | |
| "loss": 0.8489, | |
| "step": 45250 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 1.2152107915650821e-05, | |
| "loss": 0.8485, | |
| "step": 45500 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 1.0847352432017387e-05, | |
| "loss": 0.8487, | |
| "step": 45750 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 9.615389581093124e-06, | |
| "loss": 0.8484, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 8.456529502937504e-06, | |
| "loss": 0.8484, | |
| "step": 46250 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 7.371063934377885e-06, | |
| "loss": 0.8483, | |
| "step": 46500 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 6.35926613556641e-06, | |
| "loss": 0.8482, | |
| "step": 46750 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 5.421390821187988e-06, | |
| "loss": 0.8479, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 4.557674096337593e-06, | |
| "loss": 0.8478, | |
| "step": 47250 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 3.768333397081713e-06, | |
| "loss": 0.8484, | |
| "step": 47500 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "eval_loss": 0.8137823343276978, | |
| "eval_runtime": 728.9533, | |
| "eval_samples_per_second": 140.475, | |
| "eval_steps_per_second": 4.39, | |
| "step": 47500 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 3.0535674357201944e-06, | |
| "loss": 0.848, | |
| "step": 47750 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 2.4135561507613975e-06, | |
| "loss": 0.8481, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 1.848460661623763e-06, | |
| "loss": 0.8478, | |
| "step": 48250 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 1.3584232280746231e-06, | |
| "loss": 0.8475, | |
| "step": 48500 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 9.435672144173178e-07, | |
| "loss": 0.8482, | |
| "step": 48750 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 6.03997058434702e-07, | |
| "loss": 0.8484, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 3.397982450976111e-07, | |
| "loss": 0.848, | |
| "step": 49250 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 1.5103728504447522e-07, | |
| "loss": 0.8481, | |
| "step": 49500 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 3.776169783747951e-08, | |
| "loss": 0.8484, | |
| "step": 49750 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 0.0, | |
| "loss": 0.8475, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.8137637972831726, | |
| "eval_runtime": 727.8603, | |
| "eval_samples_per_second": 140.686, | |
| "eval_steps_per_second": 4.396, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "step": 50000, | |
| "total_flos": 2.180439447726719e+19, | |
| "train_loss": 0.9020122967529297, | |
| "train_runtime": 575442.638, | |
| "train_samples_per_second": 44.487, | |
| "train_steps_per_second": 0.087 | |
| } | |
| ], | |
| "max_steps": 50000, | |
| "num_train_epochs": 1, | |
| "total_flos": 2.180439447726719e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |