| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.0, | |
| "eval_steps": 500, | |
| "global_step": 217, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": 5.18852740291647, | |
| "learning_rate": 9.090909090909091e-06, | |
| "loss": 5.3084, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 5.03901362055586, | |
| "learning_rate": 4.545454545454546e-05, | |
| "loss": 5.0728, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 4.459588393324012, | |
| "learning_rate": 9.090909090909092e-05, | |
| "loss": 2.37, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 0.43459848938578693, | |
| "learning_rate": 0.00013636363636363637, | |
| "loss": 0.1307, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 0.0299929460934137, | |
| "learning_rate": 0.00018181818181818183, | |
| "loss": 0.0483, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 0.013812266692594782, | |
| "learning_rate": 0.00019988322268323268, | |
| "loss": 0.0225, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 0.009773102416496118, | |
| "learning_rate": 0.0001991705709821562, | |
| "loss": 0.0198, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 0.0039386428485718465, | |
| "learning_rate": 0.00019781476007338058, | |
| "loss": 0.015, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 0.002196082201771653, | |
| "learning_rate": 0.00019582458291091663, | |
| "loss": 0.0096, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "grad_norm": 0.018430687632200037, | |
| "learning_rate": 0.0001932129465573568, | |
| "loss": 0.0024, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 0.9834813463855321, | |
| "learning_rate": 0.0001899967884766212, | |
| "loss": 0.0281, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 0.005069811100933593, | |
| "learning_rate": 0.00018619696668800492, | |
| "loss": 0.0001, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "grad_norm": 0.007249668030582534, | |
| "learning_rate": 0.0001818381244939187, | |
| "loss": 0.0001, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 0.0071458935911795545, | |
| "learning_rate": 0.00017694853065861662, | |
| "loss": 0.0001, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 0.004757395926328272, | |
| "learning_rate": 0.00017155989607441213, | |
| "loss": 0.0001, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "grad_norm": 0.0027371110149472387, | |
| "learning_rate": 0.0001657071681043731, | |
| "loss": 0.0001, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "grad_norm": 0.0015303804705887707, | |
| "learning_rate": 0.00015942830393526176, | |
| "loss": 0.0, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "grad_norm": 0.0019029295128688208, | |
| "learning_rate": 0.0001527640244106133, | |
| "loss": 0.0191, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "grad_norm": 0.002913339239522618, | |
| "learning_rate": 0.00014575754994043956, | |
| "loss": 0.0, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "grad_norm": 0.002526914206171285, | |
| "learning_rate": 0.0001384543202002851, | |
| "loss": 0.0, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "grad_norm": 0.002364320350741189, | |
| "learning_rate": 0.00013090169943749476, | |
| "loss": 0.0, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 0.001415040670977836, | |
| "learning_rate": 0.00012314866929589432, | |
| "loss": 0.0, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "grad_norm": 0.0014806660372118452, | |
| "learning_rate": 0.00011524551115103454, | |
| "loss": 0.0, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "grad_norm": 0.0009238152395960922, | |
| "learning_rate": 0.00010724348001617625, | |
| "loss": 0.0, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "grad_norm": 0.316858259539684, | |
| "learning_rate": 9.919447213386103e-05, | |
| "loss": 0.0175, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "grad_norm": 0.0013793498850558623, | |
| "learning_rate": 9.115068840886417e-05, | |
| "loss": 0.0, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 0.03179793467382126, | |
| "learning_rate": 8.316429586529615e-05, | |
| "loss": 0.0, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "grad_norm": 0.008302429594092641, | |
| "learning_rate": 7.528708932343304e-05, | |
| "loss": 0.0104, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "grad_norm": 0.024284403402931108, | |
| "learning_rate": 6.757015549043175e-05, | |
| "loss": 0.0001, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "grad_norm": 0.007160721938093322, | |
| "learning_rate": 6.006354164343046e-05, | |
| "loss": 0.0001, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "grad_norm": 0.00442645759914904, | |
| "learning_rate": 5.28159310537518e-05, | |
| "loss": 0.0, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "grad_norm": 0.001965826516583593, | |
| "learning_rate": 4.587432725720687e-05, | |
| "loss": 0.0, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "grad_norm": 0.0012986955816234175, | |
| "learning_rate": 3.9283749218128885e-05, | |
| "loss": 0.0, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "grad_norm": 0.0010247882146462446, | |
| "learning_rate": 3.308693936411421e-05, | |
| "loss": 0.0, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "grad_norm": 0.000977980722589858, | |
| "learning_rate": 2.7324086384977698e-05, | |
| "loss": 0.0, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "grad_norm": 0.0008640860356404602, | |
| "learning_rate": 2.2032564593677774e-05, | |
| "loss": 0.0, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "grad_norm": 0.0007152715573634965, | |
| "learning_rate": 1.7246691539555028e-05, | |
| "loss": 0.0, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "grad_norm": 0.0008068934592673394, | |
| "learning_rate": 1.2997505445856084e-05, | |
| "loss": 0.0, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "grad_norm": 0.000785102458262206, | |
| "learning_rate": 9.31256391494546e-06, | |
| "loss": 0.0, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "grad_norm": 0.0006662070791456656, | |
| "learning_rate": 6.215765206679569e-06, | |
| "loss": 0.0, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "grad_norm": 0.0006676407776271578, | |
| "learning_rate": 3.7271932490209328e-06, | |
| "loss": 0.0, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "grad_norm": 0.0006248827839313129, | |
| "learning_rate": 1.8629873860586566e-06, | |
| "loss": 0.0, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "grad_norm": 0.0006849249794092451, | |
| "learning_rate": 6.352377081687011e-07, | |
| "loss": 0.0, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "grad_norm": 0.0007915981269703633, | |
| "learning_rate": 5.190664313851068e-08, | |
| "loss": 0.0056, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": NaN, | |
| "eval_runtime": 745.5269, | |
| "eval_samples_per_second": 1.551, | |
| "eval_steps_per_second": 0.388, | |
| "step": 217 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "step": 217, | |
| "total_flos": 1973257544663040.0, | |
| "train_loss": 0.0019402640651178403, | |
| "train_runtime": 4565.6051, | |
| "train_samples_per_second": 2.276, | |
| "train_steps_per_second": 0.048 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 217, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 40, | |
| "total_flos": 1973257544663040.0, | |
| "train_batch_size": 6, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |