| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.0, | |
| "eval_steps": 500, | |
| "global_step": 250, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 5.3643364906311035, | |
| "learning_rate": 0.00019987369566060176, | |
| "loss": 5.4501, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 7.935824394226074, | |
| "learning_rate": 0.00019949510169813003, | |
| "loss": 4.9761, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 6.881031036376953, | |
| "learning_rate": 0.0001986685944207868, | |
| "loss": 4.0797, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 6.994723320007324, | |
| "learning_rate": 0.00019745268727865774, | |
| "loss": 2.9864, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 1.695618748664856, | |
| "learning_rate": 0.0001958521789017376, | |
| "loss": 2.8715, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 5.028966903686523, | |
| "learning_rate": 0.00019387338576538744, | |
| "loss": 2.4938, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 3.31644344329834, | |
| "learning_rate": 0.00019152411726209176, | |
| "loss": 2.4925, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 10.450948715209961, | |
| "learning_rate": 0.0001899405251566371, | |
| "loss": 2.5641, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 2.144054412841797, | |
| "learning_rate": 0.0001870183754669526, | |
| "loss": 2.8727, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 92.50595092773438, | |
| "learning_rate": 0.0001837528040042142, | |
| "loss": 2.9301, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 26.027944564819336, | |
| "learning_rate": 0.00018015669848708767, | |
| "loss": 2.5419, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 12.144055366516113, | |
| "learning_rate": 0.0001762442511011448, | |
| "loss": 2.5677, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 12.395885467529297, | |
| "learning_rate": 0.0001720309024887907, | |
| "loss": 2.5559, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "grad_norm": 3.1185450553894043, | |
| "learning_rate": 0.00016753328081210245, | |
| "loss": 2.5763, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 5.239765167236328, | |
| "learning_rate": 0.00016276913612907007, | |
| "loss": 2.4963, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 6.314518451690674, | |
| "learning_rate": 0.00015775727034222675, | |
| "loss": 2.5602, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "grad_norm": 2.354008674621582, | |
| "learning_rate": 0.0001525174629961296, | |
| "loss": 2.6872, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "grad_norm": 1.4496679306030273, | |
| "learning_rate": 0.0001470703932165333, | |
| "loss": 2.4969, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "grad_norm": 9.645381927490234, | |
| "learning_rate": 0.00014143755809932845, | |
| "loss": 2.4725, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 1.5071710348129272, | |
| "learning_rate": 0.00013564118787132506, | |
| "loss": 2.3662, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "grad_norm": 3.7371649742126465, | |
| "learning_rate": 0.0001297041581577035, | |
| "loss": 2.3748, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "grad_norm": 3.71077823638916, | |
| "learning_rate": 0.00012364989970237248, | |
| "loss": 2.4374, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "grad_norm": 1.1883786916732788, | |
| "learning_rate": 0.00011750230589752762, | |
| "loss": 2.2459, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 2.8230981826782227, | |
| "learning_rate": 0.00011128563848734816, | |
| "loss": 2.6068, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 6.169120788574219, | |
| "learning_rate": 0.00010502443181797697, | |
| "loss": 2.6697, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "grad_norm": 1.7131496667861938, | |
| "learning_rate": 9.874339601166473e-05, | |
| "loss": 2.1541, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "grad_norm": 2.3437445163726807, | |
| "learning_rate": 9.246731944720675e-05, | |
| "loss": 2.5094, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 1.0078409910202026, | |
| "learning_rate": 8.62209709315362e-05, | |
| "loss": 2.693, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "grad_norm": 0.9157446026802063, | |
| "learning_rate": 8.002900194855932e-05, | |
| "loss": 2.4028, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 6.016546249389648, | |
| "learning_rate": 7.391584937101033e-05, | |
| "loss": 2.5198, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "grad_norm": 2.6909899711608887, | |
| "learning_rate": 6.790563901927907e-05, | |
| "loss": 2.3054, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 3.064624071121216, | |
| "learning_rate": 6.20220904478199e-05, | |
| "loss": 2.3357, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "grad_norm": 1.098503589630127, | |
| "learning_rate": 5.6288423334906735e-05, | |
| "loss": 2.1626, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "grad_norm": 0.9467936158180237, | |
| "learning_rate": 5.072726584517086e-05, | |
| "loss": 2.398, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 2.01023006439209, | |
| "learning_rate": 4.5360565326573104e-05, | |
| "loss": 2.4083, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "grad_norm": 1.5574626922607422, | |
| "learning_rate": 4.020950169424815e-05, | |
| "loss": 2.3016, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "grad_norm": 1.4903650283813477, | |
| "learning_rate": 3.52944038430556e-05, | |
| "loss": 2.1456, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "grad_norm": 2.6484365463256836, | |
| "learning_rate": 3.063466941871952e-05, | |
| "loss": 2.3501, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "grad_norm": 1.781996488571167, | |
| "learning_rate": 2.624868826418262e-05, | |
| "loss": 2.4699, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 1.1791914701461792, | |
| "learning_rate": 2.2153769843297667e-05, | |
| "loss": 2.5444, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "grad_norm": 9.811576843261719, | |
| "learning_rate": 1.8366074928281607e-05, | |
| "loss": 2.3116, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "grad_norm": 1.9491240978240967, | |
| "learning_rate": 1.4900551820530828e-05, | |
| "loss": 2.5006, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "grad_norm": 8.290270805358887, | |
| "learning_rate": 1.1770877356504683e-05, | |
| "loss": 2.2932, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "grad_norm": 1.6194989681243896, | |
| "learning_rate": 8.989402931500434e-06, | |
| "loss": 2.22, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "grad_norm": 2.6504359245300293, | |
| "learning_rate": 6.5671057543387985e-06, | |
| "loss": 2.4223, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "grad_norm": 2.504225969314575, | |
| "learning_rate": 4.513545525335705e-06, | |
| "loss": 2.294, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "grad_norm": 2.893841505050659, | |
| "learning_rate": 2.836826708532603e-06, | |
| "loss": 2.2694, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 1.2419345378875732, | |
| "learning_rate": 1.543566547079467e-06, | |
| "loss": 2.4213, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "grad_norm": 2.36327862739563, | |
| "learning_rate": 6.388689479991605e-07, | |
| "loss": 2.6469, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 2.0228891372680664, | |
| "learning_rate": 1.2630433939825327e-07, | |
| "loss": 2.6638, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "step": 250, | |
| "total_flos": 2336253085188096.0, | |
| "train_loss": 2.622329357147217, | |
| "train_runtime": 354.2831, | |
| "train_samples_per_second": 2.823, | |
| "train_steps_per_second": 0.706 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 250, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 100, | |
| "total_flos": 2336253085188096.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |