| { | |
| "best_metric": 0.06499230116605759, | |
| "best_model_checkpoint": "./beans_outputs/checkpoint-520", | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 650, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.07692307692307693, | |
| "grad_norm": 2.198392391204834, | |
| "learning_rate": 1.9692307692307696e-05, | |
| "loss": 1.0245, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.15384615384615385, | |
| "grad_norm": 1.846859335899353, | |
| "learning_rate": 1.9384615384615386e-05, | |
| "loss": 0.9453, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.23076923076923078, | |
| "grad_norm": 2.059973955154419, | |
| "learning_rate": 1.907692307692308e-05, | |
| "loss": 0.8405, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.3076923076923077, | |
| "grad_norm": 2.618006944656372, | |
| "learning_rate": 1.876923076923077e-05, | |
| "loss": 0.6956, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.38461538461538464, | |
| "grad_norm": 3.8514087200164795, | |
| "learning_rate": 1.8461538461538465e-05, | |
| "loss": 0.6615, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.46153846153846156, | |
| "grad_norm": 2.1362836360931396, | |
| "learning_rate": 1.8153846153846155e-05, | |
| "loss": 0.5829, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.5384615384615384, | |
| "grad_norm": 1.7998223304748535, | |
| "learning_rate": 1.784615384615385e-05, | |
| "loss": 0.5202, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.6153846153846154, | |
| "grad_norm": 1.2562752962112427, | |
| "learning_rate": 1.753846153846154e-05, | |
| "loss": 0.3936, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.6923076923076923, | |
| "grad_norm": 2.561612606048584, | |
| "learning_rate": 1.7230769230769234e-05, | |
| "loss": 0.3544, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.7692307692307693, | |
| "grad_norm": 2.900808811187744, | |
| "learning_rate": 1.6923076923076924e-05, | |
| "loss": 0.3681, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.8461538461538461, | |
| "grad_norm": 2.010755777359009, | |
| "learning_rate": 1.6615384615384618e-05, | |
| "loss": 0.3327, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.9230769230769231, | |
| "grad_norm": 2.478074550628662, | |
| "learning_rate": 1.630769230769231e-05, | |
| "loss": 0.2444, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 6.432145118713379, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 0.2815, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9624060150375939, | |
| "eval_loss": 0.21647371351718903, | |
| "eval_runtime": 0.9083, | |
| "eval_samples_per_second": 146.421, | |
| "eval_steps_per_second": 18.715, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.0769230769230769, | |
| "grad_norm": 1.6476765871047974, | |
| "learning_rate": 1.5692307692307693e-05, | |
| "loss": 0.1987, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.1538461538461537, | |
| "grad_norm": 3.0852138996124268, | |
| "learning_rate": 1.5384615384615387e-05, | |
| "loss": 0.2238, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.2307692307692308, | |
| "grad_norm": 1.8425233364105225, | |
| "learning_rate": 1.5076923076923078e-05, | |
| "loss": 0.2053, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.3076923076923077, | |
| "grad_norm": 0.5158002972602844, | |
| "learning_rate": 1.4769230769230772e-05, | |
| "loss": 0.2207, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.3846153846153846, | |
| "grad_norm": 0.6001647710800171, | |
| "learning_rate": 1.4461538461538462e-05, | |
| "loss": 0.2113, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.4615384615384617, | |
| "grad_norm": 4.286410808563232, | |
| "learning_rate": 1.4153846153846156e-05, | |
| "loss": 0.165, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.5384615384615383, | |
| "grad_norm": 0.4559226334095001, | |
| "learning_rate": 1.3846153846153847e-05, | |
| "loss": 0.2555, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.6153846153846154, | |
| "grad_norm": 0.6161900162696838, | |
| "learning_rate": 1.353846153846154e-05, | |
| "loss": 0.1819, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.6923076923076923, | |
| "grad_norm": 2.301481008529663, | |
| "learning_rate": 1.3230769230769231e-05, | |
| "loss": 0.2104, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.7692307692307692, | |
| "grad_norm": 1.8308448791503906, | |
| "learning_rate": 1.2923076923076925e-05, | |
| "loss": 0.1762, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.8461538461538463, | |
| "grad_norm": 5.278977394104004, | |
| "learning_rate": 1.2615384615384616e-05, | |
| "loss": 0.2573, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.9230769230769231, | |
| "grad_norm": 0.40304750204086304, | |
| "learning_rate": 1.230769230769231e-05, | |
| "loss": 0.1377, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.350562185049057, | |
| "learning_rate": 1.2e-05, | |
| "loss": 0.1287, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9699248120300752, | |
| "eval_loss": 0.13134175539016724, | |
| "eval_runtime": 0.9453, | |
| "eval_samples_per_second": 140.692, | |
| "eval_steps_per_second": 17.983, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.076923076923077, | |
| "grad_norm": 0.5022653341293335, | |
| "learning_rate": 1.1692307692307694e-05, | |
| "loss": 0.0932, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 2.1538461538461537, | |
| "grad_norm": 0.3385148346424103, | |
| "learning_rate": 1.1384615384615385e-05, | |
| "loss": 0.145, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 2.230769230769231, | |
| "grad_norm": 7.167346477508545, | |
| "learning_rate": 1.1076923076923079e-05, | |
| "loss": 0.1236, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 2.3076923076923075, | |
| "grad_norm": 11.584583282470703, | |
| "learning_rate": 1.076923076923077e-05, | |
| "loss": 0.138, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.3846153846153846, | |
| "grad_norm": 4.579661846160889, | |
| "learning_rate": 1.0461538461538463e-05, | |
| "loss": 0.1726, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 2.4615384615384617, | |
| "grad_norm": 0.30623820424079895, | |
| "learning_rate": 1.0153846153846154e-05, | |
| "loss": 0.1543, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 2.5384615384615383, | |
| "grad_norm": 5.757983684539795, | |
| "learning_rate": 9.846153846153848e-06, | |
| "loss": 0.1897, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 2.6153846153846154, | |
| "grad_norm": 1.116064190864563, | |
| "learning_rate": 9.53846153846154e-06, | |
| "loss": 0.0727, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 2.6923076923076925, | |
| "grad_norm": 0.3445896506309509, | |
| "learning_rate": 9.230769230769232e-06, | |
| "loss": 0.1346, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.769230769230769, | |
| "grad_norm": 12.3712158203125, | |
| "learning_rate": 8.923076923076925e-06, | |
| "loss": 0.1349, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 2.8461538461538463, | |
| "grad_norm": 0.29760444164276123, | |
| "learning_rate": 8.615384615384617e-06, | |
| "loss": 0.1519, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 2.9230769230769234, | |
| "grad_norm": 1.1171053647994995, | |
| "learning_rate": 8.307692307692309e-06, | |
| "loss": 0.1278, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 0.3179745078086853, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 0.1408, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9699248120300752, | |
| "eval_loss": 0.09979864209890366, | |
| "eval_runtime": 0.9058, | |
| "eval_samples_per_second": 146.838, | |
| "eval_steps_per_second": 18.769, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 3.076923076923077, | |
| "grad_norm": 2.1718528270721436, | |
| "learning_rate": 7.692307692307694e-06, | |
| "loss": 0.1321, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.1538461538461537, | |
| "grad_norm": 6.615433216094971, | |
| "learning_rate": 7.384615384615386e-06, | |
| "loss": 0.1546, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 3.230769230769231, | |
| "grad_norm": 8.979787826538086, | |
| "learning_rate": 7.076923076923078e-06, | |
| "loss": 0.1429, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 3.3076923076923075, | |
| "grad_norm": 0.2338067889213562, | |
| "learning_rate": 6.76923076923077e-06, | |
| "loss": 0.0768, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 3.3846153846153846, | |
| "grad_norm": 5.406207084655762, | |
| "learning_rate": 6.461538461538463e-06, | |
| "loss": 0.1158, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 3.4615384615384617, | |
| "grad_norm": 12.034098625183105, | |
| "learning_rate": 6.153846153846155e-06, | |
| "loss": 0.0831, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 3.5384615384615383, | |
| "grad_norm": 5.9186906814575195, | |
| "learning_rate": 5.846153846153847e-06, | |
| "loss": 0.0963, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 3.6153846153846154, | |
| "grad_norm": 8.250151634216309, | |
| "learning_rate": 5.538461538461539e-06, | |
| "loss": 0.0791, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 3.6923076923076925, | |
| "grad_norm": 0.21918846666812897, | |
| "learning_rate": 5.230769230769232e-06, | |
| "loss": 0.0971, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 3.769230769230769, | |
| "grad_norm": 0.21208502352237701, | |
| "learning_rate": 4.923076923076924e-06, | |
| "loss": 0.0718, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 3.8461538461538463, | |
| "grad_norm": 0.9157105684280396, | |
| "learning_rate": 4.615384615384616e-06, | |
| "loss": 0.1413, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.9230769230769234, | |
| "grad_norm": 0.20776814222335815, | |
| "learning_rate": 4.307692307692308e-06, | |
| "loss": 0.1272, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.30272677540779114, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 0.0846, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9924812030075187, | |
| "eval_loss": 0.06499230116605759, | |
| "eval_runtime": 0.9407, | |
| "eval_samples_per_second": 141.382, | |
| "eval_steps_per_second": 18.071, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 4.076923076923077, | |
| "grad_norm": 7.613215923309326, | |
| "learning_rate": 3.692307692307693e-06, | |
| "loss": 0.1002, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 4.153846153846154, | |
| "grad_norm": 1.5558936595916748, | |
| "learning_rate": 3.384615384615385e-06, | |
| "loss": 0.1657, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 4.230769230769231, | |
| "grad_norm": 0.3433710038661957, | |
| "learning_rate": 3.0769230769230774e-06, | |
| "loss": 0.1559, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 4.3076923076923075, | |
| "grad_norm": 1.3209067583084106, | |
| "learning_rate": 2.7692307692307697e-06, | |
| "loss": 0.0526, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 4.384615384615385, | |
| "grad_norm": 0.307217538356781, | |
| "learning_rate": 2.461538461538462e-06, | |
| "loss": 0.1305, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 4.461538461538462, | |
| "grad_norm": 0.273890882730484, | |
| "learning_rate": 2.153846153846154e-06, | |
| "loss": 0.0813, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 4.538461538461538, | |
| "grad_norm": 9.096426963806152, | |
| "learning_rate": 1.8461538461538465e-06, | |
| "loss": 0.0783, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 4.615384615384615, | |
| "grad_norm": 3.9279935359954834, | |
| "learning_rate": 1.5384615384615387e-06, | |
| "loss": 0.0952, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 4.6923076923076925, | |
| "grad_norm": 5.055220127105713, | |
| "learning_rate": 1.230769230769231e-06, | |
| "loss": 0.0868, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 4.769230769230769, | |
| "grad_norm": 0.2579015791416168, | |
| "learning_rate": 9.230769230769232e-07, | |
| "loss": 0.0834, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 4.846153846153846, | |
| "grad_norm": 0.2216470092535019, | |
| "learning_rate": 6.153846153846155e-07, | |
| "loss": 0.0644, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 4.923076923076923, | |
| "grad_norm": 0.23264001309871674, | |
| "learning_rate": 3.0769230769230774e-07, | |
| "loss": 0.0775, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.5718781352043152, | |
| "learning_rate": 0.0, | |
| "loss": 0.1105, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9774436090225563, | |
| "eval_loss": 0.08365403115749359, | |
| "eval_runtime": 0.9587, | |
| "eval_samples_per_second": 138.736, | |
| "eval_steps_per_second": 17.733, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 650, | |
| "total_flos": 4.006371770595533e+17, | |
| "train_loss": 0.21972015215800358, | |
| "train_runtime": 113.9852, | |
| "train_samples_per_second": 45.357, | |
| "train_steps_per_second": 5.702 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 650, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 4.006371770595533e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |