| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "global_step": 30414, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 4.917801012691524e-05, | |
| "loss": 4.2387, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 4.8356020253830476e-05, | |
| "loss": 3.1858, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 4.7534030380745706e-05, | |
| "loss": 2.9, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 4.671204050766095e-05, | |
| "loss": 2.8336, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_e": 0.3347315436241611, | |
| "eval_f1": 0.3097366216626532, | |
| "eval_loss": 3.770604372024536, | |
| "eval_runtime": 82.6518, | |
| "eval_samples_per_second": 43.266, | |
| "eval_steps_per_second": 43.266, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 4.589005063457618e-05, | |
| "loss": 2.6593, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 4.506806076149142e-05, | |
| "loss": 2.6292, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 4.424607088840665e-05, | |
| "loss": 2.5001, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 4.3424081015321896e-05, | |
| "loss": 2.779, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "eval_e": 0.4082774049217002, | |
| "eval_f1": 0.3790107109138272, | |
| "eval_loss": 3.6473984718322754, | |
| "eval_runtime": 82.1465, | |
| "eval_samples_per_second": 43.532, | |
| "eval_steps_per_second": 43.532, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 4.2602091142237125e-05, | |
| "loss": 2.4774, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 4.178010126915237e-05, | |
| "loss": 2.2356, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 4.09581113960676e-05, | |
| "loss": 2.312, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 4.013612152298284e-05, | |
| "loss": 2.1882, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "eval_e": 0.47511185682326623, | |
| "eval_f1": 0.42880102269324966, | |
| "eval_loss": 3.5800905227661133, | |
| "eval_runtime": 82.0337, | |
| "eval_samples_per_second": 43.592, | |
| "eval_steps_per_second": 43.592, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 3.931413164989807e-05, | |
| "loss": 2.2187, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 3.849214177681331e-05, | |
| "loss": 1.9332, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 3.7670151903728545e-05, | |
| "loss": 2.0221, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 3.684816203064378e-05, | |
| "loss": 2.1572, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "eval_e": 0.465324384787472, | |
| "eval_f1": 0.43269356216847144, | |
| "eval_loss": 3.3040895462036133, | |
| "eval_runtime": 82.0712, | |
| "eval_samples_per_second": 43.572, | |
| "eval_steps_per_second": 43.572, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 3.602617215755902e-05, | |
| "loss": 1.8899, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 3.5204182284474255e-05, | |
| "loss": 2.0571, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 3.438219241138949e-05, | |
| "loss": 1.8117, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 3.356020253830473e-05, | |
| "loss": 1.9861, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "eval_e": 0.5022371364653244, | |
| "eval_f1": 0.4571106421513448, | |
| "eval_loss": 3.3798608779907227, | |
| "eval_runtime": 82.3014, | |
| "eval_samples_per_second": 43.45, | |
| "eval_steps_per_second": 43.45, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 3.2738212665219965e-05, | |
| "loss": 1.5352, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 3.19162227921352e-05, | |
| "loss": 1.4703, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "learning_rate": 3.109423291905044e-05, | |
| "loss": 1.5819, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 3.0272243045965675e-05, | |
| "loss": 1.3955, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "eval_e": 0.5125838926174496, | |
| "eval_f1": 0.47356235492642834, | |
| "eval_loss": 3.511906147003174, | |
| "eval_runtime": 81.9656, | |
| "eval_samples_per_second": 43.628, | |
| "eval_steps_per_second": 43.628, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 2.9450253172880908e-05, | |
| "loss": 1.4744, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 2.8628263299796148e-05, | |
| "loss": 1.2676, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 2.780627342671138e-05, | |
| "loss": 1.3808, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 2.698428355362662e-05, | |
| "loss": 1.4232, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "eval_e": 0.47231543624161076, | |
| "eval_f1": 0.4436310133766935, | |
| "eval_loss": 4.007138252258301, | |
| "eval_runtime": 82.0736, | |
| "eval_samples_per_second": 43.571, | |
| "eval_steps_per_second": 43.571, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "learning_rate": 2.6162293680541855e-05, | |
| "loss": 1.3547, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "learning_rate": 2.5340303807457095e-05, | |
| "loss": 1.4022, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "learning_rate": 2.451831393437233e-05, | |
| "loss": 1.2848, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "learning_rate": 2.3696324061287568e-05, | |
| "loss": 1.3669, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "eval_e": 0.5131431767337807, | |
| "eval_f1": 0.47548209797210494, | |
| "eval_loss": 2.720799207687378, | |
| "eval_runtime": 82.4071, | |
| "eval_samples_per_second": 43.394, | |
| "eval_steps_per_second": 43.394, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "learning_rate": 2.2874334188202805e-05, | |
| "loss": 1.1994, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 2.205234431511804e-05, | |
| "loss": 1.2361, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "learning_rate": 2.1230354442033275e-05, | |
| "loss": 1.3868, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "learning_rate": 2.040836456894851e-05, | |
| "loss": 1.2285, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "eval_e": 0.5497762863534675, | |
| "eval_f1": 0.5054523815466766, | |
| "eval_loss": 3.2848546504974365, | |
| "eval_runtime": 82.2488, | |
| "eval_samples_per_second": 43.478, | |
| "eval_steps_per_second": 43.478, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 1.9586374695863748e-05, | |
| "loss": 1.0095, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "learning_rate": 1.8764384822778985e-05, | |
| "loss": 1.3675, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "learning_rate": 1.794239494969422e-05, | |
| "loss": 1.2251, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "learning_rate": 1.7120405076609458e-05, | |
| "loss": 1.1866, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "eval_e": 0.5528523489932886, | |
| "eval_f1": 0.5108568928693717, | |
| "eval_loss": 3.511319398880005, | |
| "eval_runtime": 82.0354, | |
| "eval_samples_per_second": 43.591, | |
| "eval_steps_per_second": 43.591, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 1.6298415203524694e-05, | |
| "loss": 1.1769, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "learning_rate": 1.547642533043993e-05, | |
| "loss": 0.8723, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "learning_rate": 1.4654435457355168e-05, | |
| "loss": 0.6951, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "learning_rate": 1.3832445584270404e-05, | |
| "loss": 0.7215, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "eval_e": 0.5564876957494407, | |
| "eval_f1": 0.49972487210398886, | |
| "eval_loss": 3.31522798538208, | |
| "eval_runtime": 82.1318, | |
| "eval_samples_per_second": 43.54, | |
| "eval_steps_per_second": 43.54, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "learning_rate": 1.301045571118564e-05, | |
| "loss": 0.8564, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "learning_rate": 1.2188465838100876e-05, | |
| "loss": 0.7706, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "learning_rate": 1.1366475965016111e-05, | |
| "loss": 0.7556, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "learning_rate": 1.0544486091931348e-05, | |
| "loss": 0.8466, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "eval_e": 0.581096196868009, | |
| "eval_f1": 0.5250151109087391, | |
| "eval_loss": 3.5018973350524902, | |
| "eval_runtime": 82.0779, | |
| "eval_samples_per_second": 43.568, | |
| "eval_steps_per_second": 43.568, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "learning_rate": 9.722496218846584e-06, | |
| "loss": 0.8433, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "learning_rate": 8.90050634576182e-06, | |
| "loss": 0.7424, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 2.52, | |
| "learning_rate": 8.078516472677057e-06, | |
| "loss": 0.7016, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "learning_rate": 7.256526599592293e-06, | |
| "loss": 0.7638, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "eval_e": 0.5746644295302014, | |
| "eval_f1": 0.5266203379718399, | |
| "eval_loss": 3.3142940998077393, | |
| "eval_runtime": 82.0427, | |
| "eval_samples_per_second": 43.587, | |
| "eval_steps_per_second": 43.587, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "learning_rate": 6.43453672650753e-06, | |
| "loss": 0.6414, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 2.66, | |
| "learning_rate": 5.6125468534227665e-06, | |
| "loss": 0.7472, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 2.71, | |
| "learning_rate": 4.790556980338002e-06, | |
| "loss": 0.6423, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "learning_rate": 3.968567107253239e-06, | |
| "loss": 0.78, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "eval_e": 0.5953579418344519, | |
| "eval_f1": 0.5431688990260519, | |
| "eval_loss": 3.2026774883270264, | |
| "eval_runtime": 82.1365, | |
| "eval_samples_per_second": 43.537, | |
| "eval_steps_per_second": 43.537, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 2.81, | |
| "learning_rate": 3.1465772341684756e-06, | |
| "loss": 0.6853, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 2.86, | |
| "learning_rate": 2.3245873610837114e-06, | |
| "loss": 0.6939, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 2.91, | |
| "learning_rate": 1.5025974879989478e-06, | |
| "loss": 0.6681, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "learning_rate": 6.806076149141843e-07, | |
| "loss": 0.6097, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "eval_e": 0.5953579418344519, | |
| "eval_f1": 0.5371973709058047, | |
| "eval_loss": 3.181272029876709, | |
| "eval_runtime": 82.1847, | |
| "eval_samples_per_second": 43.512, | |
| "eval_steps_per_second": 43.512, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 30414, | |
| "total_flos": 1.541864598291324e+16, | |
| "train_loss": 1.5040293447896245, | |
| "train_runtime": 3366.7923, | |
| "train_samples_per_second": 9.034, | |
| "train_steps_per_second": 9.034 | |
| } | |
| ], | |
| "max_steps": 30414, | |
| "num_train_epochs": 3, | |
| "total_flos": 1.541864598291324e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |