|
{ |
|
"best_metric": null, |
|
"best_model_checkpoint": null, |
|
"epoch": 37.0, |
|
"global_step": 3922, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 1.0, |
|
"learning_rate": 1.3832186408159307e-05, |
|
"loss": 3.0358, |
|
"step": 106 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"eval_accuracy": 0.543347000166316, |
|
"eval_loss": 2.7585761547088623, |
|
"eval_runtime": 25.1028, |
|
"eval_samples_per_second": 70.51, |
|
"eval_steps_per_second": 0.239, |
|
"step": 106 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"learning_rate": 1.5888124272106204e-05, |
|
"loss": 2.7114, |
|
"step": 212 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"eval_accuracy": 0.55558049451544, |
|
"eval_loss": 2.6144614219665527, |
|
"eval_runtime": 24.4673, |
|
"eval_samples_per_second": 72.341, |
|
"eval_steps_per_second": 0.245, |
|
"step": 212 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"learning_rate": 1.7090770826327895e-05, |
|
"loss": 2.4868, |
|
"step": 318 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"eval_accuracy": 0.6212061200839091, |
|
"eval_loss": 2.034228563308716, |
|
"eval_runtime": 24.2933, |
|
"eval_samples_per_second": 72.86, |
|
"eval_steps_per_second": 0.247, |
|
"step": 318 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"learning_rate": 1.7944062136053104e-05, |
|
"loss": 2.1388, |
|
"step": 424 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"eval_accuracy": 0.6854297952501626, |
|
"eval_loss": 1.5866005420684814, |
|
"eval_runtime": 24.3894, |
|
"eval_samples_per_second": 72.572, |
|
"eval_steps_per_second": 0.246, |
|
"step": 424 |
|
}, |
|
{ |
|
"epoch": 5.0, |
|
"learning_rate": 1.860592629580032e-05, |
|
"loss": 1.8311, |
|
"step": 530 |
|
}, |
|
{ |
|
"epoch": 5.0, |
|
"eval_accuracy": 0.7191362268359381, |
|
"eval_loss": 1.3652117252349854, |
|
"eval_runtime": 24.3545, |
|
"eval_samples_per_second": 72.677, |
|
"eval_steps_per_second": 0.246, |
|
"step": 530 |
|
}, |
|
{ |
|
"epoch": 6.0, |
|
"learning_rate": 1.9146708690274792e-05, |
|
"loss": 1.5704, |
|
"step": 636 |
|
}, |
|
{ |
|
"epoch": 6.0, |
|
"eval_accuracy": 0.7248188353074707, |
|
"eval_loss": 1.3337750434875488, |
|
"eval_runtime": 24.3217, |
|
"eval_samples_per_second": 72.775, |
|
"eval_steps_per_second": 0.247, |
|
"step": 636 |
|
}, |
|
{ |
|
"epoch": 7.0, |
|
"learning_rate": 1.9603933689955228e-05, |
|
"loss": 1.4498, |
|
"step": 742 |
|
}, |
|
{ |
|
"epoch": 7.0, |
|
"eval_accuracy": 0.7381350781490347, |
|
"eval_loss": 1.2443161010742188, |
|
"eval_runtime": 24.5188, |
|
"eval_samples_per_second": 72.189, |
|
"eval_steps_per_second": 0.245, |
|
"step": 742 |
|
}, |
|
{ |
|
"epoch": 8.0, |
|
"learning_rate": 2e-05, |
|
"loss": 1.3744, |
|
"step": 848 |
|
}, |
|
{ |
|
"epoch": 8.0, |
|
"eval_accuracy": 0.7413300423856433, |
|
"eval_loss": 1.2314091920852661, |
|
"eval_runtime": 24.3764, |
|
"eval_samples_per_second": 72.611, |
|
"eval_steps_per_second": 0.246, |
|
"step": 848 |
|
}, |
|
{ |
|
"epoch": 9.0, |
|
"learning_rate": 2e-05, |
|
"loss": 1.3207, |
|
"step": 954 |
|
}, |
|
{ |
|
"epoch": 9.0, |
|
"eval_accuracy": 0.7439575072718211, |
|
"eval_loss": 1.2015495300292969, |
|
"eval_runtime": 24.5167, |
|
"eval_samples_per_second": 72.196, |
|
"eval_steps_per_second": 0.245, |
|
"step": 954 |
|
}, |
|
{ |
|
"epoch": 10.0, |
|
"learning_rate": 2e-05, |
|
"loss": 1.2892, |
|
"step": 1060 |
|
}, |
|
{ |
|
"epoch": 10.0, |
|
"eval_accuracy": 0.7558549767302207, |
|
"eval_loss": 1.1335448026657104, |
|
"eval_runtime": 24.3508, |
|
"eval_samples_per_second": 72.688, |
|
"eval_steps_per_second": 0.246, |
|
"step": 1060 |
|
}, |
|
{ |
|
"epoch": 11.0, |
|
"learning_rate": 2e-05, |
|
"loss": 1.25, |
|
"step": 1166 |
|
}, |
|
{ |
|
"epoch": 11.0, |
|
"eval_accuracy": 0.7594773916347587, |
|
"eval_loss": 1.1179267168045044, |
|
"eval_runtime": 24.3594, |
|
"eval_samples_per_second": 72.662, |
|
"eval_steps_per_second": 0.246, |
|
"step": 1166 |
|
}, |
|
{ |
|
"epoch": 12.0, |
|
"learning_rate": 2e-05, |
|
"loss": 1.2274, |
|
"step": 1272 |
|
}, |
|
{ |
|
"epoch": 12.0, |
|
"eval_accuracy": 0.7580528411147304, |
|
"eval_loss": 1.1233829259872437, |
|
"eval_runtime": 24.3279, |
|
"eval_samples_per_second": 72.756, |
|
"eval_steps_per_second": 0.247, |
|
"step": 1272 |
|
}, |
|
{ |
|
"epoch": 13.0, |
|
"learning_rate": 2e-05, |
|
"loss": 1.2027, |
|
"step": 1378 |
|
}, |
|
{ |
|
"epoch": 13.0, |
|
"eval_accuracy": 0.7637997528180235, |
|
"eval_loss": 1.0827549695968628, |
|
"eval_runtime": 24.4052, |
|
"eval_samples_per_second": 72.525, |
|
"eval_steps_per_second": 0.246, |
|
"step": 1378 |
|
}, |
|
{ |
|
"epoch": 14.0, |
|
"learning_rate": 2e-05, |
|
"loss": 1.1838, |
|
"step": 1484 |
|
}, |
|
{ |
|
"epoch": 14.0, |
|
"eval_accuracy": 0.7666054329273585, |
|
"eval_loss": 1.0769394636154175, |
|
"eval_runtime": 24.4017, |
|
"eval_samples_per_second": 72.536, |
|
"eval_steps_per_second": 0.246, |
|
"step": 1484 |
|
}, |
|
{ |
|
"epoch": 15.0, |
|
"learning_rate": 2e-05, |
|
"loss": 1.1736, |
|
"step": 1590 |
|
}, |
|
{ |
|
"epoch": 15.0, |
|
"eval_accuracy": 0.7698640416703452, |
|
"eval_loss": 1.0607359409332275, |
|
"eval_runtime": 24.3587, |
|
"eval_samples_per_second": 72.664, |
|
"eval_steps_per_second": 0.246, |
|
"step": 1590 |
|
}, |
|
{ |
|
"epoch": 16.0, |
|
"learning_rate": 2e-05, |
|
"loss": 1.1534, |
|
"step": 1696 |
|
}, |
|
{ |
|
"epoch": 16.0, |
|
"eval_accuracy": 0.7714122091985482, |
|
"eval_loss": 1.0373061895370483, |
|
"eval_runtime": 24.4342, |
|
"eval_samples_per_second": 72.439, |
|
"eval_steps_per_second": 0.246, |
|
"step": 1696 |
|
}, |
|
{ |
|
"epoch": 17.0, |
|
"learning_rate": 2e-05, |
|
"loss": 1.1388, |
|
"step": 1802 |
|
}, |
|
{ |
|
"epoch": 17.0, |
|
"eval_accuracy": 0.7687794470619611, |
|
"eval_loss": 1.0571210384368896, |
|
"eval_runtime": 24.3831, |
|
"eval_samples_per_second": 72.591, |
|
"eval_steps_per_second": 0.246, |
|
"step": 1802 |
|
}, |
|
{ |
|
"epoch": 18.0, |
|
"learning_rate": 2e-05, |
|
"loss": 1.1273, |
|
"step": 1908 |
|
}, |
|
{ |
|
"epoch": 18.0, |
|
"eval_accuracy": 0.7713278633378673, |
|
"eval_loss": 1.0348763465881348, |
|
"eval_runtime": 24.3437, |
|
"eval_samples_per_second": 72.709, |
|
"eval_steps_per_second": 0.246, |
|
"step": 1908 |
|
}, |
|
{ |
|
"epoch": 19.0, |
|
"learning_rate": 2e-05, |
|
"loss": 1.1161, |
|
"step": 2014 |
|
}, |
|
{ |
|
"epoch": 19.0, |
|
"eval_accuracy": 0.777816642254917, |
|
"eval_loss": 1.0058482885360718, |
|
"eval_runtime": 24.086, |
|
"eval_samples_per_second": 73.487, |
|
"eval_steps_per_second": 0.249, |
|
"step": 2014 |
|
}, |
|
{ |
|
"epoch": 20.0, |
|
"learning_rate": 2e-05, |
|
"loss": 1.1037, |
|
"step": 2120 |
|
}, |
|
{ |
|
"epoch": 20.0, |
|
"eval_accuracy": 0.778376203991329, |
|
"eval_loss": 1.0106089115142822, |
|
"eval_runtime": 24.3319, |
|
"eval_samples_per_second": 72.744, |
|
"eval_steps_per_second": 0.247, |
|
"step": 2120 |
|
}, |
|
{ |
|
"epoch": 21.0, |
|
"learning_rate": 2e-05, |
|
"loss": 1.0897, |
|
"step": 2226 |
|
}, |
|
{ |
|
"epoch": 21.0, |
|
"eval_accuracy": 0.7792838874680307, |
|
"eval_loss": 0.9971184730529785, |
|
"eval_runtime": 24.2829, |
|
"eval_samples_per_second": 72.891, |
|
"eval_steps_per_second": 0.247, |
|
"step": 2226 |
|
}, |
|
{ |
|
"epoch": 22.0, |
|
"learning_rate": 2e-05, |
|
"loss": 1.0833, |
|
"step": 2332 |
|
}, |
|
{ |
|
"epoch": 22.0, |
|
"eval_accuracy": 0.7836239126411253, |
|
"eval_loss": 0.9799665212631226, |
|
"eval_runtime": 24.3154, |
|
"eval_samples_per_second": 72.793, |
|
"eval_steps_per_second": 0.247, |
|
"step": 2332 |
|
}, |
|
{ |
|
"epoch": 23.0, |
|
"learning_rate": 2e-05, |
|
"loss": 1.0747, |
|
"step": 2438 |
|
}, |
|
{ |
|
"epoch": 23.0, |
|
"eval_accuracy": 0.7836160467857196, |
|
"eval_loss": 0.9719156622886658, |
|
"eval_runtime": 24.3874, |
|
"eval_samples_per_second": 72.578, |
|
"eval_steps_per_second": 0.246, |
|
"step": 2438 |
|
}, |
|
{ |
|
"epoch": 24.0, |
|
"learning_rate": 2e-05, |
|
"loss": 1.0708, |
|
"step": 2544 |
|
}, |
|
{ |
|
"epoch": 24.0, |
|
"eval_accuracy": 0.7880468638591382, |
|
"eval_loss": 0.9512822031974792, |
|
"eval_runtime": 26.1688, |
|
"eval_samples_per_second": 67.638, |
|
"eval_steps_per_second": 0.229, |
|
"step": 2544 |
|
}, |
|
{ |
|
"epoch": 25.0, |
|
"learning_rate": 2e-05, |
|
"loss": 1.0577, |
|
"step": 2650 |
|
}, |
|
{ |
|
"epoch": 25.0, |
|
"eval_accuracy": 0.7839817403876498, |
|
"eval_loss": 0.9685712456703186, |
|
"eval_runtime": 24.2295, |
|
"eval_samples_per_second": 73.052, |
|
"eval_steps_per_second": 0.248, |
|
"step": 2650 |
|
}, |
|
{ |
|
"epoch": 26.0, |
|
"learning_rate": 2e-05, |
|
"loss": 1.0551, |
|
"step": 2756 |
|
}, |
|
{ |
|
"epoch": 26.0, |
|
"eval_accuracy": 0.7858663548179133, |
|
"eval_loss": 0.9629907608032227, |
|
"eval_runtime": 24.4249, |
|
"eval_samples_per_second": 72.467, |
|
"eval_steps_per_second": 0.246, |
|
"step": 2756 |
|
}, |
|
{ |
|
"epoch": 27.0, |
|
"learning_rate": 2e-05, |
|
"loss": 1.0489, |
|
"step": 2862 |
|
}, |
|
{ |
|
"epoch": 27.0, |
|
"eval_accuracy": 0.7881803814914183, |
|
"eval_loss": 0.9433470964431763, |
|
"eval_runtime": 24.4582, |
|
"eval_samples_per_second": 72.368, |
|
"eval_steps_per_second": 0.245, |
|
"step": 2862 |
|
}, |
|
{ |
|
"epoch": 28.0, |
|
"learning_rate": 2e-05, |
|
"loss": 1.0483, |
|
"step": 2968 |
|
}, |
|
{ |
|
"epoch": 28.0, |
|
"eval_accuracy": 0.7872097449562239, |
|
"eval_loss": 0.9574456214904785, |
|
"eval_runtime": 24.3523, |
|
"eval_samples_per_second": 72.683, |
|
"eval_steps_per_second": 0.246, |
|
"step": 2968 |
|
}, |
|
{ |
|
"epoch": 29.0, |
|
"learning_rate": 2e-05, |
|
"loss": 1.0327, |
|
"step": 3074 |
|
}, |
|
{ |
|
"epoch": 29.0, |
|
"eval_accuracy": 0.7909898637829464, |
|
"eval_loss": 0.9401029348373413, |
|
"eval_runtime": 24.3913, |
|
"eval_samples_per_second": 72.567, |
|
"eval_steps_per_second": 0.246, |
|
"step": 3074 |
|
}, |
|
{ |
|
"epoch": 30.0, |
|
"learning_rate": 2e-05, |
|
"loss": 1.0362, |
|
"step": 3180 |
|
}, |
|
{ |
|
"epoch": 30.0, |
|
"eval_accuracy": 0.7919005554242496, |
|
"eval_loss": 0.9270448684692383, |
|
"eval_runtime": 24.4335, |
|
"eval_samples_per_second": 72.441, |
|
"eval_steps_per_second": 0.246, |
|
"step": 3180 |
|
}, |
|
{ |
|
"epoch": 31.0, |
|
"learning_rate": 2e-05, |
|
"loss": 1.0243, |
|
"step": 3286 |
|
}, |
|
{ |
|
"epoch": 31.0, |
|
"eval_accuracy": 0.7986601387439376, |
|
"eval_loss": 0.8908094167709351, |
|
"eval_runtime": 24.1448, |
|
"eval_samples_per_second": 73.308, |
|
"eval_steps_per_second": 0.249, |
|
"step": 3286 |
|
}, |
|
{ |
|
"epoch": 32.0, |
|
"learning_rate": 2e-05, |
|
"loss": 1.0189, |
|
"step": 3392 |
|
}, |
|
{ |
|
"epoch": 32.0, |
|
"eval_accuracy": 0.7921468329354021, |
|
"eval_loss": 0.9229845404624939, |
|
"eval_runtime": 24.3268, |
|
"eval_samples_per_second": 72.759, |
|
"eval_steps_per_second": 0.247, |
|
"step": 3392 |
|
}, |
|
{ |
|
"epoch": 33.0, |
|
"learning_rate": 2e-05, |
|
"loss": 1.0079, |
|
"step": 3498 |
|
}, |
|
{ |
|
"epoch": 33.0, |
|
"eval_accuracy": 0.7888655988210297, |
|
"eval_loss": 0.9461591243743896, |
|
"eval_runtime": 24.4302, |
|
"eval_samples_per_second": 72.451, |
|
"eval_steps_per_second": 0.246, |
|
"step": 3498 |
|
}, |
|
{ |
|
"epoch": 34.0, |
|
"learning_rate": 2e-05, |
|
"loss": 1.0018, |
|
"step": 3604 |
|
}, |
|
{ |
|
"epoch": 34.0, |
|
"eval_accuracy": 0.7890031028955456, |
|
"eval_loss": 0.9435957074165344, |
|
"eval_runtime": 24.4529, |
|
"eval_samples_per_second": 72.384, |
|
"eval_steps_per_second": 0.245, |
|
"step": 3604 |
|
}, |
|
{ |
|
"epoch": 35.0, |
|
"learning_rate": 2e-05, |
|
"loss": 1.0063, |
|
"step": 3710 |
|
}, |
|
{ |
|
"epoch": 35.0, |
|
"eval_accuracy": 0.7977425395315658, |
|
"eval_loss": 0.8957004547119141, |
|
"eval_runtime": 24.4025, |
|
"eval_samples_per_second": 72.534, |
|
"eval_steps_per_second": 0.246, |
|
"step": 3710 |
|
}, |
|
{ |
|
"epoch": 36.0, |
|
"learning_rate": 2e-05, |
|
"loss": 0.9982, |
|
"step": 3816 |
|
}, |
|
{ |
|
"epoch": 36.0, |
|
"eval_accuracy": 0.7993480649855951, |
|
"eval_loss": 0.8885225057601929, |
|
"eval_runtime": 24.4063, |
|
"eval_samples_per_second": 72.522, |
|
"eval_steps_per_second": 0.246, |
|
"step": 3816 |
|
}, |
|
{ |
|
"epoch": 37.0, |
|
"learning_rate": 2e-05, |
|
"loss": 0.9911, |
|
"step": 3922 |
|
}, |
|
{ |
|
"epoch": 37.0, |
|
"eval_accuracy": 0.8001705662056435, |
|
"eval_loss": 0.8866317868232727, |
|
"eval_runtime": 24.3478, |
|
"eval_samples_per_second": 72.697, |
|
"eval_steps_per_second": 0.246, |
|
"step": 3922 |
|
} |
|
], |
|
"max_steps": 4240, |
|
"num_train_epochs": 40, |
|
"total_flos": 472663961108480.0, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|