| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.9839005449146736, | |
| "global_step": 76000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 0.0004997842653586619, | |
| "loss": 4.023, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "eval_loss": 2.9591097831726074, | |
| "eval_runtime": 235.529, | |
| "eval_samples_per_second": 167.083, | |
| "eval_steps_per_second": 10.445, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 0.0004980606215742609, | |
| "loss": 2.9094, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "eval_loss": 2.7322702407836914, | |
| "eval_runtime": 235.6062, | |
| "eval_samples_per_second": 167.029, | |
| "eval_steps_per_second": 10.441, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 0.0004946252280608798, | |
| "loss": 2.7599, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "eval_loss": 2.633601665496826, | |
| "eval_runtime": 235.5566, | |
| "eval_samples_per_second": 167.064, | |
| "eval_steps_per_second": 10.443, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 0.0004895017908540765, | |
| "loss": 2.6821, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "eval_loss": 2.571882724761963, | |
| "eval_runtime": 235.5446, | |
| "eval_samples_per_second": 167.072, | |
| "eval_steps_per_second": 10.444, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 0.0004827332514022428, | |
| "loss": 2.6301, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "eval_loss": 2.5301053524017334, | |
| "eval_runtime": 236.1109, | |
| "eval_samples_per_second": 166.672, | |
| "eval_steps_per_second": 10.419, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 0.0004743573557450783, | |
| "loss": 2.5918, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "eval_loss": 2.4954473972320557, | |
| "eval_runtime": 235.5537, | |
| "eval_samples_per_second": 167.066, | |
| "eval_steps_per_second": 10.443, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 0.00046442948254696206, | |
| "loss": 2.5616, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "eval_loss": 2.4691474437713623, | |
| "eval_runtime": 235.6133, | |
| "eval_samples_per_second": 167.024, | |
| "eval_steps_per_second": 10.441, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 0.00045302193229931784, | |
| "loss": 2.5361, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "eval_loss": 2.4461960792541504, | |
| "eval_runtime": 235.5392, | |
| "eval_samples_per_second": 167.076, | |
| "eval_steps_per_second": 10.444, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 0.00044023364212028847, | |
| "loss": 2.5158, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "eval_loss": 2.4347801208496094, | |
| "eval_runtime": 235.4954, | |
| "eval_samples_per_second": 167.107, | |
| "eval_steps_per_second": 10.446, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 0.0004261144576270032, | |
| "loss": 2.4993, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "eval_loss": 2.412471294403076, | |
| "eval_runtime": 235.4551, | |
| "eval_samples_per_second": 167.136, | |
| "eval_steps_per_second": 10.448, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 0.00041077999005815623, | |
| "loss": 2.4848, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "eval_loss": 2.3976333141326904, | |
| "eval_runtime": 236.1174, | |
| "eval_samples_per_second": 166.667, | |
| "eval_steps_per_second": 10.419, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 0.00039433605536526187, | |
| "loss": 2.4689, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "eval_loss": 2.3828279972076416, | |
| "eval_runtime": 235.5611, | |
| "eval_samples_per_second": 167.061, | |
| "eval_steps_per_second": 10.443, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 0.00037690507447365334, | |
| "loss": 2.4562, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "eval_loss": 2.3714354038238525, | |
| "eval_runtime": 237.177, | |
| "eval_samples_per_second": 165.923, | |
| "eval_steps_per_second": 10.372, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 0.00035860861171489625, | |
| "loss": 2.4415, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "eval_loss": 2.3587660789489746, | |
| "eval_runtime": 235.9789, | |
| "eval_samples_per_second": 166.765, | |
| "eval_steps_per_second": 10.425, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 0.0003395447945183574, | |
| "loss": 2.4321, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "eval_loss": 2.3464348316192627, | |
| "eval_runtime": 235.9733, | |
| "eval_samples_per_second": 166.769, | |
| "eval_steps_per_second": 10.425, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 0.0003198730435893789, | |
| "loss": 2.4217, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "eval_loss": 2.3358864784240723, | |
| "eval_runtime": 235.8391, | |
| "eval_samples_per_second": 166.864, | |
| "eval_steps_per_second": 10.431, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 0.00029970943417239666, | |
| "loss": 2.4062, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "eval_loss": 2.3258254528045654, | |
| "eval_runtime": 235.6751, | |
| "eval_samples_per_second": 166.98, | |
| "eval_steps_per_second": 10.438, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 0.00027922343434053505, | |
| "loss": 2.3968, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "eval_loss": 2.3131136894226074, | |
| "eval_runtime": 235.4005, | |
| "eval_samples_per_second": 167.175, | |
| "eval_steps_per_second": 10.45, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 0.000258556941398551, | |
| "loss": 2.3882, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "eval_loss": 2.3022758960723877, | |
| "eval_runtime": 234.8551, | |
| "eval_samples_per_second": 167.563, | |
| "eval_steps_per_second": 10.475, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 0.00023779013959131367, | |
| "loss": 2.3708, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "eval_loss": 2.292479991912842, | |
| "eval_runtime": 234.8376, | |
| "eval_samples_per_second": 167.575, | |
| "eval_steps_per_second": 10.475, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 0.00021710759229054673, | |
| "loss": 2.3597, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "eval_loss": 2.2818188667297363, | |
| "eval_runtime": 234.9302, | |
| "eval_samples_per_second": 167.509, | |
| "eval_steps_per_second": 10.471, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 0.0001966520200317064, | |
| "loss": 2.3516, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "eval_loss": 2.271631956100464, | |
| "eval_runtime": 234.8493, | |
| "eval_samples_per_second": 167.567, | |
| "eval_steps_per_second": 10.475, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 0.00017656457710219216, | |
| "loss": 2.3402, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "eval_loss": 2.262141466140747, | |
| "eval_runtime": 234.8012, | |
| "eval_samples_per_second": 167.601, | |
| "eval_steps_per_second": 10.477, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 0.0001569838775019702, | |
| "loss": 2.3317, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "eval_loss": 2.250854730606079, | |
| "eval_runtime": 234.8456, | |
| "eval_samples_per_second": 167.57, | |
| "eval_steps_per_second": 10.475, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 0.00013804503843352594, | |
| "loss": 2.3208, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "eval_loss": 2.2419745922088623, | |
| "eval_runtime": 235.4331, | |
| "eval_samples_per_second": 167.151, | |
| "eval_steps_per_second": 10.449, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "learning_rate": 0.00011988761690506509, | |
| "loss": 2.3129, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "eval_loss": 2.232053518295288, | |
| "eval_runtime": 235.1368, | |
| "eval_samples_per_second": 167.362, | |
| "eval_steps_per_second": 10.462, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "learning_rate": 0.00010262714265881779, | |
| "loss": 2.3034, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "eval_loss": 2.223116874694824, | |
| "eval_runtime": 235.501, | |
| "eval_samples_per_second": 167.103, | |
| "eval_steps_per_second": 10.446, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 8.638260310471869e-05, | |
| "loss": 2.2944, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "eval_loss": 2.215869188308716, | |
| "eval_runtime": 235.4682, | |
| "eval_samples_per_second": 167.127, | |
| "eval_steps_per_second": 10.447, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "learning_rate": 7.126598216971828e-05, | |
| "loss": 2.2867, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "eval_loss": 2.2056751251220703, | |
| "eval_runtime": 235.5179, | |
| "eval_samples_per_second": 167.091, | |
| "eval_steps_per_second": 10.445, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "learning_rate": 5.737486728530572e-05, | |
| "loss": 2.2775, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "eval_loss": 2.19879150390625, | |
| "eval_runtime": 235.4838, | |
| "eval_samples_per_second": 167.116, | |
| "eval_steps_per_second": 10.447, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "learning_rate": 4.481890186103055e-05, | |
| "loss": 2.2706, | |
| "step": 62000 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "eval_loss": 2.191859006881714, | |
| "eval_runtime": 235.3807, | |
| "eval_samples_per_second": 167.189, | |
| "eval_steps_per_second": 10.451, | |
| "step": 62000 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 3.36825867617602e-05, | |
| "loss": 2.2653, | |
| "step": 64000 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "eval_loss": 2.185826301574707, | |
| "eval_runtime": 235.8171, | |
| "eval_samples_per_second": 166.879, | |
| "eval_steps_per_second": 10.432, | |
| "step": 64000 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "learning_rate": 2.4031550628330818e-05, | |
| "loss": 2.2595, | |
| "step": 66000 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "eval_loss": 2.1810295581817627, | |
| "eval_runtime": 235.6965, | |
| "eval_samples_per_second": 166.965, | |
| "eval_steps_per_second": 10.437, | |
| "step": 66000 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "learning_rate": 1.5934609437343344e-05, | |
| "loss": 2.2546, | |
| "step": 68000 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "eval_loss": 2.1772620677948, | |
| "eval_runtime": 234.5679, | |
| "eval_samples_per_second": 167.768, | |
| "eval_steps_per_second": 10.487, | |
| "step": 68000 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "learning_rate": 9.452843465689393e-06, | |
| "loss": 2.2515, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "eval_loss": 2.174480676651001, | |
| "eval_runtime": 234.8677, | |
| "eval_samples_per_second": 167.554, | |
| "eval_steps_per_second": 10.474, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "learning_rate": 4.632970471463188e-06, | |
| "loss": 2.25, | |
| "step": 72000 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "eval_loss": 2.1722278594970703, | |
| "eval_runtime": 234.8936, | |
| "eval_samples_per_second": 167.535, | |
| "eval_steps_per_second": 10.473, | |
| "step": 72000 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "learning_rate": 1.504568227717773e-06, | |
| "loss": 2.2491, | |
| "step": 74000 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "eval_loss": 2.1710853576660156, | |
| "eval_runtime": 234.9822, | |
| "eval_samples_per_second": 167.472, | |
| "eval_steps_per_second": 10.469, | |
| "step": 74000 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "learning_rate": 8.920279775520568e-08, | |
| "loss": 2.2505, | |
| "step": 76000 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "eval_loss": 2.1710739135742188, | |
| "eval_runtime": 234.8578, | |
| "eval_samples_per_second": 167.561, | |
| "eval_steps_per_second": 10.474, | |
| "step": 76000 | |
| } | |
| ], | |
| "max_steps": 76616, | |
| "num_train_epochs": 2, | |
| "total_flos": 3.177337760907264e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |