| { | |
| "best_metric": 0.06640350073575974, | |
| "best_model_checkpoint": "clock_output_cvt/checkpoint-475", | |
| "epoch": 24.675324675324674, | |
| "eval_steps": 500, | |
| "global_step": 475, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 1.0416666666666668e-05, | |
| "loss": 2.4077, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "eval_accuracy": 0.03625954198473282, | |
| "eval_loss": 2.3564352989196777, | |
| "eval_runtime": 0.6407, | |
| "eval_samples_per_second": 817.896, | |
| "eval_steps_per_second": 51.509, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 2.0833333333333336e-05, | |
| "loss": 2.388, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 3.125e-05, | |
| "loss": 2.3477, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "eval_accuracy": 0.4122137404580153, | |
| "eval_loss": 2.189662218093872, | |
| "eval_runtime": 0.6085, | |
| "eval_samples_per_second": 861.15, | |
| "eval_steps_per_second": 54.233, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 2.2977, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "learning_rate": 4.9765807962529274e-05, | |
| "loss": 2.1449, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "eval_accuracy": 0.816793893129771, | |
| "eval_loss": 1.8227535486221313, | |
| "eval_runtime": 0.613, | |
| "eval_samples_per_second": 854.76, | |
| "eval_steps_per_second": 53.83, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "learning_rate": 4.8594847775175645e-05, | |
| "loss": 2.0451, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 3.64, | |
| "learning_rate": 4.742388758782202e-05, | |
| "loss": 1.7728, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8549618320610687, | |
| "eval_loss": 1.2455334663391113, | |
| "eval_runtime": 0.6145, | |
| "eval_samples_per_second": 852.784, | |
| "eval_steps_per_second": 53.706, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 4.16, | |
| "learning_rate": 4.625292740046839e-05, | |
| "loss": 1.5092, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 4.68, | |
| "learning_rate": 4.508196721311476e-05, | |
| "loss": 1.3611, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 4.99, | |
| "eval_accuracy": 0.9408396946564885, | |
| "eval_loss": 0.8229649662971497, | |
| "eval_runtime": 0.7007, | |
| "eval_samples_per_second": 747.777, | |
| "eval_steps_per_second": 47.093, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 5.19, | |
| "learning_rate": 4.3911007025761124e-05, | |
| "loss": 1.254, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 5.71, | |
| "learning_rate": 4.2740046838407495e-05, | |
| "loss": 1.1696, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 5.97, | |
| "eval_accuracy": 0.9580152671755725, | |
| "eval_loss": 0.6116180419921875, | |
| "eval_runtime": 0.6501, | |
| "eval_samples_per_second": 806.052, | |
| "eval_steps_per_second": 50.763, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 6.23, | |
| "learning_rate": 4.156908665105387e-05, | |
| "loss": 1.0323, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 6.75, | |
| "learning_rate": 4.039812646370024e-05, | |
| "loss": 1.0261, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 6.96, | |
| "eval_accuracy": 0.9866412213740458, | |
| "eval_loss": 0.4360557198524475, | |
| "eval_runtime": 0.6176, | |
| "eval_samples_per_second": 848.435, | |
| "eval_steps_per_second": 53.432, | |
| "step": 134 | |
| }, | |
| { | |
| "epoch": 7.27, | |
| "learning_rate": 3.92271662763466e-05, | |
| "loss": 0.7872, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 7.79, | |
| "learning_rate": 3.8056206088992974e-05, | |
| "loss": 0.8913, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.9790076335877863, | |
| "eval_loss": 0.3085819184780121, | |
| "eval_runtime": 0.6116, | |
| "eval_samples_per_second": 856.814, | |
| "eval_steps_per_second": 53.96, | |
| "step": 154 | |
| }, | |
| { | |
| "epoch": 8.31, | |
| "learning_rate": 3.6885245901639346e-05, | |
| "loss": 0.7895, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 8.83, | |
| "learning_rate": 3.571428571428572e-05, | |
| "loss": 0.6873, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 8.99, | |
| "eval_accuracy": 0.9904580152671756, | |
| "eval_loss": 0.21129438281059265, | |
| "eval_runtime": 0.6415, | |
| "eval_samples_per_second": 816.831, | |
| "eval_steps_per_second": 51.442, | |
| "step": 173 | |
| }, | |
| { | |
| "epoch": 9.35, | |
| "learning_rate": 3.454332552693209e-05, | |
| "loss": 0.6743, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 9.87, | |
| "learning_rate": 3.337236533957845e-05, | |
| "loss": 0.7179, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 9.97, | |
| "eval_accuracy": 0.9904580152671756, | |
| "eval_loss": 0.18043160438537598, | |
| "eval_runtime": 0.6319, | |
| "eval_samples_per_second": 829.26, | |
| "eval_steps_per_second": 52.224, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 10.39, | |
| "learning_rate": 3.220140515222483e-05, | |
| "loss": 0.7141, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 10.91, | |
| "learning_rate": 3.1030444964871196e-05, | |
| "loss": 0.6876, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 10.96, | |
| "eval_accuracy": 0.9923664122137404, | |
| "eval_loss": 0.14928863942623138, | |
| "eval_runtime": 0.6689, | |
| "eval_samples_per_second": 783.382, | |
| "eval_steps_per_second": 49.335, | |
| "step": 211 | |
| }, | |
| { | |
| "epoch": 11.43, | |
| "learning_rate": 2.9859484777517567e-05, | |
| "loss": 0.5847, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 11.95, | |
| "learning_rate": 2.8688524590163935e-05, | |
| "loss": 0.5725, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.9923664122137404, | |
| "eval_loss": 0.12330139428377151, | |
| "eval_runtime": 0.6335, | |
| "eval_samples_per_second": 827.089, | |
| "eval_steps_per_second": 52.088, | |
| "step": 231 | |
| }, | |
| { | |
| "epoch": 12.47, | |
| "learning_rate": 2.7517564402810303e-05, | |
| "loss": 0.6264, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 12.99, | |
| "learning_rate": 2.6346604215456678e-05, | |
| "loss": 0.6068, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 12.99, | |
| "eval_accuracy": 0.9923664122137404, | |
| "eval_loss": 0.10889697074890137, | |
| "eval_runtime": 0.6123, | |
| "eval_samples_per_second": 855.795, | |
| "eval_steps_per_second": 53.895, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 13.51, | |
| "learning_rate": 2.5175644028103046e-05, | |
| "loss": 0.7263, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 13.97, | |
| "eval_accuracy": 0.9923664122137404, | |
| "eval_loss": 0.09741852432489395, | |
| "eval_runtime": 0.6166, | |
| "eval_samples_per_second": 849.786, | |
| "eval_steps_per_second": 53.517, | |
| "step": 269 | |
| }, | |
| { | |
| "epoch": 14.03, | |
| "learning_rate": 2.4004683840749414e-05, | |
| "loss": 0.5785, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 14.55, | |
| "learning_rate": 2.2833723653395785e-05, | |
| "loss": 0.5596, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 14.96, | |
| "eval_accuracy": 0.9923664122137404, | |
| "eval_loss": 0.08803129196166992, | |
| "eval_runtime": 0.6127, | |
| "eval_samples_per_second": 855.286, | |
| "eval_steps_per_second": 53.863, | |
| "step": 288 | |
| }, | |
| { | |
| "epoch": 15.06, | |
| "learning_rate": 2.1662763466042153e-05, | |
| "loss": 0.604, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 15.58, | |
| "learning_rate": 2.0491803278688525e-05, | |
| "loss": 0.5638, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.9923664122137404, | |
| "eval_loss": 0.08413369953632355, | |
| "eval_runtime": 0.6464, | |
| "eval_samples_per_second": 810.624, | |
| "eval_steps_per_second": 51.051, | |
| "step": 308 | |
| }, | |
| { | |
| "epoch": 16.1, | |
| "learning_rate": 1.9320843091334896e-05, | |
| "loss": 0.5579, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 16.62, | |
| "learning_rate": 1.8149882903981267e-05, | |
| "loss": 0.5601, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 16.99, | |
| "eval_accuracy": 0.9923664122137404, | |
| "eval_loss": 0.07752922177314758, | |
| "eval_runtime": 0.6764, | |
| "eval_samples_per_second": 774.698, | |
| "eval_steps_per_second": 48.788, | |
| "step": 327 | |
| }, | |
| { | |
| "epoch": 17.14, | |
| "learning_rate": 1.6978922716627635e-05, | |
| "loss": 0.3986, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 17.66, | |
| "learning_rate": 1.5807962529274007e-05, | |
| "loss": 0.606, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 17.97, | |
| "eval_accuracy": 0.9923664122137404, | |
| "eval_loss": 0.07416040450334549, | |
| "eval_runtime": 0.6307, | |
| "eval_samples_per_second": 830.823, | |
| "eval_steps_per_second": 52.323, | |
| "step": 346 | |
| }, | |
| { | |
| "epoch": 18.18, | |
| "learning_rate": 1.4637002341920375e-05, | |
| "loss": 0.6279, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 18.7, | |
| "learning_rate": 1.3466042154566746e-05, | |
| "loss": 0.5748, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 18.96, | |
| "eval_accuracy": 0.9923664122137404, | |
| "eval_loss": 0.07282942533493042, | |
| "eval_runtime": 0.8001, | |
| "eval_samples_per_second": 654.911, | |
| "eval_steps_per_second": 41.244, | |
| "step": 365 | |
| }, | |
| { | |
| "epoch": 19.22, | |
| "learning_rate": 1.2295081967213116e-05, | |
| "loss": 0.5834, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 19.74, | |
| "learning_rate": 1.1124121779859485e-05, | |
| "loss": 0.4852, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.9923664122137404, | |
| "eval_loss": 0.07014743238687515, | |
| "eval_runtime": 0.7282, | |
| "eval_samples_per_second": 719.555, | |
| "eval_steps_per_second": 45.315, | |
| "step": 385 | |
| }, | |
| { | |
| "epoch": 20.26, | |
| "learning_rate": 9.953161592505855e-06, | |
| "loss": 0.5781, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 20.78, | |
| "learning_rate": 8.782201405152225e-06, | |
| "loss": 0.4564, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 20.99, | |
| "eval_accuracy": 0.9923664122137404, | |
| "eval_loss": 0.06891454756259918, | |
| "eval_runtime": 0.7574, | |
| "eval_samples_per_second": 691.858, | |
| "eval_steps_per_second": 43.571, | |
| "step": 404 | |
| }, | |
| { | |
| "epoch": 21.3, | |
| "learning_rate": 7.611241217798595e-06, | |
| "loss": 0.5382, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 21.82, | |
| "learning_rate": 6.440281030444965e-06, | |
| "loss": 0.4278, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 21.97, | |
| "eval_accuracy": 0.9923664122137404, | |
| "eval_loss": 0.06956692785024643, | |
| "eval_runtime": 0.7478, | |
| "eval_samples_per_second": 700.728, | |
| "eval_steps_per_second": 44.13, | |
| "step": 423 | |
| }, | |
| { | |
| "epoch": 22.34, | |
| "learning_rate": 5.2693208430913356e-06, | |
| "loss": 0.4226, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 22.86, | |
| "learning_rate": 4.098360655737704e-06, | |
| "loss": 0.481, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 22.96, | |
| "eval_accuracy": 0.9923664122137404, | |
| "eval_loss": 0.06689619272947311, | |
| "eval_runtime": 0.7742, | |
| "eval_samples_per_second": 676.868, | |
| "eval_steps_per_second": 42.627, | |
| "step": 442 | |
| }, | |
| { | |
| "epoch": 23.38, | |
| "learning_rate": 2.9274004683840754e-06, | |
| "loss": 0.5381, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 23.9, | |
| "learning_rate": 1.7564402810304448e-06, | |
| "loss": 0.4477, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.9923664122137404, | |
| "eval_loss": 0.06719958037137985, | |
| "eval_runtime": 0.6158, | |
| "eval_samples_per_second": 850.91, | |
| "eval_steps_per_second": 53.588, | |
| "step": 462 | |
| }, | |
| { | |
| "epoch": 24.42, | |
| "learning_rate": 5.85480093676815e-07, | |
| "loss": 0.598, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 24.68, | |
| "eval_accuracy": 0.9923664122137404, | |
| "eval_loss": 0.06640350073575974, | |
| "eval_runtime": 0.6103, | |
| "eval_samples_per_second": 858.552, | |
| "eval_steps_per_second": 54.069, | |
| "step": 475 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 475, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 25, | |
| "save_steps": 500, | |
| "total_flos": 2737124951961600.0, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |