| { | |
| "best_metric": 0.9870629938253499, | |
| "best_model_checkpoint": "/home3/s5431786/nlp-final-project/results/roberta-base-e-snli-classification-nli_explanation-base/checkpoint-4800", | |
| "epoch": 0.6989748369058714, | |
| "global_step": 6000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 3.1055900621118013e-06, | |
| "loss": 0.7992, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "eval_accuracy": 0.9624060150375939, | |
| "eval_f1": 0.9623351677726931, | |
| "eval_loss": 0.1367802619934082, | |
| "eval_runtime": 4.2857, | |
| "eval_samples_per_second": 2296.452, | |
| "eval_steps_per_second": 35.933, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 6.2111801242236025e-06, | |
| "loss": 0.1514, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "eval_accuracy": 0.975716317821581, | |
| "eval_f1": 0.9756730714212107, | |
| "eval_loss": 0.08741448074579239, | |
| "eval_runtime": 4.5673, | |
| "eval_samples_per_second": 2154.901, | |
| "eval_steps_per_second": 33.718, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 9.316770186335405e-06, | |
| "loss": 0.121, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "eval_accuracy": 0.9778500304816095, | |
| "eval_f1": 0.9778119667614332, | |
| "eval_loss": 0.07925494015216827, | |
| "eval_runtime": 4.3138, | |
| "eval_samples_per_second": 2281.508, | |
| "eval_steps_per_second": 35.699, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 9.87246566383257e-06, | |
| "loss": 0.1088, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "eval_accuracy": 0.9818126397073765, | |
| "eval_f1": 0.9817742348532473, | |
| "eval_loss": 0.07087873667478561, | |
| "eval_runtime": 4.671, | |
| "eval_samples_per_second": 2107.055, | |
| "eval_steps_per_second": 32.97, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 9.708960104643559e-06, | |
| "loss": 0.095, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "eval_accuracy": 0.9833367201788255, | |
| "eval_f1": 0.983300542812346, | |
| "eval_loss": 0.07264702767133713, | |
| "eval_runtime": 4.2709, | |
| "eval_samples_per_second": 2304.409, | |
| "eval_steps_per_second": 36.058, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 9.545454545454547e-06, | |
| "loss": 0.0888, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "eval_accuracy": 0.983539930908352, | |
| "eval_f1": 0.9834936652245868, | |
| "eval_loss": 0.06838709861040115, | |
| "eval_runtime": 4.2889, | |
| "eval_samples_per_second": 2294.739, | |
| "eval_steps_per_second": 35.906, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 9.381948986265533e-06, | |
| "loss": 0.0833, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "eval_accuracy": 0.9846575899207478, | |
| "eval_f1": 0.984626969048982, | |
| "eval_loss": 0.05881618708372116, | |
| "eval_runtime": 4.3142, | |
| "eval_samples_per_second": 2281.283, | |
| "eval_steps_per_second": 35.696, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 9.218443427076522e-06, | |
| "loss": 0.0802, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "eval_accuracy": 0.9832351148140622, | |
| "eval_f1": 0.9831849160413134, | |
| "eval_loss": 0.06230289116501808, | |
| "eval_runtime": 4.4668, | |
| "eval_samples_per_second": 2203.364, | |
| "eval_steps_per_second": 34.477, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 9.054937867887508e-06, | |
| "loss": 0.0822, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "eval_accuracy": 0.9861816703921967, | |
| "eval_f1": 0.9861489775586069, | |
| "eval_loss": 0.05556264892220497, | |
| "eval_runtime": 4.7392, | |
| "eval_samples_per_second": 2076.706, | |
| "eval_steps_per_second": 32.495, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 8.891432308698496e-06, | |
| "loss": 0.0764, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "eval_accuracy": 0.9852672221093274, | |
| "eval_f1": 0.9852285614441253, | |
| "eval_loss": 0.05288001894950867, | |
| "eval_runtime": 4.2997, | |
| "eval_samples_per_second": 2288.976, | |
| "eval_steps_per_second": 35.816, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 8.727926749509485e-06, | |
| "loss": 0.0782, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "eval_accuracy": 0.9860800650274334, | |
| "eval_f1": 0.9860421413180931, | |
| "eval_loss": 0.05745375156402588, | |
| "eval_runtime": 4.289, | |
| "eval_samples_per_second": 2294.686, | |
| "eval_steps_per_second": 35.905, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 8.564421190320473e-06, | |
| "loss": 0.0722, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "eval_accuracy": 0.9870961186750661, | |
| "eval_f1": 0.9870629938253499, | |
| "eval_loss": 0.053428150713443756, | |
| "eval_runtime": 4.4266, | |
| "eval_samples_per_second": 2223.361, | |
| "eval_steps_per_second": 34.789, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 8.40091563113146e-06, | |
| "loss": 0.071, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "eval_accuracy": 0.985876854297907, | |
| "eval_f1": 0.9858488466773426, | |
| "eval_loss": 0.0532599538564682, | |
| "eval_runtime": 4.3538, | |
| "eval_samples_per_second": 2260.534, | |
| "eval_steps_per_second": 35.371, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 8.237410071942446e-06, | |
| "loss": 0.0723, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "eval_accuracy": 0.9860800650274334, | |
| "eval_f1": 0.9860406491659012, | |
| "eval_loss": 0.04955599084496498, | |
| "eval_runtime": 4.2895, | |
| "eval_samples_per_second": 2294.439, | |
| "eval_steps_per_second": 35.902, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 8.073904512753434e-06, | |
| "loss": 0.0713, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "eval_accuracy": 0.9860800650274334, | |
| "eval_f1": 0.9860488223249062, | |
| "eval_loss": 0.051210932433605194, | |
| "eval_runtime": 4.2586, | |
| "eval_samples_per_second": 2311.113, | |
| "eval_steps_per_second": 36.163, | |
| "step": 6000 | |
| } | |
| ], | |
| "max_steps": 25752, | |
| "num_train_epochs": 3, | |
| "total_flos": 1.597164223963392e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |