| { | |
| "best_metric": 0.8932172851340642, | |
| "best_model_checkpoint": "./save_models/qqp/roberta-base_lr1e-05/checkpoint-12800", | |
| "epoch": 10.0, | |
| "eval_steps": 500, | |
| "global_step": 12800, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 6.510416666666667e-06, | |
| "loss": 0.488, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 9.807180851063832e-06, | |
| "loss": 0.3263, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.879235948880033, | |
| "eval_averaged_scores": 0.8598257175939353, | |
| "eval_f1": 0.8404154863078377, | |
| "eval_loss": 0.27544060349464417, | |
| "eval_runtime": 30.2577, | |
| "eval_samples_per_second": 1202.503, | |
| "eval_steps_per_second": 4.726, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 9.391622340425532e-06, | |
| "loss": 0.2866, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 8.976063829787235e-06, | |
| "loss": 0.2674, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "learning_rate": 8.560505319148937e-06, | |
| "loss": 0.2554, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.891768585955751, | |
| "eval_averaged_scores": 0.8743571206442515, | |
| "eval_f1": 0.8569456553327521, | |
| "eval_loss": 0.2498372197151184, | |
| "eval_runtime": 27.6277, | |
| "eval_samples_per_second": 1316.974, | |
| "eval_steps_per_second": 5.176, | |
| "step": 2560 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "learning_rate": 8.144946808510639e-06, | |
| "loss": 0.2339, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.73, | |
| "learning_rate": 7.729388297872341e-06, | |
| "loss": 0.2263, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8935825202693417, | |
| "eval_averaged_scores": 0.8724041088244425, | |
| "eval_f1": 0.8512256973795435, | |
| "eval_loss": 0.2527249753475189, | |
| "eval_runtime": 27.9, | |
| "eval_samples_per_second": 1304.123, | |
| "eval_steps_per_second": 5.125, | |
| "step": 3840 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "learning_rate": 7.313829787234044e-06, | |
| "loss": 0.2159, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 3.52, | |
| "learning_rate": 6.898271276595745e-06, | |
| "loss": 0.2026, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 3.91, | |
| "learning_rate": 6.482712765957447e-06, | |
| "loss": 0.2011, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9005634189913426, | |
| "eval_averaged_scores": 0.8845524747016884, | |
| "eval_f1": 0.8685415304120341, | |
| "eval_loss": 0.23971907794475555, | |
| "eval_runtime": 27.6643, | |
| "eval_samples_per_second": 1315.233, | |
| "eval_steps_per_second": 5.169, | |
| "step": 5120 | |
| }, | |
| { | |
| "epoch": 4.3, | |
| "learning_rate": 6.06715425531915e-06, | |
| "loss": 0.1871, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 4.69, | |
| "learning_rate": 5.651595744680851e-06, | |
| "loss": 0.1803, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9041088360588154, | |
| "eval_averaged_scores": 0.8878877207108364, | |
| "eval_f1": 0.8716666053628573, | |
| "eval_loss": 0.2404341995716095, | |
| "eval_runtime": 27.871, | |
| "eval_samples_per_second": 1305.48, | |
| "eval_steps_per_second": 5.131, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 5.08, | |
| "learning_rate": 5.236037234042554e-06, | |
| "loss": 0.1775, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 5.47, | |
| "learning_rate": 4.820478723404256e-06, | |
| "loss": 0.1644, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 5.86, | |
| "learning_rate": 4.404920212765958e-06, | |
| "loss": 0.1654, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9068297375292016, | |
| "eval_averaged_scores": 0.8913133079820112, | |
| "eval_f1": 0.8757968784348208, | |
| "eval_loss": 0.24361343681812286, | |
| "eval_runtime": 28.0056, | |
| "eval_samples_per_second": 1299.207, | |
| "eval_steps_per_second": 5.106, | |
| "step": 7680 | |
| }, | |
| { | |
| "epoch": 6.25, | |
| "learning_rate": 3.98936170212766e-06, | |
| "loss": 0.1527, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 6.64, | |
| "learning_rate": 3.5738031914893617e-06, | |
| "loss": 0.1519, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.9057303833997526, | |
| "eval_averaged_scores": 0.8910411109500566, | |
| "eval_f1": 0.8763518385003605, | |
| "eval_loss": 0.24579738080501556, | |
| "eval_runtime": 28.1352, | |
| "eval_samples_per_second": 1293.219, | |
| "eval_steps_per_second": 5.083, | |
| "step": 8960 | |
| }, | |
| { | |
| "epoch": 7.03, | |
| "learning_rate": 3.1582446808510644e-06, | |
| "loss": 0.1534, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 7.42, | |
| "learning_rate": 2.7426861702127662e-06, | |
| "loss": 0.138, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 7.81, | |
| "learning_rate": 2.327127659574468e-06, | |
| "loss": 0.1418, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.9066923182630204, | |
| "eval_averaged_scores": 0.8916345948024302, | |
| "eval_f1": 0.8765768713418401, | |
| "eval_loss": 0.25888094305992126, | |
| "eval_runtime": 27.7859, | |
| "eval_samples_per_second": 1309.477, | |
| "eval_steps_per_second": 5.146, | |
| "step": 10240 | |
| }, | |
| { | |
| "epoch": 8.2, | |
| "learning_rate": 1.9115691489361704e-06, | |
| "loss": 0.1366, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 8.59, | |
| "learning_rate": 1.4960106382978725e-06, | |
| "loss": 0.1307, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 8.98, | |
| "learning_rate": 1.0804521276595746e-06, | |
| "loss": 0.1327, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.9079840593651229, | |
| "eval_averaged_scores": 0.8931944703014327, | |
| "eval_f1": 0.8784048812377424, | |
| "eval_loss": 0.25855743885040283, | |
| "eval_runtime": 27.685, | |
| "eval_samples_per_second": 1314.247, | |
| "eval_steps_per_second": 5.165, | |
| "step": 11520 | |
| }, | |
| { | |
| "epoch": 9.38, | |
| "learning_rate": 6.648936170212766e-07, | |
| "loss": 0.1271, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 9.77, | |
| "learning_rate": 2.4933510638297876e-07, | |
| "loss": 0.1257, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.9080665109248317, | |
| "eval_averaged_scores": 0.8932172851340642, | |
| "eval_f1": 0.8783680593432966, | |
| "eval_loss": 0.26402151584625244, | |
| "eval_runtime": 28.0618, | |
| "eval_samples_per_second": 1296.601, | |
| "eval_steps_per_second": 5.096, | |
| "step": 12800 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 12800, | |
| "total_flos": 1.5831238702432614e+17, | |
| "train_loss": 0.1970924124121666, | |
| "train_runtime": 4395.9489, | |
| "train_samples_per_second": 744.915, | |
| "train_steps_per_second": 2.912 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 12800, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "total_flos": 1.5831238702432614e+17, | |
| "train_batch_size": 256, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |