| { | |
| "best_metric": 1.095082402229309, | |
| "best_model_checkpoint": "/temp/mt5_base-qg-ap-oficial/checkpoint-6930", | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 6930, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 9.855699855699856e-05, | |
| "loss": 9.102, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 9.711399711399713e-05, | |
| "loss": 2.6848, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 9.567099567099568e-05, | |
| "loss": 2.1899, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 9.422799422799424e-05, | |
| "loss": 2.0546, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 9.278499278499279e-05, | |
| "loss": 1.9973, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 9.134199134199136e-05, | |
| "loss": 1.953, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 8.98989898989899e-05, | |
| "loss": 1.9029, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 8.845598845598845e-05, | |
| "loss": 1.8914, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 8.701298701298701e-05, | |
| "loss": 1.8197, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 8.556998556998557e-05, | |
| "loss": 1.7947, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 8.412698412698413e-05, | |
| "loss": 1.8097, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 8.268398268398268e-05, | |
| "loss": 1.7513, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 8.124098124098124e-05, | |
| "loss": 1.7276, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 1.348934531211853, | |
| "eval_runtime": 231.7834, | |
| "eval_samples_per_second": 194.608, | |
| "eval_steps_per_second": 3.042, | |
| "step": 1386 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 7.97979797979798e-05, | |
| "loss": 1.7093, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 7.835497835497836e-05, | |
| "loss": 1.6345, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 7.691197691197691e-05, | |
| "loss": 1.6314, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 7.546897546897548e-05, | |
| "loss": 1.6207, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 7.402597402597404e-05, | |
| "loss": 1.589, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "learning_rate": 7.258297258297259e-05, | |
| "loss": 1.612, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 7.113997113997114e-05, | |
| "loss": 1.5812, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "learning_rate": 6.96969696969697e-05, | |
| "loss": 1.6024, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "learning_rate": 6.825396825396825e-05, | |
| "loss": 1.5719, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "learning_rate": 6.681096681096681e-05, | |
| "loss": 1.5706, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "learning_rate": 6.536796536796536e-05, | |
| "loss": 1.5704, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 6.392496392496393e-05, | |
| "loss": 1.5733, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "learning_rate": 6.248196248196248e-05, | |
| "loss": 1.5852, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "learning_rate": 6.103896103896104e-05, | |
| "loss": 1.5698, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 1.2226001024246216, | |
| "eval_runtime": 231.7119, | |
| "eval_samples_per_second": 194.668, | |
| "eval_steps_per_second": 3.043, | |
| "step": 2772 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 5.959595959595959e-05, | |
| "loss": 1.5239, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "learning_rate": 5.815295815295816e-05, | |
| "loss": 1.4899, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "learning_rate": 5.6709956709956715e-05, | |
| "loss": 1.461, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "learning_rate": 5.526695526695527e-05, | |
| "loss": 1.4747, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "learning_rate": 5.382395382395382e-05, | |
| "loss": 1.481, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 2.38, | |
| "learning_rate": 5.2380952380952384e-05, | |
| "loss": 1.4496, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "learning_rate": 5.093795093795094e-05, | |
| "loss": 1.4665, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "learning_rate": 4.94949494949495e-05, | |
| "loss": 1.4397, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "learning_rate": 4.8051948051948054e-05, | |
| "loss": 1.4711, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "learning_rate": 4.6608946608946615e-05, | |
| "loss": 1.4528, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 2.74, | |
| "learning_rate": 4.516594516594517e-05, | |
| "loss": 1.4661, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 2.81, | |
| "learning_rate": 4.3722943722943724e-05, | |
| "loss": 1.4583, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "learning_rate": 4.227994227994228e-05, | |
| "loss": 1.4645, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "learning_rate": 4.083694083694084e-05, | |
| "loss": 1.4547, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 1.1470191478729248, | |
| "eval_runtime": 231.6851, | |
| "eval_samples_per_second": 194.691, | |
| "eval_steps_per_second": 3.043, | |
| "step": 4158 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "learning_rate": 3.939393939393939e-05, | |
| "loss": 1.4251, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "learning_rate": 3.7950937950937954e-05, | |
| "loss": 1.3732, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 3.17, | |
| "learning_rate": 3.650793650793651e-05, | |
| "loss": 1.3843, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 3.25, | |
| "learning_rate": 3.506493506493507e-05, | |
| "loss": 1.3884, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 3.32, | |
| "learning_rate": 3.3621933621933624e-05, | |
| "loss": 1.3902, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 3.39, | |
| "learning_rate": 3.217893217893218e-05, | |
| "loss": 1.4071, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 3.46, | |
| "learning_rate": 3.073593073593073e-05, | |
| "loss": 1.3857, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 3.54, | |
| "learning_rate": 2.9292929292929294e-05, | |
| "loss": 1.3623, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 3.61, | |
| "learning_rate": 2.7849927849927855e-05, | |
| "loss": 1.3954, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 3.68, | |
| "learning_rate": 2.640692640692641e-05, | |
| "loss": 1.39, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 3.75, | |
| "learning_rate": 2.4963924963924963e-05, | |
| "loss": 1.3866, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 3.82, | |
| "learning_rate": 2.352092352092352e-05, | |
| "loss": 1.382, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 3.9, | |
| "learning_rate": 2.207792207792208e-05, | |
| "loss": 1.3959, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 3.97, | |
| "learning_rate": 2.0634920634920636e-05, | |
| "loss": 1.3969, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 1.1056883335113525, | |
| "eval_runtime": 231.7985, | |
| "eval_samples_per_second": 194.596, | |
| "eval_steps_per_second": 3.041, | |
| "step": 5544 | |
| }, | |
| { | |
| "epoch": 4.04, | |
| "learning_rate": 1.919191919191919e-05, | |
| "loss": 1.3549, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 4.11, | |
| "learning_rate": 1.7748917748917752e-05, | |
| "loss": 1.3454, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 4.18, | |
| "learning_rate": 1.630591630591631e-05, | |
| "loss": 1.3436, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 4.26, | |
| "learning_rate": 1.4862914862914865e-05, | |
| "loss": 1.3465, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 4.33, | |
| "learning_rate": 1.3419913419913421e-05, | |
| "loss": 1.3498, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 4.4, | |
| "learning_rate": 1.1976911976911977e-05, | |
| "loss": 1.3535, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 4.47, | |
| "learning_rate": 1.0533910533910535e-05, | |
| "loss": 1.3471, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 4.55, | |
| "learning_rate": 9.090909090909091e-06, | |
| "loss": 1.3427, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 4.62, | |
| "learning_rate": 7.647907647907649e-06, | |
| "loss": 1.3496, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 4.69, | |
| "learning_rate": 6.204906204906205e-06, | |
| "loss": 1.3428, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 4.76, | |
| "learning_rate": 4.7619047619047615e-06, | |
| "loss": 1.3362, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 4.83, | |
| "learning_rate": 3.318903318903319e-06, | |
| "loss": 1.3424, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 4.91, | |
| "learning_rate": 1.875901875901876e-06, | |
| "loss": 1.3475, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 4.98, | |
| "learning_rate": 4.329004329004329e-07, | |
| "loss": 1.3748, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 1.095082402229309, | |
| "eval_runtime": 232.1115, | |
| "eval_samples_per_second": 194.333, | |
| "eval_steps_per_second": 3.037, | |
| "step": 6930 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 6930, | |
| "total_flos": 2.6579255601659904e+17, | |
| "train_loss": 1.650609011189074, | |
| "train_runtime": 5663.0471, | |
| "train_samples_per_second": 39.143, | |
| "train_steps_per_second": 1.224 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 6930, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "total_flos": 2.6579255601659904e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |